In the Internet era,recommendation systems play a crucial role in helping users find relevant information from large datasets.Class imbalance is known to severely affect data quality,and therefore reduce the performan...In the Internet era,recommendation systems play a crucial role in helping users find relevant information from large datasets.Class imbalance is known to severely affect data quality,and therefore reduce the performance of recommendation systems.Due to the imbalance,machine learning algorithms tend to classify inputs into the positive(majority)class every time to achieve high prediction accuracy.Imbalance can be categorized such as by features and classes,but most studies consider only class imbalance.In this paper,we propose a recommendation system that can integrate multiple networks to adapt to a large number of imbalanced features and can deal with highly skewed and imbalanced datasets through a loss function.We propose a loss aware feature attention mechanism(LAFAM)to solve the issue of feature imbalance.The network incorporates an attention mechanism and uses multiple sub-networks to classify and learn features.For better results,the network can learn the weights of sub-networks and assign higher weights to important features.We propose suppression loss to address class imbalance,which favors negative loss by penalizing positive loss,and pays more attention to sample points near the decision boundary.Experiments on two large-scale datasets verify that the performance of the proposed system is greatly improved compared to baseline methods.展开更多
This study proposes a new component of the composite loss function minimised during training of the Super-Resolution(SR)algorithms—the normalised structural similarity index loss LSSIMN,which has the potential to imp...This study proposes a new component of the composite loss function minimised during training of the Super-Resolution(SR)algorithms—the normalised structural similarity index loss LSSIMN,which has the potential to improve the natural appearance of reconstructed images.Deep learning-based super-resolution(SR)algorithms reconstruct high-resolution images from low-resolution inputs,offering a practical means to enhance image quality without requiring superior imaging hardware,which is particularly important in medical applications where diagnostic accuracy is critical.Although recent SR methods employing convolutional and generative adversarial networks achieve high pixel fidelity,visual artefacts may persist,making the design of the loss function during training essential for ensuring reliable and naturalistic image reconstruction.Our research shows on two models—SR and Invertible Rescaling Neural Network(IRN)—trained on multiple benchmark datasets that the function LSSIMN significantly contributes to the visual quality,preserving the structural fidelity on the reference datasets.The quantitative analysis of results while incorporating LSSIMN shows that including this loss function component has a mean 2.88%impact on the improvement of the final structural similarity of the reconstructed images in the validation set,in comparison to leaving it out and 0.218%in comparison when this component is non-normalised.展开更多
Currently,challenges such as small object size and occlusion lead to a lack of accuracy and robustness in small object detection.Since small objects occupy only a few pixels in an image,the extracted features are limi...Currently,challenges such as small object size and occlusion lead to a lack of accuracy and robustness in small object detection.Since small objects occupy only a few pixels in an image,the extracted features are limited,and mainstream downsampling convolution operations further exacerbate feature loss.Additionally,due to the occlusionprone nature of small objects and their higher sensitivity to localization deviations,conventional Intersection over Union(IoU)loss functions struggle to achieve stable convergence.To address these limitations,LR-Net is proposed for small object detection.Specifically,the proposed Lossless Feature Fusion(LFF)method transfers spatial features into the channel domain while leveraging a hybrid attentionmechanism to focus on critical features,mitigating feature loss caused by downsampling.Furthermore,RSIoU is proposed to enhance the convergence performance of IoU-based losses for small objects.RSIoU corrects the inherent convergence direction issues in SIoU and proposes a penalty term as a Dynamic Focusing Mechanism parameter,enabling it to dynamically emphasize the loss contribution of small object samples.Ultimately,RSIoU significantly improves the convergence performance of the loss function for small objects,particularly under occlusion scenarios.Experiments demonstrate that LR-Net achieves significant improvements across variousmetrics onmultiple datasets compared with YOLOv8n,achieving a 3.7% increase in mean Average Precision(AP)on the VisDrone2019 dataset,along with improvements of 3.3% on the AI-TOD dataset and 1.2% on the COCO dataset.展开更多
Traditional meteorological downscaling methods face limitations due to the complex distribution of meteorological variables,which can lead to unstable forecasting results,especially in extreme scenarios.To overcome th...Traditional meteorological downscaling methods face limitations due to the complex distribution of meteorological variables,which can lead to unstable forecasting results,especially in extreme scenarios.To overcome this issue,we propose a convolutional graph neural network(CGNN)model,which we enhance with multilayer feature fusion and a squeeze-and-excitation block.Additionally,we introduce a spatially balanced mean squared error(SBMSE)loss function to address the imbalanced distribution and spatial variability of meteorological variables.The CGNN is capable of extracting essential spatial features and aggregating them from a global perspective,thereby improving the accuracy of prediction and enhancing the model's generalization ability.Based on the experimental results,CGNN has certain advantages in terms of bias distribution,exhibiting a smaller variance.When it comes to precipitation,both UNet and AE also demonstrate relatively small biases.As for temperature,AE and CNNdense perform outstandingly during the winter.The time correlation coefficients show an improvement of at least 10%at daily and monthly scales for both temperature and precipitation.Furthermore,the SBMSE loss function displays an advantage over existing loss functions in predicting the98th percentile and identifying areas where extreme events occur.However,the SBMSE tends to overestimate the distribution of extreme precipitation,which may be due to the theoretical assumptions about the posterior distribution of data that partially limit the effectiveness of the loss function.In future work,we will further optimize the SBMSE to improve prediction accuracy.展开更多
The current existing problem of deep learning framework for the detection and segmentation of electrical equipment is dominantly related to low precision.Because of the reliable,safe and easy-to-operate technology pro...The current existing problem of deep learning framework for the detection and segmentation of electrical equipment is dominantly related to low precision.Because of the reliable,safe and easy-to-operate technology provided by deep learning-based video surveillance for unmanned inspection of electrical equipment,this paper uses the bottleneck attention module(BAM)attention mechanism to improve the Solov2 model and proposes a new electrical equipment segmentation mode.Firstly,the BAM attention mechanism is integrated into the feature extraction network to adaptively learn the correlation between feature channels,thereby improving the expression ability of the feature map;secondly,the weighted sum of CrossEntropy Loss and Dice loss is designed as the mask loss to improve the segmentation accuracy and robustness of the model;finally,the non-maximal suppression(NMS)algorithm to better handle the overlap problem in instance segmentation.Experimental results show that the proposed method achieves an average segmentation accuracy of mAP of 80.4% on three types of electrical equipment datasets,including transformers,insulators and voltage transformers,which improve the detection accuracy by more than 5.7% compared with the original Solov2 model.The segmentation model proposed can provide a focusing technical means for the intelligent management of power systems.展开更多
Earthquakes pose significant perils to the built environment in urban areas.To avert the calamitous aftermath of earthquakes,it is imperative to construct seismic resilient cities.Due to the intricacy of the concept o...Earthquakes pose significant perils to the built environment in urban areas.To avert the calamitous aftermath of earthquakes,it is imperative to construct seismic resilient cities.Due to the intricacy of the concept of urban seismic resilience(USR),its assessment is a large-scale system engineering issue.The assessment of USR should be based on the notion of urban seismic capacity(USC)assessment,which includes casualties,economic loss,and recovery time as criteria.Functionality loss is also included in the assessment of USR in addition to these criteria.The assessment indicator system comprising five dimensions(building and lifeline infrastructure,environment,society,economy,and institution)and 20 indicators has been devised to quantify USR.The analytical hierarchy process(AHP)is utilized to compute the weights of the criteria,dimensions,and indicators in the urban seismic resilience assessment(USRA)indicator system.When the necessary data for a city are obtainable,the seismic resilience of that city can be assessed using this framework.To illustrate the proposed methodology,a moderate-sized city in China was selected as a case study.The assessment results indicate a high level of USR,suggesting that the city possesses strong capabilities to withstand and recover from potential future earthquakes.展开更多
To quantify the seismic resilience of buildings,a method for evaluating functional loss from the component level to the overall building is proposed,and the dual-parameter seismic resilience assessment method based on...To quantify the seismic resilience of buildings,a method for evaluating functional loss from the component level to the overall building is proposed,and the dual-parameter seismic resilience assessment method based on postearthquake loss and recovery time is improved.A threelevel function tree model is established,which can consider the dynamic changes in weight coefficients of different category of components relative to their functional losses.Bayesian networks are utilized to quantify the impact of weather conditions,construction technology levels,and worker skill levels on component repair time.A method for determining the real-time functional recovery curve of buildings based on the component repair process is proposed.Taking a three-story teaching building as an example,the seismic resilience indices under basic earthquakes and rare earthquakes are calculated.The results show that the seismic resilience grade of the teaching building is comprehensively judged as GradeⅢ,and its resilience grade is more significantly affected by postearthquake loss.The proposed method can be used to predict the seismic resilience of buildings prior to earthquakes,identify weak components within buildings,and provide guidance for taking measures to enhance the seismic resilience of buildings.展开更多
Title:A dual-parameter method for seismic resilience assessment of buildings Authors:LI Shuang;HU Binbin;LIU Wen;ZHAI Changhai Abstract:To quantify the seismic resilience of buildings,a method for evaluating functiona...Title:A dual-parameter method for seismic resilience assessment of buildings Authors:LI Shuang;HU Binbin;LIU Wen;ZHAI Changhai Abstract:To quantify the seismic resilience of buildings,a method for evaluating functional loss from the component level to the overall building is proposed,and the dual-parameter seismic resilience assessment method based on postearthquake loss and recovery time is improved.A three-level function tree model is established,which can consider the dynamic changes in weight coefficients of different category of components relative to their functional losses.Bayesian networks are utilized to quantify the impact of weather conditions,construction technology levels,and worker skill levels on component repair time.A method for determining the real-time functional recovery curve of buildings based on the component repair process is proposed.Taking a three-story teaching building as an example,the seismic resilience indices under basic earthquakes and rare earthquakes are calculated.The results show that the seismic resilience grade of the teaching building is comprehensively judged as GradeⅢ,and its resilience grade is more significantly affected by postearthquake loss.The proposed method can be used to predict the seismic resilience of buildings prior to earthquakes,identify weak components within buildings,and provide guidance for taking measures to enhance the seismic resilience of buildings.展开更多
In this paper we propose an absolute error loss EB estimator for parameter of one-side truncation distribution families. Under some conditions we have proved that the convergence rates of its Bayes risk is o, where 0&...In this paper we propose an absolute error loss EB estimator for parameter of one-side truncation distribution families. Under some conditions we have proved that the convergence rates of its Bayes risk is o, where 0<λ,r≤1,Mn≤lnln n (for large n),Mn→∞ as n→∞.展开更多
Detecting individuals wearing safety helmets in complex environments faces several challenges.These factors include limited detection accuracy and frequent missed or false detections.Additionally,existing algorithms o...Detecting individuals wearing safety helmets in complex environments faces several challenges.These factors include limited detection accuracy and frequent missed or false detections.Additionally,existing algorithms often have excessive parameter counts,complex network structures,and high computational demands.These challenges make it difficult to deploy such models efficiently on resource-constrained devices like embedded systems.Aiming at this problem,this research proposes an optimized and lightweight solution called FGP-YOLOv8,an improved version of YOLOv8n.The YOLOv8 backbone network is replaced with the FasterNet model to reduce parameters and computational demands while local convolution layers are added.This modification minimizes computational costs with only a minor impact on accuracy.A new GSTA(GSConv-Triplet Attention)module is introduced to enhance feature fusion and reduce computational complexity.This is achieved using attention weights generated from dimensional interactions within the feature map.Additionally,the ParNet-C2f module replaces the original C2f(CSP Bottleneck with 2 Convolutions)module,improving feature extraction for safety helmets of various shapes and sizes.The CIoU(Complete-IoU)is replaced with the WIoU(Wise-IoU)to boost performance further,enhancing detection accuracy and generalization capabilities.Experimental results validate the improvements.The proposedmodel reduces the parameter count by 19.9% and the computational load by 18.5%.At the same time,mAP(mean average precision)increases by 2.3%,and precision improves by 1.2%.These results demonstrate the model’s robust performance in detecting safety helmets across diverse environments.展开更多
UAV-based object detection is rapidly expanding in both civilian and military applications,including security surveillance,disaster assessment,and border patrol.However,challenges such as small objects,occlusions,comp...UAV-based object detection is rapidly expanding in both civilian and military applications,including security surveillance,disaster assessment,and border patrol.However,challenges such as small objects,occlusions,complex backgrounds,and variable lighting persist due to the unique perspective of UAV imagery.To address these issues,this paper introduces DAFPN-YOLO,an innovative model based on YOLOv8s(You Only Look Once version 8s).Themodel strikes a balance between detection accuracy and speed while reducing parameters,making itwell-suited for multi-object detection tasks from drone perspectives.A key feature of DAFPN-YOLO is the enhanced Drone-AFPN(Adaptive Feature Pyramid Network),which adaptively fuses multi-scale features to optimize feature extraction and enhance spatial and small-object information.To leverage Drone-AFPN’smulti-scale capabilities fully,a dedicated 160×160 small-object detection head was added,significantly boosting detection accuracy for small targets.In the backbone,the C2f_Dual(Cross Stage Partial with Cross-Stage Feature Fusion Dual)module and SPPELAN(Spatial Pyramid Pooling with Enhanced LocalAttentionNetwork)modulewere integrated.These components improve feature extraction and information aggregationwhile reducing parameters and computational complexity,enhancing inference efficiency.Additionally,Shape-IoU(Shape Intersection over Union)is used as the loss function for bounding box regression,enabling more precise shape-based object matching.Experimental results on the VisDrone 2019 dataset demonstrate the effectiveness ofDAFPN-YOLO.Compared to YOLOv8s,the proposedmodel achieves a 5.4 percentage point increase inmAP@0.5,a 3.8 percentage point improvement in mAP@0.5:0.95,and a 17.2%reduction in parameter count.These results highlight DAFPN-YOLO’s advantages in UAV-based object detection,offering valuable insights for applying deep learning to UAV-specific multi-object detection tasks.展开更多
Current spatio-temporal action detection methods lack sufficient capabilities in extracting and comprehending spatio-temporal information. This paper introduces an end-to-end Adaptive Cross-Scale Fusion Encoder-Decode...Current spatio-temporal action detection methods lack sufficient capabilities in extracting and comprehending spatio-temporal information. This paper introduces an end-to-end Adaptive Cross-Scale Fusion Encoder-Decoder (ACSF-ED) network to predict the action and locate the object efficiently. In the Adaptive Cross-Scale Fusion Spatio-Temporal Encoder (ACSF ST-Encoder), the Asymptotic Cross-scale Feature-fusion Module (ACCFM) is designed to address the issue of information degradation caused by the propagation of high-level semantic information, thereby extracting high-quality multi-scale features to provide superior features for subsequent spatio-temporal information modeling. Within the Shared-Head Decoder structure, a shared classification and regression detection head is constructed. A multi-constraint loss function composed of one-to-one, one-to-many, and contrastive denoising losses is designed to address the problem of insufficient constraint force in predicting results with traditional methods. This loss function enhances the accuracy of model classification predictions and improves the proximity of regression position predictions to ground truth objects. The proposed method model is evaluated on the popular dataset UCF101-24 and JHMDB-21. Experimental results demonstrate that the proposed method achieves an accuracy of 81.52% on the Frame-mAP metric, surpassing current existing methods.展开更多
Grating-based X-ray phase-contrast imaging enhances the contrast of imaged objects,particularly soft tissues.However,the radiation dose in computed tomography(CT)is generally excessive owing to the complex collection ...Grating-based X-ray phase-contrast imaging enhances the contrast of imaged objects,particularly soft tissues.However,the radiation dose in computed tomography(CT)is generally excessive owing to the complex collection scheme.Sparse-view CT collection reduces the radiation dose,but with reduced resolution and reconstructed artifacts particularly in analytical reconstruction methods.Recently,deep learning has been employed in sparse-view CT reconstruction and achieved stateof-the-art results.Nevertheless,its low generalization performance and requirement for abundant training datasets have hindered the practical application of deep learning in phase-contrast CT.In this study,a CT model was used to generate a substantial number of simulated training datasets,thereby circumventing the need for experimental datasets.By training a network with simulated training datasets,the proposed method achieves high generalization performance in attenuationbased CT and phase-contrast CT,despite the lack of sufficient experimental datasets.In experiments utilizing only half of the CT data,our proposed method obtained an image quality comparable to that of the filtered back-projection algorithm with full-view projection.The proposed method simultaneously addresses two challenges in phase-contrast three-dimensional imaging,namely the lack of experimental datasets and the high exposure dose,through model-driven deep learning.This method significantly accelerates the practical application of phase-contrast CT.展开更多
Aiming at problems such as large errors and low efficiency in manual counting of drill pipes during drilling depth measurement,an intelligent detection and counting method for the small targets at the end of drill pip...Aiming at problems such as large errors and low efficiency in manual counting of drill pipes during drilling depth measurement,an intelligent detection and counting method for the small targets at the end of drill pipes based on the improved YOLO11n is proposed.This method realizes the high-precision detection of targets at drill pipe ends in the image by optimizing the target detection model,and combines a post-processing correction mechanism to improve the drill pipe counting accuracy.In order to alleviate the low-precision problem of YOLO11n algorithm for small target recognition in the complex underground background,the YOLO11n algorithm is improved.First,the key module C3k2 in the backbone network was improved,and Poly Kernel Inception(PKI)Block was introduced to replace Bottleneck in it to fully integrate the target context information and the model’s capability of feature extraction;Second,within the model’s neck network,a new feature fusion pyramid ISOP(Improved Small Object Pyramid)is proposed,SPDConv is introduced to strengthen the P2 feature,and CSP and OmniKernel are combined to integrate multi-scale features;Finally,the default loss function is substituted with Powerful-IoU(PIoU)to solve the anchor box expansion problem.On the self-built dataset,experimental verification was conducted.The findings showed that the Recall rose by 6.4%,mAP@0.5 increased by 4.5%,and mAP@0.5:0.95 improved by 6%compared with the baseline model,effectively solving the issues of false detection and missed detection problems in small target detection task.Meanwhile,we conducted counting tests on drilling videos from 5 different scenarios,achieving an average accuracy of 97.3%,which meets the accuracy needs for drill pipe recognition and counting in coal mine drilling sites.The research findings offer theoretical basis and technical backing for promoting the intelligent development of coal mine gas extraction drilling sites.展开更多
Accurate prediction of ground surface settlement(GSS)adjacent to an excavation is important to prevent potential damage to the surrounding environment.Previous studies have extensively delved into this topic but all u...Accurate prediction of ground surface settlement(GSS)adjacent to an excavation is important to prevent potential damage to the surrounding environment.Previous studies have extensively delved into this topic but all under the limitations of either imprecise theories or insufficient data.In the present study,we proposed a physics-constrained neural network(PhyNN)for predicting excavation-induced GSS to fully integrate the theory of elasticity with observations and make full use of the strong fitting ability of neural networks(NNs).This model incorporates an analytical solution as an additional regularization term in the loss function to guide the training of NN.Moreover,we introduced three trainable parameters into the analytical solution so that it can be adaptively modified during the training process.The performance of the proposed PhyNN model is verified using data from a case study project.Results show that our PhyNN model achieves higher prediction accuracy,better generalization ability,and robustness than the purely data-driven NN model when confronted with data containing noise and outliers.Remarkably,by incorporating physical constraints,the admissible solution space of PhyNN is significantly narrowed,leading to a substantial reduction in the need for the amount of training data.The proposed PhyNN can be utilized as a general framework for integrating physical constraints into data-driven machine-learning models.展开更多
Learning from demonstration is widely regarded as a promising paradigm for robots to acquire diverse skills.Other than the artificial learning from observation-action pairs for machines,humans can learn to imitate in ...Learning from demonstration is widely regarded as a promising paradigm for robots to acquire diverse skills.Other than the artificial learning from observation-action pairs for machines,humans can learn to imitate in a more versatile and effective manner:acquiring skills through mere“observation”.Video to Command task is widely perceived as a promising approach for task-based learning,which yet faces two key challenges:(1)High redundancy and low frame rate of fine-grained action sequences make it difficult to manipulate objects robustly and accurately.(2)Video to Command models often prioritize accuracy and richness of output commands over physical capabilities,leading to impractical or unsafe instructions for robots.This article presents a novel Video to Command framework that employs multiple data associations and physical constraints.First,we introduce an object-level appearancecontrasting multiple data association strategy to effectively associate manipulated objects in visually complex environments,capturing dynamic changes in video content.Then,we propose a multi-task Video to Command model that utilizes object-level video content changes to compile expert demonstrations into manipulation commands.Finally,a multi-task hybrid loss function is proposed to train a Video to Command model that adheres to the constraints of the physical world and manipulation tasks.Our method achieved over 10%on BLEU_N,METEOR,ROUGE_L,and CIDEr compared to the up-to-date methods.The dual-arm robot prototype was established to demonstrate the whole process of learning from an expert demonstration of multiple skills and then executing the tasks by a robot.展开更多
Segmentation of the retinal vessels in the fundus is crucial for diagnosing ocular diseases.Retinal vessel images often suffer from category imbalance and large scale variations.This ultimately results in incomplete v...Segmentation of the retinal vessels in the fundus is crucial for diagnosing ocular diseases.Retinal vessel images often suffer from category imbalance and large scale variations.This ultimately results in incomplete vessel segmentation and poor continuity.In this study,we propose CT-MFENet to address the aforementioned issues.First,the use of context transformer(CT)allows for the integration of contextual feature information,which helps establish the connection between pixels and solve the problem of incomplete vessel continuity.Second,multi-scale dense residual networks are used instead of traditional CNN to address the issue of inadequate local feature extraction when the model encounters vessels at multiple scales.In the decoding stage,we introduce a local-global fusion module.It enhances the localization of vascular information and reduces the semantic gap between high-and low-level features.To address the class imbalance in retinal images,we propose a hybrid loss function that enhances the segmentation ability of the model for topological structures.We conducted experiments on the publicly available DRIVE,CHASEDB1,STARE,and IOSTAR datasets.The experimental results show that our CT-MFENet performs better than most existing methods,including the baseline U-Net.展开更多
Ufmylation is an ubiquitin-like post-translational modification characterized by the covalent binding of mature UFM1 to target proteins.Although the consequences of ufmylation on target proteins are not fully understo...Ufmylation is an ubiquitin-like post-translational modification characterized by the covalent binding of mature UFM1 to target proteins.Although the consequences of ufmylation on target proteins are not fully understood,its importance is evident from the disorders resulting from its dysfunction.Numerous case reports have established a link between biallelic loss-of-function and/or hypomorphic variants in ufmylation-related genes and a spectrum of pediatric neurodevelopmental disorders.展开更多
Current you only look once(YOLO)-based algorithm model is facing the challenge of overwhelming parameters and calculation complexity under the printed circuit board(PCB)defect detection application scenario.In order t...Current you only look once(YOLO)-based algorithm model is facing the challenge of overwhelming parameters and calculation complexity under the printed circuit board(PCB)defect detection application scenario.In order to solve this problem,we propose a new method,which combined the lightweight network mobile vision transformer(Mobile Vi T)with the convolutional block attention module(CBAM)mechanism and the new regression loss function.This method needed less computation resources,making it more suitable for embedded edge detection devices.Meanwhile,the new loss function improved the positioning accuracy of the bounding box and enhanced the robustness of the model.In addition,experiments on public datasets demonstrate that the improved model achieves an average accuracy of 87.9%across six typical defect detection tasks,while reducing computational costs by nearly 90%.It significantly reduces the model's computational requirements while maintaining accuracy,ensuring reliable performance for edge deployment.展开更多
This study presents a groundbreaking method named Expo-GAN(Exposition-Generative Adversarial Network)for style transfer in exhibition hall design,using a refined version of the Cycle Generative Adversarial Network(Cyc...This study presents a groundbreaking method named Expo-GAN(Exposition-Generative Adversarial Network)for style transfer in exhibition hall design,using a refined version of the Cycle Generative Adversarial Network(CycleGAN).The primary goal is to enhance the transformation of image styles while maintaining visual consistency,an areawhere current CycleGAN models often fall short.These traditionalmodels typically face difficulties in accurately capturing expansive features as well as the intricate stylistic details necessary for high-quality image transformation.To address these limitations,the research introduces several key modifications to the CycleGAN architecture.Enhancements to the generator involve integrating U-net with SpecTransformer modules.This integration incorporates the use of Fourier transform techniques coupled with multi-head self-attention mechanisms,which collectively improve the generator’s ability to depict both large-scale structural patterns and minute elements meticulously in the generated images.This enhancement allows the generator to achieve a more detailed and coherent fusion of styles,essential for exhibition hall designs where both broad aesthetic strokes and detailed nuances matter significantly.The study also proposes innovative changes to the discriminator by employing dilated convolution and global attention mechanisms.These are derived using the Differentiable Architecture Search(DARTS)Neural Architecture Search framework to expand the receptive field,which is crucial for recognizing comprehensive artistically styled images.By broadening the ability to discern complex artistic features,the model avoids previous pitfalls associated with style inconsistency and missing detailed features.Moreover,the traditional cyde-consistency loss function is replaced with the Learned Perceptual Image Patch Similarity(LPIPS)metric.This shift aims to significantly enhance the perceptual quality of the resultant images by prioritizing human-perceived similarities,which aligns better with user expectations and professional standards in design aesthetics.The experimental phase of this research demonstrates that this novel approach consistently outperforms the conventional CycleGAN across a broad range of datasets.Complementary ablation studies and qualitative assessments underscore its superiority,particularly in maintaining detail fidelity and style continuity.This is critical for creating a visually harmonious exhibitionhall designwhere everydetail contributes to the overall aesthetic appeal.The results illustrate that this refined approach effectively bridges the gap between technical capability and artistic necessity,marking a significant advancement in computational design methodologies.展开更多
基金supported by the National Key Research and Development Program of China(Grant numbers:2021YFF0901705,2021YFF0901700)the State Key Laboratory of Media Convergence and Communication,Communication University of China+1 种基金the Fundamental Research Funds for the Central Universitiesthe High-Quality and Cutting-Edge Disciplines Construction Project for Universities in Beijing(Internet Information,Communication University of China).
文摘In the Internet era,recommendation systems play a crucial role in helping users find relevant information from large datasets.Class imbalance is known to severely affect data quality,and therefore reduce the performance of recommendation systems.Due to the imbalance,machine learning algorithms tend to classify inputs into the positive(majority)class every time to achieve high prediction accuracy.Imbalance can be categorized such as by features and classes,but most studies consider only class imbalance.In this paper,we propose a recommendation system that can integrate multiple networks to adapt to a large number of imbalanced features and can deal with highly skewed and imbalanced datasets through a loss function.We propose a loss aware feature attention mechanism(LAFAM)to solve the issue of feature imbalance.The network incorporates an attention mechanism and uses multiple sub-networks to classify and learn features.For better results,the network can learn the weights of sub-networks and assign higher weights to important features.We propose suppression loss to address class imbalance,which favors negative loss by penalizing positive loss,and pays more attention to sample points near the decision boundary.Experiments on two large-scale datasets verify that the performance of the proposed system is greatly improved compared to baseline methods.
基金support from the following institutional grant.Internal Grant Agency of the Faculty of Economics and Management,Czech University of Life Sciences Prague,grant no.2023A0004(https://iga.pef.czu.cz/,accessed on 6 June 2025).
文摘This study proposes a new component of the composite loss function minimised during training of the Super-Resolution(SR)algorithms—the normalised structural similarity index loss LSSIMN,which has the potential to improve the natural appearance of reconstructed images.Deep learning-based super-resolution(SR)algorithms reconstruct high-resolution images from low-resolution inputs,offering a practical means to enhance image quality without requiring superior imaging hardware,which is particularly important in medical applications where diagnostic accuracy is critical.Although recent SR methods employing convolutional and generative adversarial networks achieve high pixel fidelity,visual artefacts may persist,making the design of the loss function during training essential for ensuring reliable and naturalistic image reconstruction.Our research shows on two models—SR and Invertible Rescaling Neural Network(IRN)—trained on multiple benchmark datasets that the function LSSIMN significantly contributes to the visual quality,preserving the structural fidelity on the reference datasets.The quantitative analysis of results while incorporating LSSIMN shows that including this loss function component has a mean 2.88%impact on the improvement of the final structural similarity of the reconstructed images in the validation set,in comparison to leaving it out and 0.218%in comparison when this component is non-normalised.
基金supported by Chongqing Municipal Commission of Housing and Urban-Rural Development(Grant No.CKZ2024-87)China Chongqing Municipal Science and Technology Bureau(Grant No.2024TIAD-CYKJCXX0121).
文摘Currently,challenges such as small object size and occlusion lead to a lack of accuracy and robustness in small object detection.Since small objects occupy only a few pixels in an image,the extracted features are limited,and mainstream downsampling convolution operations further exacerbate feature loss.Additionally,due to the occlusionprone nature of small objects and their higher sensitivity to localization deviations,conventional Intersection over Union(IoU)loss functions struggle to achieve stable convergence.To address these limitations,LR-Net is proposed for small object detection.Specifically,the proposed Lossless Feature Fusion(LFF)method transfers spatial features into the channel domain while leveraging a hybrid attentionmechanism to focus on critical features,mitigating feature loss caused by downsampling.Furthermore,RSIoU is proposed to enhance the convergence performance of IoU-based losses for small objects.RSIoU corrects the inherent convergence direction issues in SIoU and proposes a penalty term as a Dynamic Focusing Mechanism parameter,enabling it to dynamically emphasize the loss contribution of small object samples.Ultimately,RSIoU significantly improves the convergence performance of the loss function for small objects,particularly under occlusion scenarios.Experiments demonstrate that LR-Net achieves significant improvements across variousmetrics onmultiple datasets compared with YOLOv8n,achieving a 3.7% increase in mean Average Precision(AP)on the VisDrone2019 dataset,along with improvements of 3.3% on the AI-TOD dataset and 1.2% on the COCO dataset.
基金partially funded by the National Natural Science Foundation of China(U2142205)the Guangdong Major Project of Basic and Applied Basic Research(2020B0301030004)+1 种基金the Special Fund for Forecasters of China Meteorological Administration(CMAYBY2020-094)the Graduate Student Research and Innovation Program of Central South University(2023ZZTS0347)。
文摘Traditional meteorological downscaling methods face limitations due to the complex distribution of meteorological variables,which can lead to unstable forecasting results,especially in extreme scenarios.To overcome this issue,we propose a convolutional graph neural network(CGNN)model,which we enhance with multilayer feature fusion and a squeeze-and-excitation block.Additionally,we introduce a spatially balanced mean squared error(SBMSE)loss function to address the imbalanced distribution and spatial variability of meteorological variables.The CGNN is capable of extracting essential spatial features and aggregating them from a global perspective,thereby improving the accuracy of prediction and enhancing the model's generalization ability.Based on the experimental results,CGNN has certain advantages in terms of bias distribution,exhibiting a smaller variance.When it comes to precipitation,both UNet and AE also demonstrate relatively small biases.As for temperature,AE and CNNdense perform outstandingly during the winter.The time correlation coefficients show an improvement of at least 10%at daily and monthly scales for both temperature and precipitation.Furthermore,the SBMSE loss function displays an advantage over existing loss functions in predicting the98th percentile and identifying areas where extreme events occur.However,the SBMSE tends to overestimate the distribution of extreme precipitation,which may be due to the theoretical assumptions about the posterior distribution of data that partially limit the effectiveness of the loss function.In future work,we will further optimize the SBMSE to improve prediction accuracy.
基金Jilin Science and Technology Development Plan Project(No.20200403075SF)Doctoral Research Start-Up Fund of Northeast Electric Power University(No.BSJXM-2018202).
文摘The current existing problem of deep learning framework for the detection and segmentation of electrical equipment is dominantly related to low precision.Because of the reliable,safe and easy-to-operate technology provided by deep learning-based video surveillance for unmanned inspection of electrical equipment,this paper uses the bottleneck attention module(BAM)attention mechanism to improve the Solov2 model and proposes a new electrical equipment segmentation mode.Firstly,the BAM attention mechanism is integrated into the feature extraction network to adaptively learn the correlation between feature channels,thereby improving the expression ability of the feature map;secondly,the weighted sum of CrossEntropy Loss and Dice loss is designed as the mask loss to improve the segmentation accuracy and robustness of the model;finally,the non-maximal suppression(NMS)algorithm to better handle the overlap problem in instance segmentation.Experimental results show that the proposed method achieves an average segmentation accuracy of mAP of 80.4% on three types of electrical equipment datasets,including transformers,insulators and voltage transformers,which improve the detection accuracy by more than 5.7% compared with the original Solov2 model.The segmentation model proposed can provide a focusing technical means for the intelligent management of power systems.
基金supported by the National Key R&D Program of China(No.2023YFC3805100)the National Natural Science Foundation of China(Nos.52222811 and 52494963)。
文摘Earthquakes pose significant perils to the built environment in urban areas.To avert the calamitous aftermath of earthquakes,it is imperative to construct seismic resilient cities.Due to the intricacy of the concept of urban seismic resilience(USR),its assessment is a large-scale system engineering issue.The assessment of USR should be based on the notion of urban seismic capacity(USC)assessment,which includes casualties,economic loss,and recovery time as criteria.Functionality loss is also included in the assessment of USR in addition to these criteria.The assessment indicator system comprising five dimensions(building and lifeline infrastructure,environment,society,economy,and institution)and 20 indicators has been devised to quantify USR.The analytical hierarchy process(AHP)is utilized to compute the weights of the criteria,dimensions,and indicators in the urban seismic resilience assessment(USRA)indicator system.When the necessary data for a city are obtainable,the seismic resilience of that city can be assessed using this framework.To illustrate the proposed methodology,a moderate-sized city in China was selected as a case study.The assessment results indicate a high level of USR,suggesting that the city possesses strong capabilities to withstand and recover from potential future earthquakes.
基金The National Key Research and Development Program of China(No.2023YFC3805003)。
文摘To quantify the seismic resilience of buildings,a method for evaluating functional loss from the component level to the overall building is proposed,and the dual-parameter seismic resilience assessment method based on postearthquake loss and recovery time is improved.A threelevel function tree model is established,which can consider the dynamic changes in weight coefficients of different category of components relative to their functional losses.Bayesian networks are utilized to quantify the impact of weather conditions,construction technology levels,and worker skill levels on component repair time.A method for determining the real-time functional recovery curve of buildings based on the component repair process is proposed.Taking a three-story teaching building as an example,the seismic resilience indices under basic earthquakes and rare earthquakes are calculated.The results show that the seismic resilience grade of the teaching building is comprehensively judged as GradeⅢ,and its resilience grade is more significantly affected by postearthquake loss.The proposed method can be used to predict the seismic resilience of buildings prior to earthquakes,identify weak components within buildings,and provide guidance for taking measures to enhance the seismic resilience of buildings.
文摘Title:A dual-parameter method for seismic resilience assessment of buildings Authors:LI Shuang;HU Binbin;LIU Wen;ZHAI Changhai Abstract:To quantify the seismic resilience of buildings,a method for evaluating functional loss from the component level to the overall building is proposed,and the dual-parameter seismic resilience assessment method based on postearthquake loss and recovery time is improved.A three-level function tree model is established,which can consider the dynamic changes in weight coefficients of different category of components relative to their functional losses.Bayesian networks are utilized to quantify the impact of weather conditions,construction technology levels,and worker skill levels on component repair time.A method for determining the real-time functional recovery curve of buildings based on the component repair process is proposed.Taking a three-story teaching building as an example,the seismic resilience indices under basic earthquakes and rare earthquakes are calculated.The results show that the seismic resilience grade of the teaching building is comprehensively judged as GradeⅢ,and its resilience grade is more significantly affected by postearthquake loss.The proposed method can be used to predict the seismic resilience of buildings prior to earthquakes,identify weak components within buildings,and provide guidance for taking measures to enhance the seismic resilience of buildings.
文摘In this paper we propose an absolute error loss EB estimator for parameter of one-side truncation distribution families. Under some conditions we have proved that the convergence rates of its Bayes risk is o, where 0<λ,r≤1,Mn≤lnln n (for large n),Mn→∞ as n→∞.
基金funded by National Natural Science Foundation of China(61741303)the Foundation Project of Guangxi Key Laboratory of Spatial Information andMapping(No.21-238-21-16).
文摘Detecting individuals wearing safety helmets in complex environments faces several challenges.These factors include limited detection accuracy and frequent missed or false detections.Additionally,existing algorithms often have excessive parameter counts,complex network structures,and high computational demands.These challenges make it difficult to deploy such models efficiently on resource-constrained devices like embedded systems.Aiming at this problem,this research proposes an optimized and lightweight solution called FGP-YOLOv8,an improved version of YOLOv8n.The YOLOv8 backbone network is replaced with the FasterNet model to reduce parameters and computational demands while local convolution layers are added.This modification minimizes computational costs with only a minor impact on accuracy.A new GSTA(GSConv-Triplet Attention)module is introduced to enhance feature fusion and reduce computational complexity.This is achieved using attention weights generated from dimensional interactions within the feature map.Additionally,the ParNet-C2f module replaces the original C2f(CSP Bottleneck with 2 Convolutions)module,improving feature extraction for safety helmets of various shapes and sizes.The CIoU(Complete-IoU)is replaced with the WIoU(Wise-IoU)to boost performance further,enhancing detection accuracy and generalization capabilities.Experimental results validate the improvements.The proposedmodel reduces the parameter count by 19.9% and the computational load by 18.5%.At the same time,mAP(mean average precision)increases by 2.3%,and precision improves by 1.2%.These results demonstrate the model’s robust performance in detecting safety helmets across diverse environments.
基金supported by the National Natural Science Foundation of China(Grant Nos.62101275 and 62101274).
文摘UAV-based object detection is rapidly expanding in both civilian and military applications,including security surveillance,disaster assessment,and border patrol.However,challenges such as small objects,occlusions,complex backgrounds,and variable lighting persist due to the unique perspective of UAV imagery.To address these issues,this paper introduces DAFPN-YOLO,an innovative model based on YOLOv8s(You Only Look Once version 8s).Themodel strikes a balance between detection accuracy and speed while reducing parameters,making itwell-suited for multi-object detection tasks from drone perspectives.A key feature of DAFPN-YOLO is the enhanced Drone-AFPN(Adaptive Feature Pyramid Network),which adaptively fuses multi-scale features to optimize feature extraction and enhance spatial and small-object information.To leverage Drone-AFPN’smulti-scale capabilities fully,a dedicated 160×160 small-object detection head was added,significantly boosting detection accuracy for small targets.In the backbone,the C2f_Dual(Cross Stage Partial with Cross-Stage Feature Fusion Dual)module and SPPELAN(Spatial Pyramid Pooling with Enhanced LocalAttentionNetwork)modulewere integrated.These components improve feature extraction and information aggregationwhile reducing parameters and computational complexity,enhancing inference efficiency.Additionally,Shape-IoU(Shape Intersection over Union)is used as the loss function for bounding box regression,enabling more precise shape-based object matching.Experimental results on the VisDrone 2019 dataset demonstrate the effectiveness ofDAFPN-YOLO.Compared to YOLOv8s,the proposedmodel achieves a 5.4 percentage point increase inmAP@0.5,a 3.8 percentage point improvement in mAP@0.5:0.95,and a 17.2%reduction in parameter count.These results highlight DAFPN-YOLO’s advantages in UAV-based object detection,offering valuable insights for applying deep learning to UAV-specific multi-object detection tasks.
基金support for this work was supported by Key Lab of Intelligent and Green Flexographic Printing under Grant ZBKT202301.
文摘Current spatio-temporal action detection methods lack sufficient capabilities in extracting and comprehending spatio-temporal information. This paper introduces an end-to-end Adaptive Cross-Scale Fusion Encoder-Decoder (ACSF-ED) network to predict the action and locate the object efficiently. In the Adaptive Cross-Scale Fusion Spatio-Temporal Encoder (ACSF ST-Encoder), the Asymptotic Cross-scale Feature-fusion Module (ACCFM) is designed to address the issue of information degradation caused by the propagation of high-level semantic information, thereby extracting high-quality multi-scale features to provide superior features for subsequent spatio-temporal information modeling. Within the Shared-Head Decoder structure, a shared classification and regression detection head is constructed. A multi-constraint loss function composed of one-to-one, one-to-many, and contrastive denoising losses is designed to address the problem of insufficient constraint force in predicting results with traditional methods. This loss function enhances the accuracy of model classification predictions and improves the proximity of regression position predictions to ground truth objects. The proposed method model is evaluated on the popular dataset UCF101-24 and JHMDB-21. Experimental results demonstrate that the proposed method achieves an accuracy of 81.52% on the Frame-mAP metric, surpassing current existing methods.
基金supported by the National Natural Science Foundation of China(Nos.U2032148,U2032157,11775224)USTC Research Funds of the Double First-Class Initiative(No.YD2310002008)the National Key Research and Development Program of China(No.2017YFA0402904),the Youth Innovation Promotion Association,CAS(No.2020457)。
文摘Grating-based X-ray phase-contrast imaging enhances the contrast of imaged objects,particularly soft tissues.However,the radiation dose in computed tomography(CT)is generally excessive owing to the complex collection scheme.Sparse-view CT collection reduces the radiation dose,but with reduced resolution and reconstructed artifacts particularly in analytical reconstruction methods.Recently,deep learning has been employed in sparse-view CT reconstruction and achieved stateof-the-art results.Nevertheless,its low generalization performance and requirement for abundant training datasets have hindered the practical application of deep learning in phase-contrast CT.In this study,a CT model was used to generate a substantial number of simulated training datasets,thereby circumventing the need for experimental datasets.By training a network with simulated training datasets,the proposed method achieves high generalization performance in attenuationbased CT and phase-contrast CT,despite the lack of sufficient experimental datasets.In experiments utilizing only half of the CT data,our proposed method obtained an image quality comparable to that of the filtered back-projection algorithm with full-view projection.The proposed method simultaneously addresses two challenges in phase-contrast three-dimensional imaging,namely the lack of experimental datasets and the high exposure dose,through model-driven deep learning.This method significantly accelerates the practical application of phase-contrast CT.
基金Henan Province University Science and Technology Innovation Team Support Program Project(22IRTSTHN005)..
文摘Aiming at problems such as large errors and low efficiency in manual counting of drill pipes during drilling depth measurement,an intelligent detection and counting method for the small targets at the end of drill pipes based on the improved YOLO11n is proposed.This method realizes the high-precision detection of targets at drill pipe ends in the image by optimizing the target detection model,and combines a post-processing correction mechanism to improve the drill pipe counting accuracy.In order to alleviate the low-precision problem of YOLO11n algorithm for small target recognition in the complex underground background,the YOLO11n algorithm is improved.First,the key module C3k2 in the backbone network was improved,and Poly Kernel Inception(PKI)Block was introduced to replace Bottleneck in it to fully integrate the target context information and the model’s capability of feature extraction;Second,within the model’s neck network,a new feature fusion pyramid ISOP(Improved Small Object Pyramid)is proposed,SPDConv is introduced to strengthen the P2 feature,and CSP and OmniKernel are combined to integrate multi-scale features;Finally,the default loss function is substituted with Powerful-IoU(PIoU)to solve the anchor box expansion problem.On the self-built dataset,experimental verification was conducted.The findings showed that the Recall rose by 6.4%,mAP@0.5 increased by 4.5%,and mAP@0.5:0.95 improved by 6%compared with the baseline model,effectively solving the issues of false detection and missed detection problems in small target detection task.Meanwhile,we conducted counting tests on drilling videos from 5 different scenarios,achieving an average accuracy of 97.3%,which meets the accuracy needs for drill pipe recognition and counting in coal mine drilling sites.The research findings offer theoretical basis and technical backing for promoting the intelligent development of coal mine gas extraction drilling sites.
基金funded by the National Natural Science Foundation of China(Grant No.52090082)support provided by the China Scholarship Council(Grant No.202206260206).
文摘Accurate prediction of ground surface settlement(GSS)adjacent to an excavation is important to prevent potential damage to the surrounding environment.Previous studies have extensively delved into this topic but all under the limitations of either imprecise theories or insufficient data.In the present study,we proposed a physics-constrained neural network(PhyNN)for predicting excavation-induced GSS to fully integrate the theory of elasticity with observations and make full use of the strong fitting ability of neural networks(NNs).This model incorporates an analytical solution as an additional regularization term in the loss function to guide the training of NN.Moreover,we introduced three trainable parameters into the analytical solution so that it can be adaptively modified during the training process.The performance of the proposed PhyNN model is verified using data from a case study project.Results show that our PhyNN model achieves higher prediction accuracy,better generalization ability,and robustness than the purely data-driven NN model when confronted with data containing noise and outliers.Remarkably,by incorporating physical constraints,the admissible solution space of PhyNN is significantly narrowed,leading to a substantial reduction in the need for the amount of training data.The proposed PhyNN can be utilized as a general framework for integrating physical constraints into data-driven machine-learning models.
基金Supported by Zhejiang Provincial Key Research and Development Program(Grant No.2021C04015)。
文摘Learning from demonstration is widely regarded as a promising paradigm for robots to acquire diverse skills.Other than the artificial learning from observation-action pairs for machines,humans can learn to imitate in a more versatile and effective manner:acquiring skills through mere“observation”.Video to Command task is widely perceived as a promising approach for task-based learning,which yet faces two key challenges:(1)High redundancy and low frame rate of fine-grained action sequences make it difficult to manipulate objects robustly and accurately.(2)Video to Command models often prioritize accuracy and richness of output commands over physical capabilities,leading to impractical or unsafe instructions for robots.This article presents a novel Video to Command framework that employs multiple data associations and physical constraints.First,we introduce an object-level appearancecontrasting multiple data association strategy to effectively associate manipulated objects in visually complex environments,capturing dynamic changes in video content.Then,we propose a multi-task Video to Command model that utilizes object-level video content changes to compile expert demonstrations into manipulation commands.Finally,a multi-task hybrid loss function is proposed to train a Video to Command model that adheres to the constraints of the physical world and manipulation tasks.Our method achieved over 10%on BLEU_N,METEOR,ROUGE_L,and CIDEr compared to the up-to-date methods.The dual-arm robot prototype was established to demonstrate the whole process of learning from an expert demonstration of multiple skills and then executing the tasks by a robot.
基金the National Natural Science Foundation of China(No.62266025)。
文摘Segmentation of the retinal vessels in the fundus is crucial for diagnosing ocular diseases.Retinal vessel images often suffer from category imbalance and large scale variations.This ultimately results in incomplete vessel segmentation and poor continuity.In this study,we propose CT-MFENet to address the aforementioned issues.First,the use of context transformer(CT)allows for the integration of contextual feature information,which helps establish the connection between pixels and solve the problem of incomplete vessel continuity.Second,multi-scale dense residual networks are used instead of traditional CNN to address the issue of inadequate local feature extraction when the model encounters vessels at multiple scales.In the decoding stage,we introduce a local-global fusion module.It enhances the localization of vascular information and reduces the semantic gap between high-and low-level features.To address the class imbalance in retinal images,we propose a hybrid loss function that enhances the segmentation ability of the model for topological structures.We conducted experiments on the publicly available DRIVE,CHASEDB1,STARE,and IOSTAR datasets.The experimental results show that our CT-MFENet performs better than most existing methods,including the baseline U-Net.
文摘Ufmylation is an ubiquitin-like post-translational modification characterized by the covalent binding of mature UFM1 to target proteins.Although the consequences of ufmylation on target proteins are not fully understood,its importance is evident from the disorders resulting from its dysfunction.Numerous case reports have established a link between biallelic loss-of-function and/or hypomorphic variants in ufmylation-related genes and a spectrum of pediatric neurodevelopmental disorders.
基金supported by the National Natural Science Foundation of China(Nos.62373215,62373219 and 62073193)the Natural Science Foundation of Shandong Province(No.ZR2023MF100)+1 种基金the Key Projects of the Ministry of Industry and Information Technology(No.TC220H057-2022)the Independently Developed Instrument Funds of Shandong University(No.zy20240201)。
文摘Current you only look once(YOLO)-based algorithm model is facing the challenge of overwhelming parameters and calculation complexity under the printed circuit board(PCB)defect detection application scenario.In order to solve this problem,we propose a new method,which combined the lightweight network mobile vision transformer(Mobile Vi T)with the convolutional block attention module(CBAM)mechanism and the new regression loss function.This method needed less computation resources,making it more suitable for embedded edge detection devices.Meanwhile,the new loss function improved the positioning accuracy of the bounding box and enhanced the robustness of the model.In addition,experiments on public datasets demonstrate that the improved model achieves an average accuracy of 87.9%across six typical defect detection tasks,while reducing computational costs by nearly 90%.It significantly reduces the model's computational requirements while maintaining accuracy,ensuring reliable performance for edge deployment.
文摘This study presents a groundbreaking method named Expo-GAN(Exposition-Generative Adversarial Network)for style transfer in exhibition hall design,using a refined version of the Cycle Generative Adversarial Network(CycleGAN).The primary goal is to enhance the transformation of image styles while maintaining visual consistency,an areawhere current CycleGAN models often fall short.These traditionalmodels typically face difficulties in accurately capturing expansive features as well as the intricate stylistic details necessary for high-quality image transformation.To address these limitations,the research introduces several key modifications to the CycleGAN architecture.Enhancements to the generator involve integrating U-net with SpecTransformer modules.This integration incorporates the use of Fourier transform techniques coupled with multi-head self-attention mechanisms,which collectively improve the generator’s ability to depict both large-scale structural patterns and minute elements meticulously in the generated images.This enhancement allows the generator to achieve a more detailed and coherent fusion of styles,essential for exhibition hall designs where both broad aesthetic strokes and detailed nuances matter significantly.The study also proposes innovative changes to the discriminator by employing dilated convolution and global attention mechanisms.These are derived using the Differentiable Architecture Search(DARTS)Neural Architecture Search framework to expand the receptive field,which is crucial for recognizing comprehensive artistically styled images.By broadening the ability to discern complex artistic features,the model avoids previous pitfalls associated with style inconsistency and missing detailed features.Moreover,the traditional cyde-consistency loss function is replaced with the Learned Perceptual Image Patch Similarity(LPIPS)metric.This shift aims to significantly enhance the perceptual quality of the resultant images by prioritizing human-perceived similarities,which aligns better with user expectations and professional standards in design aesthetics.The experimental phase of this research demonstrates that this novel approach consistently outperforms the conventional CycleGAN across a broad range of datasets.Complementary ablation studies and qualitative assessments underscore its superiority,particularly in maintaining detail fidelity and style continuity.This is critical for creating a visually harmonious exhibitionhall designwhere everydetail contributes to the overall aesthetic appeal.The results illustrate that this refined approach effectively bridges the gap between technical capability and artistic necessity,marking a significant advancement in computational design methodologies.
