Structural Health Monitoring(SHM)systems play a key role in managing buildings and infrastructure by delivering vital insights into their strength and structural integrity.There is a need for more efficient techniques...Structural Health Monitoring(SHM)systems play a key role in managing buildings and infrastructure by delivering vital insights into their strength and structural integrity.There is a need for more efficient techniques to detect defects,as traditional methods are often prone to human error,and this issue is also addressed through image processing(IP).In addition to IP,automated,accurate,and real-time detection of structural defects,such as cracks,corrosion,and material degradation that conventional inspection techniques may miss,is made possible by Artificial Intelligence(AI)technologies like Machine Learning(ML)and Deep Learning(DL).This review examines the integration of computer vision and AI techniques in Structural Health Monitoring(SHM),investigating their effectiveness in detecting various forms of structural deterioration.Also,it evaluates ML and DL models in SHM for their accuracy in identifying and assessing structural damage,ultimately enhancing safety,durability,and maintenance practices in the field.Key findings reveal that AI-powered approaches,especially those utilizing IP and DL models like CNNs,significantly improve detection efficiency and accuracy,with reported accuracies in various SHM tasks.However,significant research gaps remain,including challenges with the consistency,quality,and environmental resilience of image data,a notable lack of standardized models and datasets for training across diverse structures,and concerns regarding computational costs,model interpretability,and seamless integration with existing systems.Future work should focus on developing more robust models through data augmentation,transfer learning,and hybrid approaches,standardizing protocols,and fostering interdisciplinary collaboration to overcome these limitations and achieve more reliable,scalable,and affordable SHM systems.展开更多
Breast cancer remains one of the most pressing global health concerns,and early detection plays a crucial role in improving survival rates.Integrating digital mammography with computational techniques and advanced ima...Breast cancer remains one of the most pressing global health concerns,and early detection plays a crucial role in improving survival rates.Integrating digital mammography with computational techniques and advanced image processing has significantly enhanced the ability to identify abnormalities.However,existing methodologies face persistent challenges,including low image contrast,noise interference,and inaccuracies in segmenting regions of interest.To address these limitations,this study introduces a novel computational framework for analyzing mammographic images,evaluated using the Mammographic Image Analysis Society(MIAS)dataset comprising 322 samples.The proposed methodology follows a structured three-stage approach.Initially,mammographic scans are classified using the Breast Imaging Reporting and Data System(BI-RADS),ensuring systematic and standardized image analysis.Next,the pectoral muscle,which can interfere with accurate segmentation,is effectively removed to refine the region of interest(ROI).The final stage involves an advanced image pre-processing module utilizing Independent Component Analysis(ICA)to enhance contrast,suppress noise,and improve image clarity.Following these enhancements,a robust segmentation technique is employed to delineated abnormal regions.Experimental results validate the efficiency of the proposed framework,demonstrating a significant improvement in the Effective Measure of Enhancement(EME)and a 3 dB increase in Peak Signal-to-Noise Ratio(PSNR),indicating superior image quality.The model also achieves an accuracy of approximately 97%,surpassing contemporary techniques evaluated on the MIAS dataset.Furthermore,its ability to process mammograms across all BI-RADS categories highlights its adaptability and reliability for clinical applications.This study presents an advanced and dependable computational framework for mammographic image analysis,effectively addressing critical challenges in noise reduction,contrast enhancement,and segmentation precision.The proposed approach lays the groundwork for seamless integration into computer-aided diagnostic(CAD)systems,with the potential to significantly enhance early breast cancer detection and contribute to improved patient outcomes.展开更多
This paper provides a comprehensive introduction to the mini-Si Tian Real-time Image Processing pipeline(STRIP)and evaluates its operational performance.The STRIP pipeline is specifically designed for real-time alert ...This paper provides a comprehensive introduction to the mini-Si Tian Real-time Image Processing pipeline(STRIP)and evaluates its operational performance.The STRIP pipeline is specifically designed for real-time alert triggering and light curve generation for transient sources.By applying the STRIP pipeline to both simulated and real observational data of the Mini-Si Tian survey,it successfully identified various types of variable sources,including stellar flares,supernovae,variable stars,and asteroids,while meeting requirements of reduction speed within 5 minutes.For the real observational data set,the pipeline detected one flare event,127 variable stars,and14 asteroids from three monitored sky regions.Additionally,two data sets were generated:one,a real-bogus training data set comprising 218,818 training samples,and the other,a variable star light curve data set with 421instances.These data sets will be used to train machine learning algorithms,which are planned for future integration into STRIP.展开更多
Machine learning(ML)is increasingly applied for medical image processing with appropriate learning paradigms.These applications include analyzing images of various organs,such as the brain,lung,eye,etc.,to identify sp...Machine learning(ML)is increasingly applied for medical image processing with appropriate learning paradigms.These applications include analyzing images of various organs,such as the brain,lung,eye,etc.,to identify specific flaws/diseases for diagnosis.The primary concern of ML applications is the precise selection of flexible image features for pattern detection and region classification.Most of the extracted image features are irrelevant and lead to an increase in computation time.Therefore,this article uses an analytical learning paradigm to design a Congruent Feature Selection Method to select the most relevant image features.This process trains the learning paradigm using similarity and correlation-based features over different textural intensities and pixel distributions.The similarity between the pixels over the various distribution patterns with high indexes is recommended for disease diagnosis.Later,the correlation based on intensity and distribution is analyzed to improve the feature selection congruency.Therefore,the more congruent pixels are sorted in the descending order of the selection,which identifies better regions than the distribution.Now,the learning paradigm is trained using intensity and region-based similarity to maximize the chances of selection.Therefore,the probability of feature selection,regardless of the textures and medical image patterns,is improved.This process enhances the performance of ML applications for different medical image processing.The proposed method improves the accuracy,precision,and training rate by 13.19%,10.69%,and 11.06%,respectively,compared to other models for the selected dataset.The mean error and selection time is also reduced by 12.56%and 13.56%,respectively,compared to the same models and dataset.展开更多
Deep learning now underpins many state-of-the-art systems for biomedical image and signal processing,enabling automated lesion detection,physiological monitoring,and therapy planning with accuracy that rivals expert p...Deep learning now underpins many state-of-the-art systems for biomedical image and signal processing,enabling automated lesion detection,physiological monitoring,and therapy planning with accuracy that rivals expert performance.This survey reviews the principal model families as convolutional,recurrent,generative,reinforcement,autoencoder,and transfer-learning approaches as emphasising how their architectural choices map to tasks such as segmentation,classification,reconstruction,and anomaly detection.A dedicated treatment of multimodal fusion networks shows how imaging features can be integrated with genomic profiles and clinical records to yield more robust,context-aware predictions.To support clinical adoption,we outline post-hoc explainability techniques(Grad-CAM,SHAP,LIME)and describe emerging intrinsically interpretable designs that expose decision logic to end users.Regulatory guidance from the U.S.FDA,the European Medicines Agency,and the EU AI Act is summarised,linking transparency and lifecycle-monitoring requirements to concrete development practices.Remaining challenges as data imbalance,computational cost,privacy constraints,and cross-domain generalization are discussed alongside promising solutions such as federated learning,uncertainty quantification,and lightweight 3-D architectures.The article therefore offers researchers,clinicians,and policymakers a concise,practice-oriented roadmap for deploying trustworthy deep-learning systems in healthcare.展开更多
Foundation models(FMs)have rapidly evolved and have achieved signicant accomplishments in computer vision tasks.Specically,the prompt mechanism conveniently allows users to integrate image prior information into the m...Foundation models(FMs)have rapidly evolved and have achieved signicant accomplishments in computer vision tasks.Specically,the prompt mechanism conveniently allows users to integrate image prior information into the model,making it possible to apply models without any training.Therefore,we proposed a workflow based on foundation models and zero training to solve the tasks of photoacoustic(PA)image processing.We employed the Segment Anything Model(SAM)by setting simple prompts and integrating the model's outputs with prior knowledge of the imaged objects to accomplish various tasks,including:(1)removing the skin signal in three-dimensional PA image rendering;(2)dual speed-of-sound reconstruction,and(3)segmentation ofnger blood vessels.Through these demonstrations,we have concluded that FMs can be directly applied in PA imaging without the requirement for network design and training.This potentially allows for a hands-on,convenient approach to achieving efficient and accurate segmentation of PA images.This paper serves as a comprehensive tutorial,facilitating the mastery of the technique through the provision of code and sample datasets.展开更多
1 Introduction onMultimodal Learning in Image Processing IP(Image processing),as a classical research domain in computer application technology,has been researched for decades.It is one of the most important research ...1 Introduction onMultimodal Learning in Image Processing IP(Image processing),as a classical research domain in computer application technology,has been researched for decades.It is one of the most important research directions in computer vision,which is the basis for many current hotspots such as intelligent transportation/education/industry,etc.Because image processing is the strongest link for AI(artificial intelligence)applying to real world application,it has been a challenging research field with the development of AI,from DNN(deep convolutional network),Attention/LSTM(long-short term memory),to Transformer/Diffusion/Mamba based GAI(generated AI)models,e.g.,GPT and Sora[1].Today,the description ability of single-model feature limits the performance of image processing.More comprehensive description of the image is required to match the computational performance of current large scale models.展开更多
Medical image processing technology plays an indispensable role in the field of modern medicine.By processing and analyzing medical images,it provides doctors with more comprehensive and accurate medical information,t...Medical image processing technology plays an indispensable role in the field of modern medicine.By processing and analyzing medical images,it provides doctors with more comprehensive and accurate medical information,thereby effectively aiding them in generating higher-quality treatment plans.In recent years,with the rapid development of deep learning technology,medical image processing techniques has been powered by providing more accurate information for diagnosis of disease.展开更多
The growing spectrum of Generative Adversarial Network (GAN) applications in medical imaging, cyber security, data augmentation, and the field of remote sensing tasks necessitate a sharp spike in the criticality of re...The growing spectrum of Generative Adversarial Network (GAN) applications in medical imaging, cyber security, data augmentation, and the field of remote sensing tasks necessitate a sharp spike in the criticality of review of Generative Adversarial Networks. Earlier reviews that targeted reviewing certain architecture of the GAN or emphasizing a specific application-oriented area have done so in a narrow spirit and lacked the systematic comparative analysis of the models’ performance metrics. Numerous reviews do not apply standardized frameworks, showing gaps in the efficiency evaluation of GANs, training stability, and suitability for specific tasks. In this work, a systemic review of GAN models using the PRISMA framework is developed in detail to fill the gap by structurally evaluating GAN architectures. A wide variety of GAN models have been discussed in this review, starting from the basic Conditional GAN, Wasserstein GAN, and Deep Convolutional GAN, and have gone down to many specialized models, such as EVAGAN, FCGAN, and SIF-GAN, for different applications across various domains like fault diagnosis, network security, medical imaging, and image segmentation. The PRISMA methodology systematically filters relevant studies by inclusion and exclusion criteria to ensure transparency and replicability in the review process. Hence, all models are assessed relative to specific performance metrics such as accuracy, stability, and computational efficiency. There are multiple benefits to using the PRISMA approach in this setup. Not only does this help in finding optimal models suitable for various applications, but it also provides an explicit framework for comparing GAN performance. In addition to this, diverse types of GAN are included to ensure a comprehensive view of the state-of-the-art techniques. This work is essential not only in terms of its result but also because it guides the direction of future research by pinpointing which types of applications require some GAN architectures, works to improve specific task model selection, and points out areas for further research on the development and application of GANs.展开更多
Real-time water-medium endoscopic images can assist doctors in performing operations such as tissue cleaning and nucleus pulpous removal.During medical operating procedures,it is inevitable that tissue particles,debri...Real-time water-medium endoscopic images can assist doctors in performing operations such as tissue cleaning and nucleus pulpous removal.During medical operating procedures,it is inevitable that tissue particles,debris and other contaminants will be suspended within the viewing area,resulting in blurred images and the loss of surface details in biological tissues.Currently,few studies have focused on enhancing such endoscopic images.This paper proposes a water-medium endoscopic image processing method based on dual transmittance in accordance with the imaging characteristics of spinal endoscopy.By establishing an underwater imaging model for spinal endoscopy,we estimate the transmittance of the endoscopic images based on the boundary constraints and local image contrast.The two transmittances are then fused and combined with transmittance maps and ambient light estimations to restore the images before attenuation,ultimately enhancing the details and texture of the images.Experiments comparing classical image enhancement algorithms demonstrate that the proposed algorithm could effectively improve the quality of spinal endoscopic images.展开更多
This study aims to enhance automated crop detection using high-resolution Unmanned Aerial Vehicle(UAV)imagery by integrating the Visible Atmospherically Resistant Index(VARI)with deep learning models.The primary chall...This study aims to enhance automated crop detection using high-resolution Unmanned Aerial Vehicle(UAV)imagery by integrating the Visible Atmospherically Resistant Index(VARI)with deep learning models.The primary challenge addressed is the detection of bananas interplanted with betel nuts,a scenario where traditional image processing techniques struggle due to color similarities and canopy overlap.The research explores the effectiveness of three deep learning models—Single Shot MultiBox Detector(SSD),You Only Look Once version 3(YOLOv3),and Faster Region-Based Convolutional Neural Network(Faster RCNN)—using Red,Green,Blue(RGB)and VARI images for banana detection.Results show that VARI significantly improves detection accuracy,with YOLOv3 achieving the best performance,achieving a precision of 73.77%,recall of 100%,and reduced training time by 95 seconds.Additionally,the average Intersection over Union(IoU)increased by 4%–25%across models with VARI-enhanced images.This study confirms that incorporating VARI improves the performance of deep learning models,offering a promising solution for precise crop detection in complex agricultural environments.展开更多
The precise identification of quartz minerals is crucial in mineralogy and geology due to their widespread occurrence and industrial significance.Traditional methods of quartz identification in thin sections are labor...The precise identification of quartz minerals is crucial in mineralogy and geology due to their widespread occurrence and industrial significance.Traditional methods of quartz identification in thin sections are labor-intensive and require significant expertise,often complicated by the coexistence of other minerals.This study presents a novel approach leveraging deep learning techniques combined with hyperspectral imaging to automate the identification process of quartz minerals.The utilizied four advanced deep learning models—PSPNet,U-Net,FPN,and LinkNet—has significant advancements in efficiency and accuracy.Among these models,PSPNet exhibited superior performance,achieving the highest intersection over union(IoU)scores and demonstrating exceptional reliability in segmenting quartz minerals,even in complex scenarios.The study involved a comprehensive dataset of 120 thin sections,encompassing 2470 hyperspectral images prepared from 20 rock samples.Expert-reviewed masks were used for model training,ensuring robust segmentation results.This automated approach not only expedites the recognition process but also enhances reliability,providing a valuable tool for geologists and advancing the field of mineralogical analysis.展开更多
In-optical-sensor computing architectures based on neuro-inspired optical sensor arrays have become key milestones for in-sensor artificial intelligence(AI)technology,enabling intelligent vision sensing and extensive ...In-optical-sensor computing architectures based on neuro-inspired optical sensor arrays have become key milestones for in-sensor artificial intelligence(AI)technology,enabling intelligent vision sensing and extensive data processing.These architectures must demonstrate potential advantages in terms of mass production and complementary metal oxide semiconductor compatibility.Here,we introduce a visible-light-driven neuromorphic vision system that integrates front-end retinomorphic photosensors with a back-end artificial neural network(ANN),employing a single neuro-inspired indium-g allium-zinc-oxide photo transistor(NIP)featuring an aluminum sensitization layer(ASL).By methodically adjusting the ASL coverage on IGZO phototransistors,a fast-switching response-type and a synaptic response-type of IGZO photo transistors are successfully developed.Notably,the fabricated NIP shows a remarkable retina-like photoinduced synaptic plasticity under wavelengths up to 635 nm,with over256-states,weight update nonlinearity below 0.1,and a dynamic range of 64.01.Owing to this technology,a 6×6 neuro-inspired optical image sensor array with the NIP can perform highly integrated sensing,memory,and preprocessing functions,including contrast enhancement,and handwritten digit image recognition.The demonstrated prototype highlights the potential for efficient hardware implementations in in-sensor AI technologies.展开更多
Digital watermarking technology plays an important role in detecting malicious tampering and protecting image copyright.However,in practical applications,this technology faces various problems such as severe image dis...Digital watermarking technology plays an important role in detecting malicious tampering and protecting image copyright.However,in practical applications,this technology faces various problems such as severe image distortion,inaccurate localization of the tampered regions,and difficulty in recovering content.Given these shortcomings,a fragile image watermarking algorithm for tampering blind-detection and content self-recovery is proposed.The multi-feature watermarking authentication code(AC)is constructed using texture feature of local binary patterns(LBP),direct coefficient of discrete cosine transform(DCT)and contrast feature of gray level co-occurrence matrix(GLCM)for detecting the tampered region,and the recovery code(RC)is designed according to the average grayscale value of pixels in image blocks for recovering the tampered content.Optimal pixel adjustment process(OPAP)and least significant bit(LSB)algorithms are used to embed the recovery code and authentication code into the image in a staggered manner.When detecting the integrity of the image,the authentication code comparison method and threshold judgment method are used to perform two rounds of tampering detection on the image and blindly recover the tampered content.Experimental results show that this algorithm has good transparency,strong and blind detection,and self-recovery performance against four types of malicious attacks and some conventional signal processing operations.When resisting copy-paste,text addition,cropping and vector quantization under the tampering rate(TR)10%,the average tampering detection rate is up to 94.09%,and the peak signal-to-noise ratio(PSNR)of the watermarked image and the recovered image are both greater than 41.47 and 40.31 dB,which demonstrates its excellent advantages compared with other related algorithms in recent years.展开更多
Images taken in dim environments frequently exhibit issues like insufficient brightness,noise,color shifts,and loss of detail.These problems pose significant challenges to dark image enhancement tasks.Current approach...Images taken in dim environments frequently exhibit issues like insufficient brightness,noise,color shifts,and loss of detail.These problems pose significant challenges to dark image enhancement tasks.Current approaches,while effective in global illumination modeling,often struggle to simultaneously suppress noise and preserve structural details,especially under heterogeneous lighting.Furthermore,misalignment between luminance and color channels introduces additional challenges to accurate enhancement.In response to the aforementioned difficulties,we introduce a single-stage framework,M2ATNet,using the multi-scale multi-attention and Transformer architecture.First,to address the problems of texture blurring and residual noise,we design a multi-scale multi-attention denoising module(MMAD),which is applied separately to the luminance and color channels to enhance the structural and texture modeling capabilities.Secondly,to solve the non-alignment problem of the luminance and color channels,we introduce the multi-channel feature fusion Transformer(CFFT)module,which effectively recovers the dark details and corrects the color shifts through cross-channel alignment and deep feature interaction.To guide the model to learn more stably and efficiently,we also fuse multiple types of loss functions to form a hybrid loss term.We extensively evaluate the proposed method on various standard datasets,including LOL-v1,LOL-v2,DICM,LIME,and NPE.Evaluation in terms of numerical metrics and visual quality demonstrate that M2ATNet consistently outperforms existing advanced approaches.Ablation studies further confirm the critical roles played by the MMAD and CFFT modules to detail preservation and visual fidelity under challenging illumination-deficient environments.展开更多
High-resolution remote sensing images(HRSIs)are now an essential data source for gathering surface information due to advancements in remote sensing data capture technologies.However,their significant scale changes an...High-resolution remote sensing images(HRSIs)are now an essential data source for gathering surface information due to advancements in remote sensing data capture technologies.However,their significant scale changes and wealth of spatial details pose challenges for semantic segmentation.While convolutional neural networks(CNNs)excel at capturing local features,they are limited in modeling long-range dependencies.Conversely,transformers utilize multihead self-attention to integrate global context effectively,but this approach often incurs a high computational cost.This paper proposes a global-local multiscale context network(GLMCNet)to extract both global and local multiscale contextual information from HRSIs.A detail-enhanced filtering module(DEFM)is proposed at the end of the encoder to refine the encoder outputs further,thereby enhancing the key details extracted by the encoder and effectively suppressing redundant information.In addition,a global-local multiscale transformer block(GLMTB)is proposed in the decoding stage to enable the modeling of rich multiscale global and local information.We also design a stair fusion mechanism to transmit deep semantic information from deep to shallow layers progressively.Finally,we propose the semantic awareness enhancement module(SAEM),which further enhances the representation of multiscale semantic features through spatial attention and covariance channel attention.Extensive ablation analyses and comparative experiments were conducted to evaluate the performance of the proposed method.Specifically,our method achieved a mean Intersection over Union(mIoU)of 86.89%on the ISPRS Potsdam dataset and 84.34%on the ISPRS Vaihingen dataset,outperforming existing models such as ABCNet and BANet.展开更多
Driven by advancements in mobile internet technology,images have become a crucial data medium.Ensuring the security of image information during transmission has thus emerged as an urgent challenge.This study proposes ...Driven by advancements in mobile internet technology,images have become a crucial data medium.Ensuring the security of image information during transmission has thus emerged as an urgent challenge.This study proposes a novel image encryption algorithm specifically designed for grayscale image security.This research introduces a new Cantor diagonal matrix permutation method.The proposed permutation method uses row and column index sequences to control the Cantor diagonal matrix,where the row and column index sequences are generated by a spatiotemporal chaotic system named coupled map lattice(CML).The high initial value sensitivity of the CML system makes the permutation method highly sensitive and secure.Additionally,leveraging fractal theory,this study introduces a chaotic fractal matrix and applies this matrix in the diffusion process.This chaotic fractal matrix exhibits selfsimilarity and irregularity.Using the Cantor diagonal matrix and chaotic fractal matrix,this paper introduces a fast image encryption algorithm involving two diffusion steps and one permutation step.Moreover,the algorithm achieves robust security with only a single encryption round,ensuring high operational efficiency.Experimental results show that the proposed algorithm features an expansive key space,robust security,high sensitivity,high efficiency,and superior statistical properties for the ciphered images.Thus,the proposed algorithm not only provides a practical solution for secure image transmission but also bridges fractal theory with image encryption techniques,thereby opening new research avenues in chaotic cryptography and advancing the development of information security technology.展开更多
Reversible data hiding(RDH)enables secret data embedding while preserving complete cover image recovery,making it crucial for applications requiring image integrity.The pixel value ordering(PVO)technique used in multi...Reversible data hiding(RDH)enables secret data embedding while preserving complete cover image recovery,making it crucial for applications requiring image integrity.The pixel value ordering(PVO)technique used in multi-stego images provides good image quality but often results in low embedding capability.To address these challenges,this paper proposes a high-capacity RDH scheme based on PVO that generates three stego images from a single cover image.The cover image is partitioned into non-overlapping blocks with pixels sorted in ascending order.Four secret bits are embedded into each block’s maximum pixel value,while three additional bits are embedded into the second-largest value when the pixel difference exceeds a predefined threshold.A similar embedding strategy is also applied to the minimum side of the block,including the second-smallest pixel value.This design enables each block to embed up to 14 bits of secret data.Experimental results demonstrate that the proposed method achieves significantly higher embedding capacity and improved visual quality compared to existing triple-stego RDH approaches,advancing the field of reversible steganography.展开更多
Remote sensing image super-resolution technology is pivotal for enhancing image quality in critical applications including environmental monitoring,urban planning,and disaster assessment.However,traditional methods ex...Remote sensing image super-resolution technology is pivotal for enhancing image quality in critical applications including environmental monitoring,urban planning,and disaster assessment.However,traditional methods exhibit deficiencies in detail recovery and noise suppression,particularly when processing complex landscapes(e.g.,forests,farmlands),leading to artifacts and spectral distortions that limit practical utility.To address this,we propose an enhanced Super-Resolution Generative Adversarial Network(SRGAN)framework featuring three key innovations:(1)Replacement of L1/L2 loss with a robust Charbonnier loss to suppress noise while preserving edge details via adaptive gradient balancing;(2)A multi-loss joint optimization strategy dynamically weighting Charbonnier loss(β=0.5),Visual Geometry Group(VGG)perceptual loss(α=1),and adversarial loss(γ=0.1)to synergize pixel-level accuracy and perceptual quality;(3)A multi-scale residual network(MSRN)capturing cross-scale texture features(e.g.,forest canopies,mountain contours).Validated on Sentinel-2(10 m)and SPOT-6/7(2.5 m)datasets covering 904 km2 in Motuo County,Xizang,our method outperforms the SRGAN baseline(SR4RS)with Peak Signal-to-Noise Ratio(PSNR)gains of 0.29 dB and Structural Similarity Index(SSIM)improvements of 3.08%on forest imagery.Visual comparisons confirm enhanced texture continuity despite marginal Learned Perceptual Image Patch Similarity(LPIPS)increases.The method significantly improves noise robustness and edge retention in complex geomorphology,demonstrating 18%faster response in forest fire early warning and providing high-resolution support for agricultural/urban monitoring.Future work will integrate spectral constraints and lightweight architectures.展开更多
In this study,an automated multimodal system for detecting,classifying,and dating fruit was developed using a two-stage YOLOv11 pipeline.In the first stage,the YOLOv11 detection model locates individual date fruits in...In this study,an automated multimodal system for detecting,classifying,and dating fruit was developed using a two-stage YOLOv11 pipeline.In the first stage,the YOLOv11 detection model locates individual date fruits in real time by drawing bounding boxes around them.These bounding boxes are subsequently passed to a YOLOv11 classification model,which analyzes cropped images and assigns class labels.An additional counting module automatically tallies the detected fruits,offering a near-instantaneous estimation of quantity.The experimental results suggest high precision and recall for detection,high classification accuracy(across 15 classes),and near-perfect counting in real time.This paper presents a multi-stage pipeline for date fruit detection,classification,and automated counting,employing YOLOv11-based models to achieve high accuracy while maintaining real-time throughput.The results demonstrated that the detection precision exceeded 90%,the classification accuracy approached 92%,and the counting module correlated closely with the manual tallies.These findings confirm the potential of reducing manual labour and enhancing operational efficiency in post-harvesting processes.Future studies will include dataset expansion,user-centric interfaces,and integration with harvesting robotics.展开更多
文摘Structural Health Monitoring(SHM)systems play a key role in managing buildings and infrastructure by delivering vital insights into their strength and structural integrity.There is a need for more efficient techniques to detect defects,as traditional methods are often prone to human error,and this issue is also addressed through image processing(IP).In addition to IP,automated,accurate,and real-time detection of structural defects,such as cracks,corrosion,and material degradation that conventional inspection techniques may miss,is made possible by Artificial Intelligence(AI)technologies like Machine Learning(ML)and Deep Learning(DL).This review examines the integration of computer vision and AI techniques in Structural Health Monitoring(SHM),investigating their effectiveness in detecting various forms of structural deterioration.Also,it evaluates ML and DL models in SHM for their accuracy in identifying and assessing structural damage,ultimately enhancing safety,durability,and maintenance practices in the field.Key findings reveal that AI-powered approaches,especially those utilizing IP and DL models like CNNs,significantly improve detection efficiency and accuracy,with reported accuracies in various SHM tasks.However,significant research gaps remain,including challenges with the consistency,quality,and environmental resilience of image data,a notable lack of standardized models and datasets for training across diverse structures,and concerns regarding computational costs,model interpretability,and seamless integration with existing systems.Future work should focus on developing more robust models through data augmentation,transfer learning,and hybrid approaches,standardizing protocols,and fostering interdisciplinary collaboration to overcome these limitations and achieve more reliable,scalable,and affordable SHM systems.
基金funded by Deanship of Graduate Studies and Scientific Research at Najran University for supporting the research project through the Nama’a program,with the project code NU/GP/MRC/13/771-4.
文摘Breast cancer remains one of the most pressing global health concerns,and early detection plays a crucial role in improving survival rates.Integrating digital mammography with computational techniques and advanced image processing has significantly enhanced the ability to identify abnormalities.However,existing methodologies face persistent challenges,including low image contrast,noise interference,and inaccuracies in segmenting regions of interest.To address these limitations,this study introduces a novel computational framework for analyzing mammographic images,evaluated using the Mammographic Image Analysis Society(MIAS)dataset comprising 322 samples.The proposed methodology follows a structured three-stage approach.Initially,mammographic scans are classified using the Breast Imaging Reporting and Data System(BI-RADS),ensuring systematic and standardized image analysis.Next,the pectoral muscle,which can interfere with accurate segmentation,is effectively removed to refine the region of interest(ROI).The final stage involves an advanced image pre-processing module utilizing Independent Component Analysis(ICA)to enhance contrast,suppress noise,and improve image clarity.Following these enhancements,a robust segmentation technique is employed to delineated abnormal regions.Experimental results validate the efficiency of the proposed framework,demonstrating a significant improvement in the Effective Measure of Enhancement(EME)and a 3 dB increase in Peak Signal-to-Noise Ratio(PSNR),indicating superior image quality.The model also achieves an accuracy of approximately 97%,surpassing contemporary techniques evaluated on the MIAS dataset.Furthermore,its ability to process mammograms across all BI-RADS categories highlights its adaptability and reliability for clinical applications.This study presents an advanced and dependable computational framework for mammographic image analysis,effectively addressing critical challenges in noise reduction,contrast enhancement,and segmentation precision.The proposed approach lays the groundwork for seamless integration into computer-aided diagnostic(CAD)systems,with the potential to significantly enhance early breast cancer detection and contribute to improved patient outcomes.
基金supported from the Strategic Pioneer Program of the Astronomy Large-Scale Scientific FacilityChinese Academy of Sciences and the Science and Education Integration Funding of University of Chinese Academy of Sciences+9 种基金the supports from the National Key Basic R&D Program of China via 2023YFA1608303the Strategic Priority Research Program of the Chinese Academy of Sciences(XDB0550103)the supports from the Strategic Priority Research Program of the Chinese Academy of Sciences under grant No.XDB0550000the National Natural Science Foundation of China(NSFC,grant Nos.12422303 and12261141690)the supports from the NSFC(grant No.12403024)supports from the NSFC through grant Nos.11988101 and 11933004the Postdoctoral Fellowship Program of CPSF under grant No.GZB20240731the Young Data Scientist Project of the National Astronomical Data Centerthe China Post-doctoral Science Foundation(No.2023M743447)supports from the New Cornerstone Science Foundation through the New Cornerstone Investigator Program and the XPLORER PRIZE。
文摘This paper provides a comprehensive introduction to the mini-Si Tian Real-time Image Processing pipeline(STRIP)and evaluates its operational performance.The STRIP pipeline is specifically designed for real-time alert triggering and light curve generation for transient sources.By applying the STRIP pipeline to both simulated and real observational data of the Mini-Si Tian survey,it successfully identified various types of variable sources,including stellar flares,supernovae,variable stars,and asteroids,while meeting requirements of reduction speed within 5 minutes.For the real observational data set,the pipeline detected one flare event,127 variable stars,and14 asteroids from three monitored sky regions.Additionally,two data sets were generated:one,a real-bogus training data set comprising 218,818 training samples,and the other,a variable star light curve data set with 421instances.These data sets will be used to train machine learning algorithms,which are planned for future integration into STRIP.
基金the Deanship of Scientifc Research at King Khalid University for funding this work through large group Research Project under grant number RGP2/421/45supported via funding from Prince Sattam bin Abdulaziz University project number(PSAU/2024/R/1446)+1 种基金supported by theResearchers Supporting Project Number(UM-DSR-IG-2023-07)Almaarefa University,Riyadh,Saudi Arabia.supported by the Basic Science Research Program through the National Research Foundation of Korea(NRF)funded by the Ministry of Education(No.2021R1F1A1055408).
文摘Machine learning(ML)is increasingly applied for medical image processing with appropriate learning paradigms.These applications include analyzing images of various organs,such as the brain,lung,eye,etc.,to identify specific flaws/diseases for diagnosis.The primary concern of ML applications is the precise selection of flexible image features for pattern detection and region classification.Most of the extracted image features are irrelevant and lead to an increase in computation time.Therefore,this article uses an analytical learning paradigm to design a Congruent Feature Selection Method to select the most relevant image features.This process trains the learning paradigm using similarity and correlation-based features over different textural intensities and pixel distributions.The similarity between the pixels over the various distribution patterns with high indexes is recommended for disease diagnosis.Later,the correlation based on intensity and distribution is analyzed to improve the feature selection congruency.Therefore,the more congruent pixels are sorted in the descending order of the selection,which identifies better regions than the distribution.Now,the learning paradigm is trained using intensity and region-based similarity to maximize the chances of selection.Therefore,the probability of feature selection,regardless of the textures and medical image patterns,is improved.This process enhances the performance of ML applications for different medical image processing.The proposed method improves the accuracy,precision,and training rate by 13.19%,10.69%,and 11.06%,respectively,compared to other models for the selected dataset.The mean error and selection time is also reduced by 12.56%and 13.56%,respectively,compared to the same models and dataset.
基金supported by the Science Committee of the Ministry of Higher Education and Science of the Republic of Kazakhstan within the framework of grant AP23489899“Applying Deep Learning and Neuroimaging Methods for Brain Stroke Diagnosis”.
文摘Deep learning now underpins many state-of-the-art systems for biomedical image and signal processing,enabling automated lesion detection,physiological monitoring,and therapy planning with accuracy that rivals expert performance.This survey reviews the principal model families as convolutional,recurrent,generative,reinforcement,autoencoder,and transfer-learning approaches as emphasising how their architectural choices map to tasks such as segmentation,classification,reconstruction,and anomaly detection.A dedicated treatment of multimodal fusion networks shows how imaging features can be integrated with genomic profiles and clinical records to yield more robust,context-aware predictions.To support clinical adoption,we outline post-hoc explainability techniques(Grad-CAM,SHAP,LIME)and describe emerging intrinsically interpretable designs that expose decision logic to end users.Regulatory guidance from the U.S.FDA,the European Medicines Agency,and the EU AI Act is summarised,linking transparency and lifecycle-monitoring requirements to concrete development practices.Remaining challenges as data imbalance,computational cost,privacy constraints,and cross-domain generalization are discussed alongside promising solutions such as federated learning,uncertainty quantification,and lightweight 3-D architectures.The article therefore offers researchers,clinicians,and policymakers a concise,practice-oriented roadmap for deploying trustworthy deep-learning systems in healthcare.
基金support from Strategic Project of Precision Surgery,Tsinghua UniversityInitiative Scientific Research Program,Institute for Intelligent Healthcare,Tsinghua University+5 种基金Tsinghua-Foshan Institute of Advanced ManufacturingNational Natural Science Foundation of China(61735016)Beijing Nova Program(20230484308)Young Elite Scientists Sponsorship Program by CAST(2023QNRC001)Youth Elite Program of Beijing Friendship Hospital(YYQCJH2022-9)Science and Technology Program of Beijing Tongzhou District(KJ2023CX012).
文摘Foundation models(FMs)have rapidly evolved and have achieved signicant accomplishments in computer vision tasks.Specically,the prompt mechanism conveniently allows users to integrate image prior information into the model,making it possible to apply models without any training.Therefore,we proposed a workflow based on foundation models and zero training to solve the tasks of photoacoustic(PA)image processing.We employed the Segment Anything Model(SAM)by setting simple prompts and integrating the model's outputs with prior knowledge of the imaged objects to accomplish various tasks,including:(1)removing the skin signal in three-dimensional PA image rendering;(2)dual speed-of-sound reconstruction,and(3)segmentation ofnger blood vessels.Through these demonstrations,we have concluded that FMs can be directly applied in PA imaging without the requirement for network design and training.This potentially allows for a hands-on,convenient approach to achieving efficient and accurate segmentation of PA images.This paper serves as a comprehensive tutorial,facilitating the mastery of the technique through the provision of code and sample datasets.
基金supported by 2023 Key Supported Project of the 14th Five Year Plan for Education and Science in Hunan Province with No.XJK23AXX0012021 Supported Project of the Educational Science Plan in Hunan Province with No.XJK21BXX010.
文摘1 Introduction onMultimodal Learning in Image Processing IP(Image processing),as a classical research domain in computer application technology,has been researched for decades.It is one of the most important research directions in computer vision,which is the basis for many current hotspots such as intelligent transportation/education/industry,etc.Because image processing is the strongest link for AI(artificial intelligence)applying to real world application,it has been a challenging research field with the development of AI,from DNN(deep convolutional network),Attention/LSTM(long-short term memory),to Transformer/Diffusion/Mamba based GAI(generated AI)models,e.g.,GPT and Sora[1].Today,the description ability of single-model feature limits the performance of image processing.More comprehensive description of the image is required to match the computational performance of current large scale models.
文摘Medical image processing technology plays an indispensable role in the field of modern medicine.By processing and analyzing medical images,it provides doctors with more comprehensive and accurate medical information,thereby effectively aiding them in generating higher-quality treatment plans.In recent years,with the rapid development of deep learning technology,medical image processing techniques has been powered by providing more accurate information for diagnosis of disease.
文摘The growing spectrum of Generative Adversarial Network (GAN) applications in medical imaging, cyber security, data augmentation, and the field of remote sensing tasks necessitate a sharp spike in the criticality of review of Generative Adversarial Networks. Earlier reviews that targeted reviewing certain architecture of the GAN or emphasizing a specific application-oriented area have done so in a narrow spirit and lacked the systematic comparative analysis of the models’ performance metrics. Numerous reviews do not apply standardized frameworks, showing gaps in the efficiency evaluation of GANs, training stability, and suitability for specific tasks. In this work, a systemic review of GAN models using the PRISMA framework is developed in detail to fill the gap by structurally evaluating GAN architectures. A wide variety of GAN models have been discussed in this review, starting from the basic Conditional GAN, Wasserstein GAN, and Deep Convolutional GAN, and have gone down to many specialized models, such as EVAGAN, FCGAN, and SIF-GAN, for different applications across various domains like fault diagnosis, network security, medical imaging, and image segmentation. The PRISMA methodology systematically filters relevant studies by inclusion and exclusion criteria to ensure transparency and replicability in the review process. Hence, all models are assessed relative to specific performance metrics such as accuracy, stability, and computational efficiency. There are multiple benefits to using the PRISMA approach in this setup. Not only does this help in finding optimal models suitable for various applications, but it also provides an explicit framework for comparing GAN performance. In addition to this, diverse types of GAN are included to ensure a comprehensive view of the state-of-the-art techniques. This work is essential not only in terms of its result but also because it guides the direction of future research by pinpointing which types of applications require some GAN architectures, works to improve specific task model selection, and points out areas for further research on the development and application of GANs.
文摘Real-time water-medium endoscopic images can assist doctors in performing operations such as tissue cleaning and nucleus pulpous removal.During medical operating procedures,it is inevitable that tissue particles,debris and other contaminants will be suspended within the viewing area,resulting in blurred images and the loss of surface details in biological tissues.Currently,few studies have focused on enhancing such endoscopic images.This paper proposes a water-medium endoscopic image processing method based on dual transmittance in accordance with the imaging characteristics of spinal endoscopy.By establishing an underwater imaging model for spinal endoscopy,we estimate the transmittance of the endoscopic images based on the boundary constraints and local image contrast.The two transmittances are then fused and combined with transmittance maps and ambient light estimations to restore the images before attenuation,ultimately enhancing the details and texture of the images.Experiments comparing classical image enhancement algorithms demonstrate that the proposed algorithm could effectively improve the quality of spinal endoscopic images.
文摘This study aims to enhance automated crop detection using high-resolution Unmanned Aerial Vehicle(UAV)imagery by integrating the Visible Atmospherically Resistant Index(VARI)with deep learning models.The primary challenge addressed is the detection of bananas interplanted with betel nuts,a scenario where traditional image processing techniques struggle due to color similarities and canopy overlap.The research explores the effectiveness of three deep learning models—Single Shot MultiBox Detector(SSD),You Only Look Once version 3(YOLOv3),and Faster Region-Based Convolutional Neural Network(Faster RCNN)—using Red,Green,Blue(RGB)and VARI images for banana detection.Results show that VARI significantly improves detection accuracy,with YOLOv3 achieving the best performance,achieving a precision of 73.77%,recall of 100%,and reduced training time by 95 seconds.Additionally,the average Intersection over Union(IoU)increased by 4%–25%across models with VARI-enhanced images.This study confirms that incorporating VARI improves the performance of deep learning models,offering a promising solution for precise crop detection in complex agricultural environments.
文摘The precise identification of quartz minerals is crucial in mineralogy and geology due to their widespread occurrence and industrial significance.Traditional methods of quartz identification in thin sections are labor-intensive and require significant expertise,often complicated by the coexistence of other minerals.This study presents a novel approach leveraging deep learning techniques combined with hyperspectral imaging to automate the identification process of quartz minerals.The utilizied four advanced deep learning models—PSPNet,U-Net,FPN,and LinkNet—has significant advancements in efficiency and accuracy.Among these models,PSPNet exhibited superior performance,achieving the highest intersection over union(IoU)scores and demonstrating exceptional reliability in segmenting quartz minerals,even in complex scenarios.The study involved a comprehensive dataset of 120 thin sections,encompassing 2470 hyperspectral images prepared from 20 rock samples.Expert-reviewed masks were used for model training,ensuring robust segmentation results.This automated approach not only expedites the recognition process but also enhances reliability,providing a valuable tool for geologists and advancing the field of mineralogical analysis.
基金supported by the National Research Foundation of Korea(NRF)Grant funded by the Korea government(MSIT)(Grant No.RS-2023-00256917)Samsung Display。
文摘In-optical-sensor computing architectures based on neuro-inspired optical sensor arrays have become key milestones for in-sensor artificial intelligence(AI)technology,enabling intelligent vision sensing and extensive data processing.These architectures must demonstrate potential advantages in terms of mass production and complementary metal oxide semiconductor compatibility.Here,we introduce a visible-light-driven neuromorphic vision system that integrates front-end retinomorphic photosensors with a back-end artificial neural network(ANN),employing a single neuro-inspired indium-g allium-zinc-oxide photo transistor(NIP)featuring an aluminum sensitization layer(ASL).By methodically adjusting the ASL coverage on IGZO phototransistors,a fast-switching response-type and a synaptic response-type of IGZO photo transistors are successfully developed.Notably,the fabricated NIP shows a remarkable retina-like photoinduced synaptic plasticity under wavelengths up to 635 nm,with over256-states,weight update nonlinearity below 0.1,and a dynamic range of 64.01.Owing to this technology,a 6×6 neuro-inspired optical image sensor array with the NIP can perform highly integrated sensing,memory,and preprocessing functions,including contrast enhancement,and handwritten digit image recognition.The demonstrated prototype highlights the potential for efficient hardware implementations in in-sensor AI technologies.
基金supported by Postgraduate Research&Practice Innovation Program of Jiangsu Province,China(Grant No.SJCX24_1332)Jiangsu Province Education Science Planning Project in 2024(Grant No.B-b/2024/01/122)High-Level Talent Scientific Research Foundation of Jinling Institute of Technology,China(Grant No.jit-b-201918).
文摘Digital watermarking technology plays an important role in detecting malicious tampering and protecting image copyright.However,in practical applications,this technology faces various problems such as severe image distortion,inaccurate localization of the tampered regions,and difficulty in recovering content.Given these shortcomings,a fragile image watermarking algorithm for tampering blind-detection and content self-recovery is proposed.The multi-feature watermarking authentication code(AC)is constructed using texture feature of local binary patterns(LBP),direct coefficient of discrete cosine transform(DCT)and contrast feature of gray level co-occurrence matrix(GLCM)for detecting the tampered region,and the recovery code(RC)is designed according to the average grayscale value of pixels in image blocks for recovering the tampered content.Optimal pixel adjustment process(OPAP)and least significant bit(LSB)algorithms are used to embed the recovery code and authentication code into the image in a staggered manner.When detecting the integrity of the image,the authentication code comparison method and threshold judgment method are used to perform two rounds of tampering detection on the image and blindly recover the tampered content.Experimental results show that this algorithm has good transparency,strong and blind detection,and self-recovery performance against four types of malicious attacks and some conventional signal processing operations.When resisting copy-paste,text addition,cropping and vector quantization under the tampering rate(TR)10%,the average tampering detection rate is up to 94.09%,and the peak signal-to-noise ratio(PSNR)of the watermarked image and the recovered image are both greater than 41.47 and 40.31 dB,which demonstrates its excellent advantages compared with other related algorithms in recent years.
基金funded by the National Natural Science Foundation of China,grant numbers 52374156 and 62476005。
文摘Images taken in dim environments frequently exhibit issues like insufficient brightness,noise,color shifts,and loss of detail.These problems pose significant challenges to dark image enhancement tasks.Current approaches,while effective in global illumination modeling,often struggle to simultaneously suppress noise and preserve structural details,especially under heterogeneous lighting.Furthermore,misalignment between luminance and color channels introduces additional challenges to accurate enhancement.In response to the aforementioned difficulties,we introduce a single-stage framework,M2ATNet,using the multi-scale multi-attention and Transformer architecture.First,to address the problems of texture blurring and residual noise,we design a multi-scale multi-attention denoising module(MMAD),which is applied separately to the luminance and color channels to enhance the structural and texture modeling capabilities.Secondly,to solve the non-alignment problem of the luminance and color channels,we introduce the multi-channel feature fusion Transformer(CFFT)module,which effectively recovers the dark details and corrects the color shifts through cross-channel alignment and deep feature interaction.To guide the model to learn more stably and efficiently,we also fuse multiple types of loss functions to form a hybrid loss term.We extensively evaluate the proposed method on various standard datasets,including LOL-v1,LOL-v2,DICM,LIME,and NPE.Evaluation in terms of numerical metrics and visual quality demonstrate that M2ATNet consistently outperforms existing advanced approaches.Ablation studies further confirm the critical roles played by the MMAD and CFFT modules to detail preservation and visual fidelity under challenging illumination-deficient environments.
基金provided by the Science Research Project of Hebei Education Department under grant No.BJK2024115.
文摘High-resolution remote sensing images(HRSIs)are now an essential data source for gathering surface information due to advancements in remote sensing data capture technologies.However,their significant scale changes and wealth of spatial details pose challenges for semantic segmentation.While convolutional neural networks(CNNs)excel at capturing local features,they are limited in modeling long-range dependencies.Conversely,transformers utilize multihead self-attention to integrate global context effectively,but this approach often incurs a high computational cost.This paper proposes a global-local multiscale context network(GLMCNet)to extract both global and local multiscale contextual information from HRSIs.A detail-enhanced filtering module(DEFM)is proposed at the end of the encoder to refine the encoder outputs further,thereby enhancing the key details extracted by the encoder and effectively suppressing redundant information.In addition,a global-local multiscale transformer block(GLMTB)is proposed in the decoding stage to enable the modeling of rich multiscale global and local information.We also design a stair fusion mechanism to transmit deep semantic information from deep to shallow layers progressively.Finally,we propose the semantic awareness enhancement module(SAEM),which further enhances the representation of multiscale semantic features through spatial attention and covariance channel attention.Extensive ablation analyses and comparative experiments were conducted to evaluate the performance of the proposed method.Specifically,our method achieved a mean Intersection over Union(mIoU)of 86.89%on the ISPRS Potsdam dataset and 84.34%on the ISPRS Vaihingen dataset,outperforming existing models such as ABCNet and BANet.
基金supported by the National Natural Science Foundation of China(62376106)The Science and Technology Development Plan of Jilin Province(20250102212JC).
文摘Driven by advancements in mobile internet technology,images have become a crucial data medium.Ensuring the security of image information during transmission has thus emerged as an urgent challenge.This study proposes a novel image encryption algorithm specifically designed for grayscale image security.This research introduces a new Cantor diagonal matrix permutation method.The proposed permutation method uses row and column index sequences to control the Cantor diagonal matrix,where the row and column index sequences are generated by a spatiotemporal chaotic system named coupled map lattice(CML).The high initial value sensitivity of the CML system makes the permutation method highly sensitive and secure.Additionally,leveraging fractal theory,this study introduces a chaotic fractal matrix and applies this matrix in the diffusion process.This chaotic fractal matrix exhibits selfsimilarity and irregularity.Using the Cantor diagonal matrix and chaotic fractal matrix,this paper introduces a fast image encryption algorithm involving two diffusion steps and one permutation step.Moreover,the algorithm achieves robust security with only a single encryption round,ensuring high operational efficiency.Experimental results show that the proposed algorithm features an expansive key space,robust security,high sensitivity,high efficiency,and superior statistical properties for the ciphered images.Thus,the proposed algorithm not only provides a practical solution for secure image transmission but also bridges fractal theory with image encryption techniques,thereby opening new research avenues in chaotic cryptography and advancing the development of information security technology.
基金funded by University of Transport and Communications(UTC)under grant number T2025-CN-004.
文摘Reversible data hiding(RDH)enables secret data embedding while preserving complete cover image recovery,making it crucial for applications requiring image integrity.The pixel value ordering(PVO)technique used in multi-stego images provides good image quality but often results in low embedding capability.To address these challenges,this paper proposes a high-capacity RDH scheme based on PVO that generates three stego images from a single cover image.The cover image is partitioned into non-overlapping blocks with pixels sorted in ascending order.Four secret bits are embedded into each block’s maximum pixel value,while three additional bits are embedded into the second-largest value when the pixel difference exceeds a predefined threshold.A similar embedding strategy is also applied to the minimum side of the block,including the second-smallest pixel value.This design enables each block to embed up to 14 bits of secret data.Experimental results demonstrate that the proposed method achieves significantly higher embedding capacity and improved visual quality compared to existing triple-stego RDH approaches,advancing the field of reversible steganography.
基金This study was supported by:Inner Mongolia Academy of Forestry Sciences Open Research Project(Grant No.KF2024MS03)The Project to Improve the Scientific Research Capacity of the Inner Mongolia Academy of Forestry Sciences(Grant No.2024NLTS04)The Innovation and Entrepreneurship Training Program for Undergraduates of Beijing Forestry University(Grant No.X202410022268).
文摘Remote sensing image super-resolution technology is pivotal for enhancing image quality in critical applications including environmental monitoring,urban planning,and disaster assessment.However,traditional methods exhibit deficiencies in detail recovery and noise suppression,particularly when processing complex landscapes(e.g.,forests,farmlands),leading to artifacts and spectral distortions that limit practical utility.To address this,we propose an enhanced Super-Resolution Generative Adversarial Network(SRGAN)framework featuring three key innovations:(1)Replacement of L1/L2 loss with a robust Charbonnier loss to suppress noise while preserving edge details via adaptive gradient balancing;(2)A multi-loss joint optimization strategy dynamically weighting Charbonnier loss(β=0.5),Visual Geometry Group(VGG)perceptual loss(α=1),and adversarial loss(γ=0.1)to synergize pixel-level accuracy and perceptual quality;(3)A multi-scale residual network(MSRN)capturing cross-scale texture features(e.g.,forest canopies,mountain contours).Validated on Sentinel-2(10 m)and SPOT-6/7(2.5 m)datasets covering 904 km2 in Motuo County,Xizang,our method outperforms the SRGAN baseline(SR4RS)with Peak Signal-to-Noise Ratio(PSNR)gains of 0.29 dB and Structural Similarity Index(SSIM)improvements of 3.08%on forest imagery.Visual comparisons confirm enhanced texture continuity despite marginal Learned Perceptual Image Patch Similarity(LPIPS)increases.The method significantly improves noise robustness and edge retention in complex geomorphology,demonstrating 18%faster response in forest fire early warning and providing high-resolution support for agricultural/urban monitoring.Future work will integrate spectral constraints and lightweight architectures.
基金supported by the Deanship of Scientific Research,Vice Presidency for Graduate Studies and Scientific Research,King Faisal University,Saudi Arabia,Grant No.KFU250098.
文摘In this study,an automated multimodal system for detecting,classifying,and dating fruit was developed using a two-stage YOLOv11 pipeline.In the first stage,the YOLOv11 detection model locates individual date fruits in real time by drawing bounding boxes around them.These bounding boxes are subsequently passed to a YOLOv11 classification model,which analyzes cropped images and assigns class labels.An additional counting module automatically tallies the detected fruits,offering a near-instantaneous estimation of quantity.The experimental results suggest high precision and recall for detection,high classification accuracy(across 15 classes),and near-perfect counting in real time.This paper presents a multi-stage pipeline for date fruit detection,classification,and automated counting,employing YOLOv11-based models to achieve high accuracy while maintaining real-time throughput.The results demonstrated that the detection precision exceeded 90%,the classification accuracy approached 92%,and the counting module correlated closely with the manual tallies.These findings confirm the potential of reducing manual labour and enhancing operational efficiency in post-harvesting processes.Future studies will include dataset expansion,user-centric interfaces,and integration with harvesting robotics.
