Image-based similar trademark retrieval is a time-consuming and labor-intensive task in the trademark examination process.This paper aims to support trademark examiners by training Deep Convolutional Neural Network(DC...Image-based similar trademark retrieval is a time-consuming and labor-intensive task in the trademark examination process.This paper aims to support trademark examiners by training Deep Convolutional Neural Network(DCNN)models for effective Trademark Image Retrieval(TIR).To achieve this goal,we first develop a novel labeling method that automatically generates hundreds of thousands of labeled similar and dissimilar trademark image pairs using accompanying data fields such as citation lists,Vienna classification(VC)codes,and trademark ownership information.This approach eliminates the need for manual labeling and provides a large-scale dataset suitable for training deep learning models.We then train DCNN models based on Siamese and Triplet architectures,evaluating various feature extractors to determine the most effective configuration.Furthermore,we present an Adapted Contrastive Loss Function(ACLF)for the trademark retrieval task,specifically engineered to mitigate the influence of noisy labels found in automatically created datasets.Experimental results indicate that our proposed model(Efficient-Net_v21_Siamese)performs best at both True Negative Rate(TNR)threshold levels,TNR 0.9 and TNR 0.95,with==respective True Positive Rates(TPRs)of 77.7%and 70.8%and accuracies of 83.9%and 80.4%.Additionally,when testing on the public trademark dataset METU_v2,our model achieves a normalized average rank(NAR)of 0.0169,outperforming the current state-of-the-art(SOTA)model.Based on these findings,we estimate that considering only approximately 10%of the returned trademarks would be sufficient,significantly reducing the review time.Therefore,the paper highlights the potential of utilizing national trademark data to enhance the accuracy and efficiency of trademark retrieval systems,ultimately supporting trademark examiners in their evaluation tasks.展开更多
Content-Based Image Retrieval(CBIR)and image mining are becoming more important study fields in computer vision due to their wide range of applications in healthcare,security,and various domains.The image retrieval sy...Content-Based Image Retrieval(CBIR)and image mining are becoming more important study fields in computer vision due to their wide range of applications in healthcare,security,and various domains.The image retrieval system mainly relies on the efficiency and accuracy of the classification models.This research addresses the challenge of enhancing the image retrieval system by developing a novel approach,EfficientNet-Convolutional Neural Network(EffNet-CNN).The key objective of this research is to evaluate the proposed EffNet-CNN model’s performance in image classification,image mining,and CBIR.The novelty of the proposed EffNet-CNN model includes the integration of different techniques and modifications.The model includes the Mahalanobis distance metric for feature matching,which enhances the similarity measurements.The model extends EfficientNet architecture by incorporating additional convolutional layers,batch normalization,dropout,and pooling layers for improved hierarchical feature extraction.A systematic hyperparameter optimization using SGD,performance evaluation with three datasets,and data normalization for improving feature representations.The EffNet-CNN is assessed utilizing precision,accuracy,F-measure,and recall metrics across MS-COCO,CIFAR-10 and 100 datasets.The model achieved accuracy values ranging from 90.60%to 95.90%for the MS-COCO dataset,96.8%to 98.3%for the CIFAR-10 dataset and 92.9%to 98.6%for the CIFAR-100 dataset.A validation of the EffNet-CNN model’s results with other models reveals the proposed model’s superior performance.The results highlight the potential of the EffNet-CNN model proposed for image classification and its usefulness in image mining and CBIR.展开更多
Research on mass gathering events is critical for ensuring public security and maintaining social order.However,most of the existing works focus on crowd behavior analysis areas such as anomaly detection and crowd cou...Research on mass gathering events is critical for ensuring public security and maintaining social order.However,most of the existing works focus on crowd behavior analysis areas such as anomaly detection and crowd counting,and there is a relative lack of research on mass gathering behaviors.We believe real-time detection and monitoring of mass gathering behaviors are essential formigrating potential security risks and emergencies.Therefore,it is imperative to develop a method capable of accurately identifying and localizing mass gatherings before disasters occur,enabling prompt and effective responses.To address this problem,we propose an innovative Event-Driven Attention Network(EDAN),which achieves image-text matching in the scenario of mass gathering events with good results for the first time.Traditional image-text retrieval methods based on global alignment are difficult to capture the local details within complex scenes,limiting retrieval accuracy.While local alignment-based methods aremore effective at extracting detailed features,they frequently process raw textual features directly,which often contain ambiguities and redundant information that can diminish retrieval efficiency and degrade model performance.To overcome these challenges,EDAN introduces an Event-Driven AttentionModule that adaptively focuses attention on image regions or textual words relevant to the event type.By calculating the semantic distance between event labels and textual content,this module effectively significantly reduces computational complexity and enhances retrieval efficiency.To validate the effectiveness of EDAN,we construct a dedicated multimodal dataset tailored for the analysis of mass gathering events,providing a reliable foundation for subsequent studies.We conduct comparative experiments with other methods on our dataset,the experimental results demonstrate the effectiveness of EDAN.In the image-to-text retrieval task,EDAN achieved the best performance on the R@5 metric,while in the text-to-image retrieval task,it showed superior results on both R@10 and R@5 metrics.Additionally,EDAN excelled in the overall Rsummetric,achieving the best performance.Finally,ablation studies further verified the effectiveness of event-driven attention module.展开更多
The retrieval of the biomass parameters from active/passive microwave remote sensing data （10.2 GHz） is performed based on an iterative inversion of BP neural network model with fuzzy optimization. The BP neural net...The retrieval of the biomass parameters from active/passive microwave remote sensing data （10.2 GHz） is performed based on an iterative inversion of BP neural network model with fuzzy optimization. The BP neural network is trained by a set of the measurements of active and passive remote sensing and the ground truth data versus Day of Year during growth. Once the network training is complete, the model can be used to retrieve the temporal variations of the biomass parameters from another set of observation data. The model was used in weights and microware observation data of wheat growth in 1989 to retrieve biomass parameters change of wheat growth this year. The retrieved biomass parameters correspond well with the real data of the growth, which shows that the BP model is scientific and sound.展开更多
To deal with a lack of semantic interoperability of traditional knowledge retrieval approaches, a semantic-based networked manufacturing （NM） knowledge retrieval architecture is proposed, which offers a series of to...To deal with a lack of semantic interoperability of traditional knowledge retrieval approaches, a semantic-based networked manufacturing （NM） knowledge retrieval architecture is proposed, which offers a series of tools for supporting the sharing of knowledge and promoting NM collaboration. A 5-tuple based semantic information retrieval model is proposed, which includes the interoperation on the semantic layer, and a test process is given for this model. The recall ratio and the precision ratio of manufacturing knowledge retrieval are proved to be greatly improved by evaluation. Thus, a practical and reliable approach based on the semantic web is provided for solving the correlated concrete problems in regional networked manufacturing.展开更多
A neural network methodology is presented to retrieve wind vectors from ERS - 1/2 scatterometer data. The wind directional ambiguities are eliminated by a circular median filter algorithm. All data come from ERS - 1/2...A neural network methodology is presented to retrieve wind vectors from ERS - 1/2 scatterometer data. The wind directional ambiguities are eliminated by a circular median filter algorithm. All data come from ERS - 1/2 scatterometer data collocated pairs with CMCD4 vector. Comparing the results with CMCD4 and ECMWF wind vector,they agree well, which indicates that it is possible to extract wind vector from the ERS-1/2 scatterometer with the neural network method.展开更多
Conventional retrieval and neural network methods are used simultaneously to retrieve sea surface wind speed(SSWS)from HH-polarized Sentinel-1(S1)SAR images.The Polarization Ratio(PR)models combined with the CMOD5.N G...Conventional retrieval and neural network methods are used simultaneously to retrieve sea surface wind speed(SSWS)from HH-polarized Sentinel-1(S1)SAR images.The Polarization Ratio(PR)models combined with the CMOD5.N Geophysical Model Function(GMF)is used for SSWS retrieval from the HH-polarized SAR data.We compared different PR models developed based on previous C-band SAR data in HH-polarization for their applications to the S1 SAR data.The recently proposed CMODH,i.e.,retrieving SSWS directly from the HHpolarized S1 data is also validated.The results indicate that the CMODH model performs better than results achieved using the PR models.We proposed a neural network method based on the backward propagation(BP)neural network to retrieve SSWS from the S1 HH-polarized data.The SSWS retrieved using the BP neural network model agrees better with the buoy measurements and ASCAT dataset than the results achieved using the conventional methods.Compared to the buoy measurements,the bias,root mean square error(RMSE)and scatter index(SI)of wind speed retrieved by the BP neural network model are 0.10 m/s,1.38 m/s and 19.85%,respectively,while compared to the ASCAT dataset the three parameters of training set are–0.01 m/s,1.33 m/s and 15.10%,respectively.It is suggested that the BP neural network model has a potential application in retrieving SSWS from Sentinel-1 images acquired at HH-polarization.展开更多
In content-based image retrieval(CBIR),primitive image signatures are critical because they represent the visual characteristics.Image signatures,which are algorithmically descriptive and accurately recognized visual ...In content-based image retrieval(CBIR),primitive image signatures are critical because they represent the visual characteristics.Image signatures,which are algorithmically descriptive and accurately recognized visual components,are used to appropriately index and retrieve comparable results.To differentiate an image in the category of qualifying contender,feature vectors must have image information's like colour,objects,shape,spatial viewpoints.Previous methods such as sketch-based image retrieval by salient contour(SBIR)and greedy learning of deep Boltzmann machine(GDBM)used spatial information to distinguish between image categories.This requires interest points and also feature analysis emerged image detection problems.Thus,a proposed model to overcome this issue and predict the repeating pattern as well as series of pixels that conclude similarity has been necessary.In this study,a technique called CBIR-similarity measure via artificial neural network interpolation(CBIR-SMANN)has been presented.By collecting datasets,the images are resized then subject to Gaussian filtering in the pre-processing stage,then by permitting them to the Hessian detector,the interesting points are gathered.Based on Skewness,mean,kurtosis and standard deviation features were extracted then given to ANN for interpolation.Interpolated results are stored in a database for retrieval.In the testing stage,the query image was inputted that is subjected to pre-processing,and feature extraction was then fed to the similarity measurement function.Thus,ANN helps to get similar images from the database.CBIR-SMANN have been implemented in the python tool and then evaluated for its performance.Results show that CBIR-SMANN exhibited a high recall value of 78%with a minimum retrieval time of 980 ms.This showed the supremacy of the proposed model was comparatively greater than the previous ones.展开更多
A convolutional neural network is employed to retrieve the time-domain envelop and phase of few-cycle femtosecond pulses from transient-grating frequency-resolved optical gating(TG-FROG) traces.We use theoretically ge...A convolutional neural network is employed to retrieve the time-domain envelop and phase of few-cycle femtosecond pulses from transient-grating frequency-resolved optical gating(TG-FROG) traces.We use theoretically generated TGFROG traces to complete supervised trainings of the convolutional neural networks,then use similarly generated traces not included in the training dataset to test how well the networks are trained.Accurate retrieval of such traces by the neural network is realized.In our case,we find that networks with exponential linear unit(ELU) activation function perform better than those with leaky rectified linear unit(LRELU) and scaled exponential linear unit(SELU).Finally,the issues that need to be addressed for the retrieval of experimental data by this method are discussed.展开更多
Satellite hyperspectral infrared sounder measurements have better horizontal resolution than other sounding techniques as it boasts the stratospheric gravity wave(GW)analysis.To accurately and efficiently derive the t...Satellite hyperspectral infrared sounder measurements have better horizontal resolution than other sounding techniques as it boasts the stratospheric gravity wave(GW)analysis.To accurately and efficiently derive the three-dimensional structure of the stratospheric GWs from the single-field-of-view(SFOV)Atmospheric Infra Red Sounder(AIRS)observations,this paper firstly focuses on the retrieval of the atmospheric temperature profiles in the altitude range of 20-60 km with an artificial neural network approach(ANN).The simulation experiments show that the retrieval bias is less than 0.5 K,and the root mean square error(RMSE)ranges from 1.8 to 4 K.Moreover,the retrieval results from 20 granules of the AIRS observations with the trained neural network(AIRS_SFOV)and the corresponding operational AIRS products(AIRS_L2)as well as the dual-regression results from the Cooperative Institute for Meteorological Satellite Studies(CIMSS)(AIRS_DR)are compared respectively with ECMWF T799 data.The comparison indicates that the standard deviation of the ANN retrieval errors is significantly less than that of the AIRS_DR.Furthermore,the analysis of the typical GW events induced by the mountain Andes and the typhoon"Soulik"using different data indicates that the AIRS_SFOV results capture more details of the stratospheric gravity waves in the perturbation amplitude and pattern than the operational AIRS products do.展开更多
Lung medical image retrieval based on content similarity plays an important role in computer-aided diagnosis of lung cancer.In recent years,binary hashing has become a hot topic in this field due to its compressed sto...Lung medical image retrieval based on content similarity plays an important role in computer-aided diagnosis of lung cancer.In recent years,binary hashing has become a hot topic in this field due to its compressed storage and fast query speed.Traditional hashing methods often rely on highdimensional features based hand-crafted methods,which might not be optimally compatible with lung nodule images.Also,different hashing bits contribute to the image retrieval differently,and therefore treating the hashing bits equally affects the retrieval accuracy.Hence,an image retrieval method of lung nodule images is proposed with the basis on convolutional neural networks and hashing.First,apre-trained and fine-tuned convolutional neural network is employed to learn multilevel semantic features of the lung nodules.Principal components analysis is utilized to remove redundant information and preserve informative semantic features of the lung nodules.Second,the proposed method relies on nine sign labels of lung nodules for the training set,and the semantic feature is combined to construct hashing functions.Finally,returned lung nodule images can be easily ranked with the query-adaptive search method based on weighted Hamming distance.Extensive experiments and evaluations on the dataset demonstrate that the proposed method can significantly improve the expression ability of lung nodule images,which further validates the effectiveness of the proposed method.展开更多
This paper presents a new integrated information retrieval support system （IIRSS） which can help Web search engines retrieve cross-lingual information from hereto geneous resources stored in multi-databases in Intra...This paper presents a new integrated information retrieval support system （IIRSS） which can help Web search engines retrieve cross-lingual information from hereto geneous resources stored in multi-databases in Intranet. The IIRSS, with a three-layer architecture, can cooperate with other application servers running in Intranet. By using intelligent agents to collect information and to create indexes on the-fly, using an access control strategy to confine a user to browsing those accessible documents for him/her through a single portal, and using a new cross-lingual translation tool to help the search engine retrieve documents, the new system provides controllable information access with different authorizations, personalized services, and real-time information retrieval.展开更多
Aiming at shortcomings of traditional image retrieval systems, a new image retrieval approach based on color features of image combining intuitive fuzzy theory with genetic algorithm is proposed. Each image is segment...Aiming at shortcomings of traditional image retrieval systems, a new image retrieval approach based on color features of image combining intuitive fuzzy theory with genetic algorithm is proposed. Each image is segmented into a constant number of sub-images in vertical direction. Color features are extracted from every sub-image to get chromosome coding. It is considered that fuzzy membership and intuitive fuzzy hesitancy degree of every pixel＇s color in image are associated to all the color histogram bins. Certain feature, fuzzy feature and intuitive fuzzy feature of colors in an image, are used together to describe the content of image. Efficient combinations of sub-image are selected according to operation of selecting, crossing and variation. Retrieval results are obtained from image matching based on these color feature combinations of sub-images. Tests show that this approach can improve the accuracy of image retrieval in the case of not decreasing the speed of image retrieval. Its mean precision is above 80 %.展开更多
Currently,the video captioning models based on an encoder-decoder mainly rely on a single video input source.The contents of video captioning are limited since few studies employed external corpus information to guide...Currently,the video captioning models based on an encoder-decoder mainly rely on a single video input source.The contents of video captioning are limited since few studies employed external corpus information to guide the generation of video captioning,which is not conducive to the accurate descrip-tion and understanding of video content.To address this issue,a novel video captioning method guided by a sentence retrieval generation network(ED-SRG)is proposed in this paper.First,a ResNeXt network model,an efficient convolutional network for online video understanding(ECO)model,and a long short-term memory(LSTM)network model are integrated to construct an encoder-decoder,which is utilized to extract the 2D features,3D features,and object features of video data respectively.These features are decoded to generate textual sentences that conform to video content for sentence retrieval.Then,a sentence-transformer network model is employed to retrieve different sentences in an external corpus that are semantically similar to the above textual sentences.The candidate sentences are screened out through similarity measurement.Finally,a novel GPT-2 network model is constructed based on GPT-2 network structure.The model introduces a designed random selector to randomly select predicted words with a high probability in the corpus,which is used to guide and generate textual sentences that are more in line with human natural language expressions.The proposed method in this paper is compared with several existing works by experiments.The results show that the indicators BLEU-4,CIDEr,ROUGE_L,and METEOR are improved by 3.1%,1.3%,0.3%,and 1.5%on a public dataset MSVD and 1.3%,0.5%,0.2%,1.9%on a public dataset MSR-VTT respectively.It can be seen that the proposed method in this paper can generate video captioning with richer semantics than several state-of-the-art approaches.展开更多
To avoid the scalability of the existing systems that employed centralized indexing,index flooding or query flooding,we proposed an efficient peer-to-peer information retrieval system SPIRS (Semantic P2P-based Informa...To avoid the scalability of the existing systems that employed centralized indexing,index flooding or query flooding,we proposed an efficient peer-to-peer information retrieval system SPIRS (Semantic P2P-based Information Retrieval System) that supported state-of-the-art content and semantic searches. SPIRS distributes document indices through P2P network hierarchically by Latent Semantic Indexing (LSI) and organizes nodes into a hierarchical overlay through CAN and TRIE. Comparing with other P2P search techniques,those based on simple keyword matching,SPIRS has better accuracy for considering the advanced relevance among documents. Given a query,only a small number of nodes are needed for SPIRS to identify the matching documents. Furthermore,both theoretical analysis and experimental results show that SPIRS possesses higher accuracy and less logic hops.展开更多
Recent advances in wireless mobile computing, digital library, and distributed multimedia technologies are stimulating the development of mobile multimedia digital library systems （ M2DLS） that allow mobile clients ...Recent advances in wireless mobile computing, digital library, and distributed multimedia technologies are stimulating the development of mobile multimedia digital library systems （ M2DLS） that allow mobile clients to access multimedia material anywhere and anytime over a cellular radio network. This paper addresses the problem of providing synchronized multimedia retrieval in these sys- tems. An efficient inter-media synchronization scheme called wireless prioritized feedback （WPF） for mobile multimedia on-demand retrieval in digital library systems over CDMA cellular radio networks in the absence of synchronized clocks is presented. In the WPF scheme, base stations use preferential lightweight message called feedback units transmitted by mobile mediaphones to detect and correct asynchronies. The algorithm is described with pseudo code, and experiments are described to demonstrate the efficiency of WPF.展开更多
Aiming at the difficulty of accurately constructing the dynamic model of subtropical high, based on the potential height field time series over 500 hPa layer of T106 numerical forecast products, by using EOF（empirica...Aiming at the difficulty of accurately constructing the dynamic model of subtropical high, based on the potential height field time series over 500 hPa layer of T106 numerical forecast products, by using EOF（empirical orthogonal function） temporal-spatial separation technique, the disassembled EOF time coefficients series were regarded as dynamical model variables, and dynamic system retrieval idea as well as genetic algorithm were introduced to make dynamical model parameters optimization search, then, a reasonable non-linear dynamic model of EOF time-coefficients was established. By dynamic model integral and EOF temporal-spatial components assembly, a mid-/long-term forecast of subtropical high was carried out. The experimental results show that the forecast results of dynamic model are superior to that of general numerical model forecast results. A new modeling idea and forecast technique is presented for diagnosing and forecasting such complicated weathers as subtropical high.展开更多
In the era of big data rich inWe Media,the single mode retrieval system has been unable to meet people’s demand for information retrieval.This paper proposes a new solution to the problem of feature extraction and un...In the era of big data rich inWe Media,the single mode retrieval system has been unable to meet people’s demand for information retrieval.This paper proposes a new solution to the problem of feature extraction and unified mapping of different modes:A Cross-Modal Hashing retrieval algorithm based on Deep Residual Network(CMHR-DRN).The model construction is divided into two stages:The first stage is the feature extraction of different modal data,including the use of Deep Residual Network(DRN)to extract the image features,using the method of combining TF-IDF with the full connection network to extract the text features,and the obtained image and text features used as the input of the second stage.In the second stage,the image and text features are mapped into Hash functions by supervised learning,and the image and text features are mapped to the common binary Hamming space.In the process of mapping,the distance measurement of the original distance measurement and the common feature space are kept unchanged as far as possible to improve the accuracy of Cross-Modal Retrieval.In training the model,adaptive moment estimation(Adam)is used to calculate the adaptive learning rate of each parameter,and the stochastic gradient descent(SGD)is calculated to obtain the minimum loss function.The whole training process is completed on Caffe deep learning framework.Experiments show that the proposed algorithm CMHR-DRN based on Deep Residual Network has better retrieval performance and stronger advantages than other Cross-Modal algorithms CMFH,CMDN and CMSSH.展开更多
In this paper, we propose the dynamically-evolving active overlay network （DEAON）, which is an efficient, scalable yet simple protocol to facilitate applications of decentralized information retrieval in P2P network...In this paper, we propose the dynamically-evolving active overlay network （DEAON）, which is an efficient, scalable yet simple protocol to facilitate applications of decentralized information retrieval in P2P networks. DEAON consists of three novel components ： a Desirable Topology Construction and Adaptation algorithm to guide the evolution of the overlay topology towards a small-world-like graph; a Semantic-based Neighbor Selection scheme to conduct an online neighbor ranking; a Topology-aware Intelligent Search mechanism to forward incoming queries to deliberately selected neighbors. We deploy and compare DEAON with other several existing distributed search techniques over static and dynamic environments. The results indicate that DEAON outperforms its competitors by achieving higher recall rate while using much less network resources, in both of the above environments.展开更多
In the present paper, we have studied the effect of soil textures,?i.e., sand, silt and clay on microwave scattering at X-band (10 GHz) at various incidence angles and like polarizations (i.e., Horizontal-Horizontal;H...In the present paper, we have studied the effect of soil textures,?i.e., sand, silt and clay on microwave scattering at X-band (10 GHz) at various incidence angles and like polarizations (i.e., Horizontal-Horizontal;HH-, Vertical-Vertical;VV-). We have proposed a retrieval technique based on Genetic Algorithm (GA) to retrieve soil texture. For this purpose, ten types of soil mixtures having different percentage of sand, silt and clay have been analyzed. The observations were carried out by ingeniously assembled X-band scatterometer. A good agreement has been noticed between estimated and observed soil texture. Study infers that soil texture is quite sensitive to radar scattering and it is possible to retrieve soil texture with radar/scatterometer data with good accuracy and this type of retrieved results can be helpful to predict soil strength as well as soil erosion of the particular area.展开更多
基金funded by the Institute of InformationTechnology,VietnamAcademy of Science and Technology(project number CSCL02.02/22-23)“Research and Development of Methods for Searching Similar Trademark Images Using Machine Learning to Support Trademark Examination in Vietnam”.
文摘Image-based similar trademark retrieval is a time-consuming and labor-intensive task in the trademark examination process.This paper aims to support trademark examiners by training Deep Convolutional Neural Network(DCNN)models for effective Trademark Image Retrieval(TIR).To achieve this goal,we first develop a novel labeling method that automatically generates hundreds of thousands of labeled similar and dissimilar trademark image pairs using accompanying data fields such as citation lists,Vienna classification(VC)codes,and trademark ownership information.This approach eliminates the need for manual labeling and provides a large-scale dataset suitable for training deep learning models.We then train DCNN models based on Siamese and Triplet architectures,evaluating various feature extractors to determine the most effective configuration.Furthermore,we present an Adapted Contrastive Loss Function(ACLF)for the trademark retrieval task,specifically engineered to mitigate the influence of noisy labels found in automatically created datasets.Experimental results indicate that our proposed model(Efficient-Net_v21_Siamese)performs best at both True Negative Rate(TNR)threshold levels,TNR 0.9 and TNR 0.95,with==respective True Positive Rates(TPRs)of 77.7%and 70.8%and accuracies of 83.9%and 80.4%.Additionally,when testing on the public trademark dataset METU_v2,our model achieves a normalized average rank(NAR)of 0.0169,outperforming the current state-of-the-art(SOTA)model.Based on these findings,we estimate that considering only approximately 10%of the returned trademarks would be sufficient,significantly reducing the review time.Therefore,the paper highlights the potential of utilizing national trademark data to enhance the accuracy and efficiency of trademark retrieval systems,ultimately supporting trademark examiners in their evaluation tasks.
基金The authors extend their appreciation to the Deanship of Research and Graduate Studies at King Khalid University,Kingdom of Saudi Arabia,for funding this work through the Small Research Group Project under Grant Number RGP.1/316/45.
文摘Content-Based Image Retrieval(CBIR)and image mining are becoming more important study fields in computer vision due to their wide range of applications in healthcare,security,and various domains.The image retrieval system mainly relies on the efficiency and accuracy of the classification models.This research addresses the challenge of enhancing the image retrieval system by developing a novel approach,EfficientNet-Convolutional Neural Network(EffNet-CNN).The key objective of this research is to evaluate the proposed EffNet-CNN model’s performance in image classification,image mining,and CBIR.The novelty of the proposed EffNet-CNN model includes the integration of different techniques and modifications.The model includes the Mahalanobis distance metric for feature matching,which enhances the similarity measurements.The model extends EfficientNet architecture by incorporating additional convolutional layers,batch normalization,dropout,and pooling layers for improved hierarchical feature extraction.A systematic hyperparameter optimization using SGD,performance evaluation with three datasets,and data normalization for improving feature representations.The EffNet-CNN is assessed utilizing precision,accuracy,F-measure,and recall metrics across MS-COCO,CIFAR-10 and 100 datasets.The model achieved accuracy values ranging from 90.60%to 95.90%for the MS-COCO dataset,96.8%to 98.3%for the CIFAR-10 dataset and 92.9%to 98.6%for the CIFAR-100 dataset.A validation of the EffNet-CNN model’s results with other models reveals the proposed model’s superior performance.The results highlight the potential of the EffNet-CNN model proposed for image classification and its usefulness in image mining and CBIR.
基金sponsored by Natural Science Foundation of Xinjiang Uygur Autonomous Region(2024D01A19).
文摘Research on mass gathering events is critical for ensuring public security and maintaining social order.However,most of the existing works focus on crowd behavior analysis areas such as anomaly detection and crowd counting,and there is a relative lack of research on mass gathering behaviors.We believe real-time detection and monitoring of mass gathering behaviors are essential formigrating potential security risks and emergencies.Therefore,it is imperative to develop a method capable of accurately identifying and localizing mass gatherings before disasters occur,enabling prompt and effective responses.To address this problem,we propose an innovative Event-Driven Attention Network(EDAN),which achieves image-text matching in the scenario of mass gathering events with good results for the first time.Traditional image-text retrieval methods based on global alignment are difficult to capture the local details within complex scenes,limiting retrieval accuracy.While local alignment-based methods aremore effective at extracting detailed features,they frequently process raw textual features directly,which often contain ambiguities and redundant information that can diminish retrieval efficiency and degrade model performance.To overcome these challenges,EDAN introduces an Event-Driven AttentionModule that adaptively focuses attention on image regions or textual words relevant to the event type.By calculating the semantic distance between event labels and textual content,this module effectively significantly reduces computational complexity and enhances retrieval efficiency.To validate the effectiveness of EDAN,we construct a dedicated multimodal dataset tailored for the analysis of mass gathering events,providing a reliable foundation for subsequent studies.We conduct comparative experiments with other methods on our dataset,the experimental results demonstrate the effectiveness of EDAN.In the image-to-text retrieval task,EDAN achieved the best performance on the R@5 metric,while in the text-to-image retrieval task,it showed superior results on both R@10 and R@5 metrics.Additionally,EDAN excelled in the overall Rsummetric,achieving the best performance.Finally,ablation studies further verified the effectiveness of event-driven attention module.
文摘The retrieval of the biomass parameters from active/passive microwave remote sensing data （10.2 GHz） is performed based on an iterative inversion of BP neural network model with fuzzy optimization. The BP neural network is trained by a set of the measurements of active and passive remote sensing and the ground truth data versus Day of Year during growth. Once the network training is complete, the model can be used to retrieve the temporal variations of the biomass parameters from another set of observation data. The model was used in weights and microware observation data of wheat growth in 1989 to retrieve biomass parameters change of wheat growth this year. The retrieved biomass parameters correspond well with the real data of the growth, which shows that the BP model is scientific and sound.
基金The National High Technology Research and Devel-opment Program of China (863Program) (No2003AA1Z2560,2002AA414060)the Key Science and Technology Program of Shaanxi Province (No2006K04-G10)
文摘To deal with a lack of semantic interoperability of traditional knowledge retrieval approaches, a semantic-based networked manufacturing （NM） knowledge retrieval architecture is proposed, which offers a series of tools for supporting the sharing of knowledge and promoting NM collaboration. A 5-tuple based semantic information retrieval model is proposed, which includes the interoperation on the semantic layer, and a test process is given for this model. The recall ratio and the precision ratio of manufacturing knowledge retrieval are proved to be greatly improved by evaluation. Thus, a practical and reliable approach based on the semantic web is provided for solving the correlated concrete problems in regional networked manufacturing.
文摘A neural network methodology is presented to retrieve wind vectors from ERS - 1/2 scatterometer data. The wind directional ambiguities are eliminated by a circular median filter algorithm. All data come from ERS - 1/2 scatterometer data collocated pairs with CMCD4 vector. Comparing the results with CMCD4 and ECMWF wind vector,they agree well, which indicates that it is possible to extract wind vector from the ERS-1/2 scatterometer with the neural network method.
基金The National Key Research and Development Program under contract Nos 2016YFC1402703 and 2018YFC1407100
文摘Conventional retrieval and neural network methods are used simultaneously to retrieve sea surface wind speed(SSWS)from HH-polarized Sentinel-1(S1)SAR images.The Polarization Ratio(PR)models combined with the CMOD5.N Geophysical Model Function(GMF)is used for SSWS retrieval from the HH-polarized SAR data.We compared different PR models developed based on previous C-band SAR data in HH-polarization for their applications to the S1 SAR data.The recently proposed CMODH,i.e.,retrieving SSWS directly from the HHpolarized S1 data is also validated.The results indicate that the CMODH model performs better than results achieved using the PR models.We proposed a neural network method based on the backward propagation(BP)neural network to retrieve SSWS from the S1 HH-polarized data.The SSWS retrieved using the BP neural network model agrees better with the buoy measurements and ASCAT dataset than the results achieved using the conventional methods.Compared to the buoy measurements,the bias,root mean square error(RMSE)and scatter index(SI)of wind speed retrieved by the BP neural network model are 0.10 m/s,1.38 m/s and 19.85%,respectively,while compared to the ASCAT dataset the three parameters of training set are–0.01 m/s,1.33 m/s and 15.10%,respectively.It is suggested that the BP neural network model has a potential application in retrieving SSWS from Sentinel-1 images acquired at HH-polarization.
文摘In content-based image retrieval(CBIR),primitive image signatures are critical because they represent the visual characteristics.Image signatures,which are algorithmically descriptive and accurately recognized visual components,are used to appropriately index and retrieve comparable results.To differentiate an image in the category of qualifying contender,feature vectors must have image information's like colour,objects,shape,spatial viewpoints.Previous methods such as sketch-based image retrieval by salient contour(SBIR)and greedy learning of deep Boltzmann machine(GDBM)used spatial information to distinguish between image categories.This requires interest points and also feature analysis emerged image detection problems.Thus,a proposed model to overcome this issue and predict the repeating pattern as well as series of pixels that conclude similarity has been necessary.In this study,a technique called CBIR-similarity measure via artificial neural network interpolation(CBIR-SMANN)has been presented.By collecting datasets,the images are resized then subject to Gaussian filtering in the pre-processing stage,then by permitting them to the Hessian detector,the interesting points are gathered.Based on Skewness,mean,kurtosis and standard deviation features were extracted then given to ANN for interpolation.Interpolated results are stored in a database for retrieval.In the testing stage,the query image was inputted that is subjected to pre-processing,and feature extraction was then fed to the similarity measurement function.Thus,ANN helps to get similar images from the database.CBIR-SMANN have been implemented in the python tool and then evaluated for its performance.Results show that CBIR-SMANN exhibited a high recall value of 78%with a minimum retrieval time of 980 ms.This showed the supremacy of the proposed model was comparatively greater than the previous ones.
基金Project supported by the National Key R&D Program of China(Grant No.2017YFB0405202)the National Natural Science Foundation of China(Grant Nos.61690221,91850209,and 11774277)。
文摘A convolutional neural network is employed to retrieve the time-domain envelop and phase of few-cycle femtosecond pulses from transient-grating frequency-resolved optical gating(TG-FROG) traces.We use theoretically generated TGFROG traces to complete supervised trainings of the convolutional neural networks,then use similarly generated traces not included in the training dataset to test how well the networks are trained.Accurate retrieval of such traces by the neural network is realized.In our case,we find that networks with exponential linear unit(ELU) activation function perform better than those with leaky rectified linear unit(LRELU) and scaled exponential linear unit(SELU).Finally,the issues that need to be addressed for the retrieval of experimental data by this method are discussed.
基金National Natural Science Foundation of China(41575031,41375024)Postdoctoral Science Foundation of China(2015M580124)Meteorology Research Special Funds for Public Welfare Projects(GYHY201406011)。
文摘Satellite hyperspectral infrared sounder measurements have better horizontal resolution than other sounding techniques as it boasts the stratospheric gravity wave(GW)analysis.To accurately and efficiently derive the three-dimensional structure of the stratospheric GWs from the single-field-of-view(SFOV)Atmospheric Infra Red Sounder(AIRS)observations,this paper firstly focuses on the retrieval of the atmospheric temperature profiles in the altitude range of 20-60 km with an artificial neural network approach(ANN).The simulation experiments show that the retrieval bias is less than 0.5 K,and the root mean square error(RMSE)ranges from 1.8 to 4 K.Moreover,the retrieval results from 20 granules of the AIRS observations with the trained neural network(AIRS_SFOV)and the corresponding operational AIRS products(AIRS_L2)as well as the dual-regression results from the Cooperative Institute for Meteorological Satellite Studies(CIMSS)(AIRS_DR)are compared respectively with ECMWF T799 data.The comparison indicates that the standard deviation of the ANN retrieval errors is significantly less than that of the AIRS_DR.Furthermore,the analysis of the typical GW events induced by the mountain Andes and the typhoon"Soulik"using different data indicates that the AIRS_SFOV results capture more details of the stratospheric gravity waves in the perturbation amplitude and pattern than the operational AIRS products do.
基金Supported by the National Natural Science Foundation of China(61373100)the Open Funding Project of State Key Laboratory of Virtual Reality Technology and Systems(BUAA-VR-16KF-13,BUAA-VR-17KF-14,BUAA-VR-17KF-15)the Research Project Supported by Shanxi Scholarship Council of China(2016-038)
文摘Lung medical image retrieval based on content similarity plays an important role in computer-aided diagnosis of lung cancer.In recent years,binary hashing has become a hot topic in this field due to its compressed storage and fast query speed.Traditional hashing methods often rely on highdimensional features based hand-crafted methods,which might not be optimally compatible with lung nodule images.Also,different hashing bits contribute to the image retrieval differently,and therefore treating the hashing bits equally affects the retrieval accuracy.Hence,an image retrieval method of lung nodule images is proposed with the basis on convolutional neural networks and hashing.First,apre-trained and fine-tuned convolutional neural network is employed to learn multilevel semantic features of the lung nodules.Principal components analysis is utilized to remove redundant information and preserve informative semantic features of the lung nodules.Second,the proposed method relies on nine sign labels of lung nodules for the training set,and the semantic feature is combined to construct hashing functions.Finally,returned lung nodule images can be easily ranked with the query-adaptive search method based on weighted Hamming distance.Extensive experiments and evaluations on the dataset demonstrate that the proposed method can significantly improve the expression ability of lung nodule images,which further validates the effectiveness of the proposed method.
基金Supported by the National Natural Science Foun-dation of China (60173010)
文摘This paper presents a new integrated information retrieval support system （IIRSS） which can help Web search engines retrieve cross-lingual information from hereto geneous resources stored in multi-databases in Intranet. The IIRSS, with a three-layer architecture, can cooperate with other application servers running in Intranet. By using intelligent agents to collect information and to create indexes on the-fly, using an access control strategy to confine a user to browsing those accessible documents for him/her through a single portal, and using a new cross-lingual translation tool to help the search engine retrieve documents, the new system provides controllable information access with different authorizations, personalized services, and real-time information retrieval.
基金Sponsored by the Ministerial Level Foundation(20061823)
文摘Aiming at shortcomings of traditional image retrieval systems, a new image retrieval approach based on color features of image combining intuitive fuzzy theory with genetic algorithm is proposed. Each image is segmented into a constant number of sub-images in vertical direction. Color features are extracted from every sub-image to get chromosome coding. It is considered that fuzzy membership and intuitive fuzzy hesitancy degree of every pixel＇s color in image are associated to all the color histogram bins. Certain feature, fuzzy feature and intuitive fuzzy feature of colors in an image, are used together to describe the content of image. Efficient combinations of sub-image are selected according to operation of selecting, crossing and variation. Retrieval results are obtained from image matching based on these color feature combinations of sub-images. Tests show that this approach can improve the accuracy of image retrieval in the case of not decreasing the speed of image retrieval. Its mean precision is above 80 %.
基金supported in part by the National Natural Science Foundation of China under Grants 62273272 and 61873277in part by the Chinese Postdoctoral Science Foundation under Grant 2020M673446+1 种基金in part by the Key Research and Development Program of Shaanxi Province under Grant 2023-YBGY-243in part by the Youth Innovation Team of Shaanxi Universities.
文摘Currently,the video captioning models based on an encoder-decoder mainly rely on a single video input source.The contents of video captioning are limited since few studies employed external corpus information to guide the generation of video captioning,which is not conducive to the accurate descrip-tion and understanding of video content.To address this issue,a novel video captioning method guided by a sentence retrieval generation network(ED-SRG)is proposed in this paper.First,a ResNeXt network model,an efficient convolutional network for online video understanding(ECO)model,and a long short-term memory(LSTM)network model are integrated to construct an encoder-decoder,which is utilized to extract the 2D features,3D features,and object features of video data respectively.These features are decoded to generate textual sentences that conform to video content for sentence retrieval.Then,a sentence-transformer network model is employed to retrieve different sentences in an external corpus that are semantically similar to the above textual sentences.The candidate sentences are screened out through similarity measurement.Finally,a novel GPT-2 network model is constructed based on GPT-2 network structure.The model introduces a designed random selector to randomly select predicted words with a high probability in the corpus,which is used to guide and generate textual sentences that are more in line with human natural language expressions.The proposed method in this paper is compared with several existing works by experiments.The results show that the indicators BLEU-4,CIDEr,ROUGE_L,and METEOR are improved by 3.1%,1.3%,0.3%,and 1.5%on a public dataset MSVD and 1.3%,0.5%,0.2%,1.9%on a public dataset MSR-VTT respectively.It can be seen that the proposed method in this paper can generate video captioning with richer semantics than several state-of-the-art approaches.
基金the Nartional Basic Research Programof China(Grant No.2002CB312002)the Science and Technology Commission of Shanghai Munic-ipality Project(Grant No.03dz15027 and 03dz15028).
文摘To avoid the scalability of the existing systems that employed centralized indexing,index flooding or query flooding,we proposed an efficient peer-to-peer information retrieval system SPIRS (Semantic P2P-based Information Retrieval System) that supported state-of-the-art content and semantic searches. SPIRS distributes document indices through P2P network hierarchically by Latent Semantic Indexing (LSI) and organizes nodes into a hierarchical overlay through CAN and TRIE. Comparing with other P2P search techniques,those based on simple keyword matching,SPIRS has better accuracy for considering the advanced relevance among documents. Given a query,only a small number of nodes are needed for SPIRS to identify the matching documents. Furthermore,both theoretical analysis and experimental results show that SPIRS possesses higher accuracy and less logic hops.
基金Project supported by National Natural Science Foundation of China (Grant No .60221120146) , National Basic Research Pro-gram of China ( Grant No . 973 -G1999032704) , and National Grand Fundamental Post Doctor Research of China (Grant No .2003034146)
文摘Recent advances in wireless mobile computing, digital library, and distributed multimedia technologies are stimulating the development of mobile multimedia digital library systems （ M2DLS） that allow mobile clients to access multimedia material anywhere and anytime over a cellular radio network. This paper addresses the problem of providing synchronized multimedia retrieval in these sys- tems. An efficient inter-media synchronization scheme called wireless prioritized feedback （WPF） for mobile multimedia on-demand retrieval in digital library systems over CDMA cellular radio networks in the absence of synchronized clocks is presented. In the WPF scheme, base stations use preferential lightweight message called feedback units transmitted by mobile mediaphones to detect and correct asynchronies. The algorithm is described with pseudo code, and experiments are described to demonstrate the efficiency of WPF.
基金Project supported by the National Natural Science Foundation of China (No.40375019) the Tropical Marine and Meteorology Science Foundation (No.200609) the Jiangsu Key Laboratory of Meteorological Disaster Foundation (No.KLME0507)
文摘Aiming at the difficulty of accurately constructing the dynamic model of subtropical high, based on the potential height field time series over 500 hPa layer of T106 numerical forecast products, by using EOF（empirical orthogonal function） temporal-spatial separation technique, the disassembled EOF time coefficients series were regarded as dynamical model variables, and dynamic system retrieval idea as well as genetic algorithm were introduced to make dynamical model parameters optimization search, then, a reasonable non-linear dynamic model of EOF time-coefficients was established. By dynamic model integral and EOF temporal-spatial components assembly, a mid-/long-term forecast of subtropical high was carried out. The experimental results show that the forecast results of dynamic model are superior to that of general numerical model forecast results. A new modeling idea and forecast technique is presented for diagnosing and forecasting such complicated weathers as subtropical high.
文摘In the era of big data rich inWe Media,the single mode retrieval system has been unable to meet people’s demand for information retrieval.This paper proposes a new solution to the problem of feature extraction and unified mapping of different modes:A Cross-Modal Hashing retrieval algorithm based on Deep Residual Network(CMHR-DRN).The model construction is divided into two stages:The first stage is the feature extraction of different modal data,including the use of Deep Residual Network(DRN)to extract the image features,using the method of combining TF-IDF with the full connection network to extract the text features,and the obtained image and text features used as the input of the second stage.In the second stage,the image and text features are mapped into Hash functions by supervised learning,and the image and text features are mapped to the common binary Hamming space.In the process of mapping,the distance measurement of the original distance measurement and the common feature space are kept unchanged as far as possible to improve the accuracy of Cross-Modal Retrieval.In training the model,adaptive moment estimation(Adam)is used to calculate the adaptive learning rate of each parameter,and the stochastic gradient descent(SGD)is calculated to obtain the minimum loss function.The whole training process is completed on Caffe deep learning framework.Experiments show that the proposed algorithm CMHR-DRN based on Deep Residual Network has better retrieval performance and stronger advantages than other Cross-Modal algorithms CMFH,CMDN and CMSSH.
文摘In this paper, we propose the dynamically-evolving active overlay network （DEAON）, which is an efficient, scalable yet simple protocol to facilitate applications of decentralized information retrieval in P2P networks. DEAON consists of three novel components ： a Desirable Topology Construction and Adaptation algorithm to guide the evolution of the overlay topology towards a small-world-like graph; a Semantic-based Neighbor Selection scheme to conduct an online neighbor ranking; a Topology-aware Intelligent Search mechanism to forward incoming queries to deliberately selected neighbors. We deploy and compare DEAON with other several existing distributed search techniques over static and dynamic environments. The results indicate that DEAON outperforms its competitors by achieving higher recall rate while using much less network resources, in both of the above environments.
文摘In the present paper, we have studied the effect of soil textures,?i.e., sand, silt and clay on microwave scattering at X-band (10 GHz) at various incidence angles and like polarizations (i.e., Horizontal-Horizontal;HH-, Vertical-Vertical;VV-). We have proposed a retrieval technique based on Genetic Algorithm (GA) to retrieve soil texture. For this purpose, ten types of soil mixtures having different percentage of sand, silt and clay have been analyzed. The observations were carried out by ingeniously assembled X-band scatterometer. A good agreement has been noticed between estimated and observed soil texture. Study infers that soil texture is quite sensitive to radar scattering and it is possible to retrieve soil texture with radar/scatterometer data with good accuracy and this type of retrieved results can be helpful to predict soil strength as well as soil erosion of the particular area.
