With the rapid development of the mobile communication and the Internet,the previous web anomaly detectionand identificationmodels were built relying on security experts’empirical knowledge and attack features.Althou...With the rapid development of the mobile communication and the Internet,the previous web anomaly detectionand identificationmodels were built relying on security experts’empirical knowledge and attack features.Althoughthis approach can achieve higher detection performance,it requires huge human labor and resources to maintainthe feature library.In contrast,semantic feature engineering can dynamically discover new semantic featuresand optimize feature selection by automatically analyzing the semantic information contained in the data itself,thus reducing dependence on prior knowledge.However,current semantic features still have the problem ofsemantic expression singularity,as they are extracted from a single semantic mode such as word segmentation,character segmentation,or arbitrary semantic feature extraction.This paper extracts features of web requestsfrom dual semantic granularity,and proposes a semantic feature fusion method to solve the above problems.Themethod first preprocesses web requests,and extracts word-level and character-level semantic features of URLs viaconvolutional neural network(CNN),respectively.By constructing three loss functions to reduce losses betweenfeatures,labels and categories.Experiments on the HTTP CSIC 2010,Malicious URLs and HttpParams datasetsverify the proposedmethod.Results show that compared withmachine learning,deep learningmethods and BERTmodel,the proposed method has better detection performance.And it achieved the best detection rate of 99.16%in the dataset HttpParams.展开更多
Automated optical inspection(AOI)is a significant process in printed circuit board assembly(PCBA)production lines which aims to detect tiny defects in PCBAs.Existing AOI equipment has several deficiencies including lo...Automated optical inspection(AOI)is a significant process in printed circuit board assembly(PCBA)production lines which aims to detect tiny defects in PCBAs.Existing AOI equipment has several deficiencies including low throughput,large computation cost,high latency,and poor flexibility,which limits the efficiency of online PCBA inspection.In this paper,a novel PCBA defect detection method based on a lightweight deep convolution neural network is proposed.In this method,the semantic segmentation model is combined with a rule-based defect recognition algorithm to build up a defect detection frame-work.To improve the performance of the model,extensive real PCBA images are collected from production lines as datasets.Some optimization methods have been applied in the model according to production demand and enable integration in lightweight computing devices.Experiment results show that the production line using our method realizes a throughput more than three times higher than traditional methods.Our method can be integrated into a lightweight inference system and pro-mote the flexibility of AOI.The proposed method builds up a general paradigm and excellent example for model design and optimization oriented towards industrial requirements.展开更多
We propose a novel image segmentation algorithm to tackle the challenge of limited recognition and segmentation performance in identifying welding seam images during robotic intelligent operations.Initially,to enhance...We propose a novel image segmentation algorithm to tackle the challenge of limited recognition and segmentation performance in identifying welding seam images during robotic intelligent operations.Initially,to enhance the capability of deep neural networks in extracting geometric attributes from depth images,we developed a novel deep geometric convolution operator(DGConv).DGConv is utilized to construct a deep local geometric feature extraction module,facilitating a more comprehensive exploration of the intrinsic geometric information within depth images.Secondly,we integrate the newly proposed deep geometric feature module with the Fully Convolutional Network(FCN8)to establish a high-performance deep neural network algorithm tailored for depth image segmentation.Concurrently,we enhance the FCN8 detection head by separating the segmentation and classification processes.This enhancement significantly boosts the network’s overall detection capability.Thirdly,for a comprehensive assessment of our proposed algorithm and its applicability in real-world industrial settings,we curated a line-scan image dataset featuring weld seams.This dataset,named the Standardized Linear Depth Profile(SLDP)dataset,was collected from actual industrial sites where autonomous robots are in operation.Ultimately,we conducted experiments utilizing the SLDP dataset,achieving an average accuracy of 92.7%.Our proposed approach exhibited a remarkable performance improvement over the prior method on the identical dataset.Moreover,we have successfully deployed the proposed algorithm in genuine industrial environments,fulfilling the prerequisites of unmanned robot operations.展开更多
With the continuous development and utilization of marine resources,the underwater target detection has gradually become a popular research topic in the field of underwater robot operations and target detection.Howeve...With the continuous development and utilization of marine resources,the underwater target detection has gradually become a popular research topic in the field of underwater robot operations and target detection.However,it is difficult to combine the environmental semantic information and the semantic information of targets at different scales by detection algorithms due to the complex underwater environment.In this paper,a cascade model based on the UGC-YOLO network structure with high detection accuracy is proposed.The YOLOv3 convolutional neural network is employed as the baseline structure.By fusing the global semantic information between two residual stages in the parallel structure of the feature extraction network,the perception of underwater targets is improved and the detection rate of hard-to-detect underwater objects is raised.Furthermore,the deformable convolution is applied to capture longrange semantic dependencies and PPM pooling is introduced in the highest layer network for aggregating semantic information.Finally,a multi-scale weighted fusion approach is presented for learning semantic information at different scales.Experiments are conducted on an underwater test dataset and the results have demonstrated that our proposed algorithm could detect aquatic targets in complex degraded underwater images.Compared with the baseline network algorithm,the Common Objects in Context(COCO)evaluation metric has been improved by 4.34%.展开更多
Log anomaly detection is an important paradigm for system troubleshooting.Existing log anomaly detection based on Long Short-Term Memory(LSTM)networks is time-consuming to handle long sequences.Transformer model is in...Log anomaly detection is an important paradigm for system troubleshooting.Existing log anomaly detection based on Long Short-Term Memory(LSTM)networks is time-consuming to handle long sequences.Transformer model is introduced to promote efficiency.However,most existing Transformer-based log anomaly detection methods convert unstructured log messages into structured templates by log parsing,which introduces parsing errors.They only extract simple semantic feature,which ignores other features,and are generally supervised,relying on the amount of labeled data.To overcome the limitations of existing methods,this paper proposes a novel unsupervised log anomaly detection method based on multi-feature(UMFLog).UMFLog includes two sub-models to consider two kinds of features:semantic feature and statistical feature,respectively.UMFLog applies the log original content with detailed parameters instead of templates or template IDs to avoid log parsing errors.In the first sub-model,UMFLog uses Bidirectional Encoder Representations from Transformers(BERT)instead of random initialization to extract effective semantic feature,and an unsupervised hypersphere-based Transformer model to learn compact log sequence representations and obtain anomaly candidates.In the second sub-model,UMFLog exploits a statistical feature-based Variational Autoencoder(VAE)about word occurrence times to identify the final anomaly from anomaly candidates.Extensive experiments and evaluations are conducted on three real public log datasets.The results show that UMFLog significantly improves F1-scores compared to the state-of-the-art(SOTA)methods because of the multi-feature.展开更多
Visual simultaneous localization and mapping(SLAM)is crucial in robotics and autonomous driving.However,traditional visual SLAM faces challenges in dynamic environments.To address this issue,researchers have proposed ...Visual simultaneous localization and mapping(SLAM)is crucial in robotics and autonomous driving.However,traditional visual SLAM faces challenges in dynamic environments.To address this issue,researchers have proposed semantic SLAM,which combines object detection,semantic segmentation,instance segmentation,and visual SLAM.Despite the growing body of literature on semantic SLAM,there is currently a lack of comprehensive research on the integration of object detection and visual SLAM.Therefore,this study aims to gather information from multiple databases and review relevant literature using specific keywords.It focuses on visual SLAM based on object detection,covering different aspects.Firstly,it discusses the current research status and challenges in this field,highlighting methods for incorporating semantic information from object detection networks into mileage measurement,closed-loop detection,and map construction.It also compares the characteristics and performance of various visual SLAM object detection algorithms.Lastly,it provides an outlook on future research directions and emerging trends in visual SLAM.Research has shown that visual SLAM based on object detection has significant improvements compared to traditional SLAM in dynamic point removal,data association,point cloud segmentation,and other technologies.It can improve the robustness and accuracy of the entire SLAM system and can run in real time.With the continuous optimization of algorithms and the improvement of hardware level,object visual SLAM has great potential for development.展开更多
This paper proposes a novel method of lane detection,which adopts VGG16 as the basis of convolutional neural network to extract lane line features by cavity convolution,wherein the lane lines are divided into dotted l...This paper proposes a novel method of lane detection,which adopts VGG16 as the basis of convolutional neural network to extract lane line features by cavity convolution,wherein the lane lines are divided into dotted lines and solid lines.Expanding the field of experience through hollow convolution,the full connection layer of the network is discarded,the last largest pooling layer of the VGG16 network is removed,and the processing of the last three convolution layers is replaced by hole convolution.At the same time,CNN adopts the encoder and decoder structure mode,and uses the index function of the maximum pooling layer in the decoder part to upsample the encoder in a counter-pooling manner,realizing semantic segmentation.And combined with the instance segmentation,and finally through the fitting to achieve the detection of the lane line.In addition,the currently disclosed lane line data sets are relatively small,and there is no distinction between lane solid lines and dashed lines.To this end,our work made a lane line data set for the lane virtual and real identification,and based on the proposed algorithm effective verification of the data set achieved by the increased segmentation.The final test shows that the proposed method has a good balance between lane detection speed and accuracy,which has good robustness.展开更多
Power Shell has been widely deployed in fileless malware and advanced persistent threat(APT)attacks due to its high stealthiness and live-off-theland technique.However,existing works mainly focus on deobfuscation and ...Power Shell has been widely deployed in fileless malware and advanced persistent threat(APT)attacks due to its high stealthiness and live-off-theland technique.However,existing works mainly focus on deobfuscation and malicious detection,lacking the malicious Power Shell families classification and behavior analysis.Moreover,the state-of-the-art methods fail to capture fine-grained features and semantic relationships,resulting in low robustness and accuracy.To this end,we propose Power Detector,a novel malicious Power Shell script detector based on multimodal semantic fusion and deep learning.Specifically,we design four feature extraction methods to extract key features from character,token,abstract syntax tree(AST),and semantic knowledge graph.Then,we intelligently design four embeddings(i.e.,Char2Vec,Token2Vec,AST2Vec,and Rela2Vec) and construct a multi-modal fusion algorithm to concatenate feature vectors from different views.Finally,we propose a combined model based on transformer and CNN-Bi LSTM to implement Power Shell family detection.Our experiments with five types of Power Shell attacks show that PowerDetector can accurately detect various obfuscated and stealth PowerShell scripts,with a 0.9402 precision,a 0.9358 recall,and a 0.9374 F1-score.Furthermore,through singlemodal and multi-modal comparison experiments,we demonstrate that PowerDetector’s multi-modal embedding and deep learning model can achieve better accuracy and even identify more unknown attacks.展开更多
In recent years,computer visionfinds wide applications in maritime surveillance with its sophisticated algorithms and advanced architecture.Auto-matic ship detection with computer vision techniques provide an efficien...In recent years,computer visionfinds wide applications in maritime surveillance with its sophisticated algorithms and advanced architecture.Auto-matic ship detection with computer vision techniques provide an efficient means to monitor as well as track ships in water bodies.Waterways being an important medium of transport require continuous monitoring for protection of national security.The remote sensing satellite images of ships in harbours and water bodies are the image data that aid the neural network models to localize ships and to facilitate early identification of possible threats at sea.This paper proposes a deep learning based model capable enough to classify between ships and no-ships as well as to localize ships in the original images using bounding box tech-nique.Furthermore,classified ships are again segmented with deep learning based auto-encoder model.The proposed model,in terms of classification,provides suc-cessful results generating 99.5%and 99.2%validation and training accuracy respectively.The auto-encoder model also produces 85.1%and 84.2%validation and training accuracies.Moreover the IoU metric of the segmented images is found to be of 0.77 value.The experimental results reveal that the model is accu-rate and can be implemented for automatic ship detection in water bodies consid-ering remote sensing satellite images as input to the computer vision system.展开更多
Data fusion is usually an important process in multi-sensor remotely sensed imagery integration environments with the aim of enriching features lacking in the sensors involved in the fusion process. This technique has...Data fusion is usually an important process in multi-sensor remotely sensed imagery integration environments with the aim of enriching features lacking in the sensors involved in the fusion process. This technique has attracted much interest in many researches especially in the field of agriculture. On the other hand, deep learning (DL) based semantic segmentation shows high performance in remote sensing classification, and it requires large datasets in a supervised learning way. In the paper, a method of fusing multi-source remote sensing images with convolution neural networks (CNN) for semantic segmentation is proposed and applied to identify crops. Venezuelan Remote Sensing Satellite-2 (VRSS-2) and the high-resolution of Google Earth (GE) imageries have been used and more than 1000 sample sets have been collected for supervised learning process. The experiment results show that the crops extraction with an average overall accuracy more than 93% has been obtained, which demonstrates that data fusion combined with DL is highly feasible to crops extraction from satellite images and GE imagery, and it shows that deep learning techniques can serve as an invaluable tools for larger remote sensing data fusion frameworks, specifically for the applications in precision farming.展开更多
Current Chinese event detection methods commonly use word embedding to capture semantic representation,but these methods find it difficult to capture the dependence relationship between the trigger words and other wor...Current Chinese event detection methods commonly use word embedding to capture semantic representation,but these methods find it difficult to capture the dependence relationship between the trigger words and other words in the same sentence.Based on the simple evaluation,it is known that a dependency parser can effectively capture dependency relationships and improve the accuracy of event categorisation.This study proposes a novel architecture that models a hybrid representation to summarise semantic and structural information from both characters and words.This model can capture rich semantic features for the event detection task by incorporating the semantic representation generated from the dependency parser.The authors evaluate different models on kbp 2017 corpus.The experimental results show that the proposed method can significantly improve performance in Chinese event detection.展开更多
A semantic unit based event detection scheme in soccer videos is proposed in this paper.The scheme can be characterized as a three-layer framework. At the lowest layer, low-level featuresincluding color, texture, edge...A semantic unit based event detection scheme in soccer videos is proposed in this paper.The scheme can be characterized as a three-layer framework. At the lowest layer, low-level featuresincluding color, texture, edge, shape, and motion are extracted. High-level semantic events aredefined at the highest layer. In order to connect low-level features and high-level semantics, wedesign and define some semantic units at the intermediate layer. A semantic unit is composed of asequence of consecutives frames with the same cue that is deduced from low-level features. Based onsemantic units, a Bayesian network is used to reason the probabilities of events. The experiments forshoot and card event detection in soccer videos show that the proposed method has an encouragingperformance.展开更多
Focusing on the problem of goal event detection in soccer videos,a novel method based on Hidden Markov Model(HMM) and the semantic rule is proposed.Firstly,a HMM for a goal event is constructed.Then a Normalized Seman...Focusing on the problem of goal event detection in soccer videos,a novel method based on Hidden Markov Model(HMM) and the semantic rule is proposed.Firstly,a HMM for a goal event is constructed.Then a Normalized Semantic Weighted Sum(NSWS) rule is established by defining a new feature of shots,semantic observation weight.The test video is detected based on the HMM and the NSWS rule,respectively.Finally,a fusion scheme based on logic distance is proposed and the detection results of the HMM and the NSWS rule are fused by optimal weights in the decision level,obtaining the final result.Experimental results indicate that the proposed method achieves 96.43% precision and 100% recall,which shows the effectiveness of this letter.展开更多
1. IntroductionHumans have the ability (or competence) to think logically, and this is an undeniable fact. However,what this ability consists in is a difficult question. It might be said that logical ability consists ...1. IntroductionHumans have the ability (or competence) to think logically, and this is an undeniable fact. However,what this ability consists in is a difficult question. It might be said that logical ability consists in theknowledge of a set of logic rules. But what are those logic rules? For centuries logicians have devel-展开更多
Semantic duplicates in databases represent today an important data quality challenge which leads to bad decisions. In large databases, we sometimes find ourselves with tens of thousands of duplicates, which necessitat...Semantic duplicates in databases represent today an important data quality challenge which leads to bad decisions. In large databases, we sometimes find ourselves with tens of thousands of duplicates, which necessitates an automatic deduplication. For this, it is necessary to detect duplicates, with a fairly reliable method to find as many duplicates as possible and powerful enough to run in a reasonable time. This paper proposes and compares on real data effective duplicates detection methods for automatic deduplication of files based on names, working with French texts or English texts, and the names of people or places, in Africa or in the West. After conducting a more complete classification of semantic duplicates than the usual classifications, we introduce several methods for detecting duplicates whose average complexity observed is less than O(2n). Through a simple model, we highlight a global efficacy rate, combining precision and recall. We propose a new metric distance between records, as well as rules for automatic duplicate detection. Analyses made on a database containing real data for an administration in Central Africa, and on a known standard database containing names of restaurants in the USA, have shown better results than those of known methods, with a lesser complexity.展开更多
The detection of phishing and legitimate websites is considered a great challenge for web service providers because the users of such websites are indistinguishable.Phishing websites also create traffic in the entire ...The detection of phishing and legitimate websites is considered a great challenge for web service providers because the users of such websites are indistinguishable.Phishing websites also create traffic in the entire network.Another phishing issue is the broadening malware of the entire network,thus highlighting the demand for their detection while massive datasets(i.e.,big data)are processed.Despite the application of boosting mechanisms in phishing detection,these methods are prone to significant errors in their output,specifically due to the combination of all website features in the training state.The upcoming big data system requires MapReduce,a popular parallel programming,to process massive datasets.To address these issues,a probabilistic latent semantic and greedy levy gradient boosting(PLS-GLGB)algorithm for website phishing detection using MapReduce is proposed.A feature selection-based model is provided using a probabilistic intersective latent semantic preprocessing model to minimize errors in website phishing detection.Here,the missing data in each URL are identified and discarded for further processing to ensure data quality.Subsequently,with the preprocessed features(URLs),feature vectors are updated by the greedy levy divergence gradient(model)that selects the optimal features in the URL and accurately detects the websites.Thus,greedy levy efficiently differentiates between phishing websites and legitimate websites.Experiments are conducted using one of the largest public corpora of a website phish tank dataset.Results show that the PLS-GLGB algorithm for website phishing detection outperforms stateof-the-art phishing detection methods.Significant amounts of phishing detection time and errors are also saved during the detection of website phishing.展开更多
Semantic change detection is extension of change detection task in which it is not only used to identify the changed regions but also to analyze the land area semantic(labels/categories)details before and after the ti...Semantic change detection is extension of change detection task in which it is not only used to identify the changed regions but also to analyze the land area semantic(labels/categories)details before and after the timelines are analyzed.Periodical land change analysis is used for many real time applications for valuation purposes.Majority of the research works are focused on Convolutional Neural Networks(CNN)which tries to analyze changes alone.Semantic information of changes appears to be missing,there by absence of communication between the different semantic timelines and changes detected over the region happens.To overcome this limitation,a CNN network is proposed incorporating the Resnet-34 pre-trained model on Fully Convolutional Network(FCN)blocks for exploring the temporal data of satellite images in different timelines and change map between these two timelines are analyzed.Further this model achieves better results by analyzing the semantic information between the timelines and based on localized information collected from skip connections which help in generating a better change map with the categories that might have changed over a land area across timelines.Proposed model effectively examines the semantic changes such as from-to changes on land over time period.The experimental results on SECOND(Semantic Change detectiON Dataset)indicates that the proposed model yields notable improvement in performance when it is compared with the existing approaches and this also improves the semantic segmentation task on images over different timelines and the changed areas of land area across timelines.展开更多
Purpose: This paper aims to provide a method to detect research communities based on research interest in researcher network, which combines the topological structure and vertex attributes in a unified manner.Design/m...Purpose: This paper aims to provide a method to detect research communities based on research interest in researcher network, which combines the topological structure and vertex attributes in a unified manner.Design/methodology/approach: A heterogeneous researcher network has been constructed by combining multiple relations of academic researchers. Vertex attributes and their similarities were considered and calculated. An approach has been proposed and tested to detect research community in research organizations based on this multi-relation researcher network.Findings: Detection of topologically well-connected, semantically coherent and meaningful research community was achieved.Research limitations: The sample size of evaluation experiments was relatively small. In the present study, a limited number of 72 researchers were analyzed for constructing researcher network and detecting research community. Therefore, a large sample size is required to give more information and reliable results.Practical implications: The proposed multi-relation researcher network and approaches for discovering research communities of similar research interests will contribute to collective innovation behavior such as brainstorming and to promote interdisciplinary cooperation.Originality/value: Recent researches on community detection devote most efforts to singlerelation researcher networks and put the main focus on the topological structure of networks.In reality, there exist multi-relation social networks. Vertex attribute also plays an important role in community detection. The present study combined multiple single-relational researcher networks into a multi-relational network and proposed a structure-attribute clustering method for detecting research community in research organizations.展开更多
基金a grant from the National Natural Science Foundation of China(Nos.11905239,12005248 and 12105303).
文摘With the rapid development of the mobile communication and the Internet,the previous web anomaly detectionand identificationmodels were built relying on security experts’empirical knowledge and attack features.Althoughthis approach can achieve higher detection performance,it requires huge human labor and resources to maintainthe feature library.In contrast,semantic feature engineering can dynamically discover new semantic featuresand optimize feature selection by automatically analyzing the semantic information contained in the data itself,thus reducing dependence on prior knowledge.However,current semantic features still have the problem ofsemantic expression singularity,as they are extracted from a single semantic mode such as word segmentation,character segmentation,or arbitrary semantic feature extraction.This paper extracts features of web requestsfrom dual semantic granularity,and proposes a semantic feature fusion method to solve the above problems.Themethod first preprocesses web requests,and extracts word-level and character-level semantic features of URLs viaconvolutional neural network(CNN),respectively.By constructing three loss functions to reduce losses betweenfeatures,labels and categories.Experiments on the HTTP CSIC 2010,Malicious URLs and HttpParams datasetsverify the proposedmethod.Results show that compared withmachine learning,deep learningmethods and BERTmodel,the proposed method has better detection performance.And it achieved the best detection rate of 99.16%in the dataset HttpParams.
基金supported in part by the IoT Intelligent Microsystem Center of Tsinghua University-China Mobile Joint Research Institute.
文摘Automated optical inspection(AOI)is a significant process in printed circuit board assembly(PCBA)production lines which aims to detect tiny defects in PCBAs.Existing AOI equipment has several deficiencies including low throughput,large computation cost,high latency,and poor flexibility,which limits the efficiency of online PCBA inspection.In this paper,a novel PCBA defect detection method based on a lightweight deep convolution neural network is proposed.In this method,the semantic segmentation model is combined with a rule-based defect recognition algorithm to build up a defect detection frame-work.To improve the performance of the model,extensive real PCBA images are collected from production lines as datasets.Some optimization methods have been applied in the model according to production demand and enable integration in lightweight computing devices.Experiment results show that the production line using our method realizes a throughput more than three times higher than traditional methods.Our method can be integrated into a lightweight inference system and pro-mote the flexibility of AOI.The proposed method builds up a general paradigm and excellent example for model design and optimization oriented towards industrial requirements.
基金This work was supported by the National Natural Science Foundation of China(Grant No.U20A20197).
文摘We propose a novel image segmentation algorithm to tackle the challenge of limited recognition and segmentation performance in identifying welding seam images during robotic intelligent operations.Initially,to enhance the capability of deep neural networks in extracting geometric attributes from depth images,we developed a novel deep geometric convolution operator(DGConv).DGConv is utilized to construct a deep local geometric feature extraction module,facilitating a more comprehensive exploration of the intrinsic geometric information within depth images.Secondly,we integrate the newly proposed deep geometric feature module with the Fully Convolutional Network(FCN8)to establish a high-performance deep neural network algorithm tailored for depth image segmentation.Concurrently,we enhance the FCN8 detection head by separating the segmentation and classification processes.This enhancement significantly boosts the network’s overall detection capability.Thirdly,for a comprehensive assessment of our proposed algorithm and its applicability in real-world industrial settings,we curated a line-scan image dataset featuring weld seams.This dataset,named the Standardized Linear Depth Profile(SLDP)dataset,was collected from actual industrial sites where autonomous robots are in operation.Ultimately,we conducted experiments utilizing the SLDP dataset,achieving an average accuracy of 92.7%.Our proposed approach exhibited a remarkable performance improvement over the prior method on the identical dataset.Moreover,we have successfully deployed the proposed algorithm in genuine industrial environments,fulfilling the prerequisites of unmanned robot operations.
基金supported by the National Natural Science Foundation of China(No.62271199)the Natural Science Foundation of Hunan Province,China(No.2020JJ5170)the Scientific Research Fund of Hunan Provincial Education Department(No.18C0299)。
文摘With the continuous development and utilization of marine resources,the underwater target detection has gradually become a popular research topic in the field of underwater robot operations and target detection.However,it is difficult to combine the environmental semantic information and the semantic information of targets at different scales by detection algorithms due to the complex underwater environment.In this paper,a cascade model based on the UGC-YOLO network structure with high detection accuracy is proposed.The YOLOv3 convolutional neural network is employed as the baseline structure.By fusing the global semantic information between two residual stages in the parallel structure of the feature extraction network,the perception of underwater targets is improved and the detection rate of hard-to-detect underwater objects is raised.Furthermore,the deformable convolution is applied to capture longrange semantic dependencies and PPM pooling is introduced in the highest layer network for aggregating semantic information.Finally,a multi-scale weighted fusion approach is presented for learning semantic information at different scales.Experiments are conducted on an underwater test dataset and the results have demonstrated that our proposed algorithm could detect aquatic targets in complex degraded underwater images.Compared with the baseline network algorithm,the Common Objects in Context(COCO)evaluation metric has been improved by 4.34%.
基金supported in part by the National Natural Science Foundation of China under Grant 62272062the Scientific Research Fund of Hunan Provincial Transportation Department(No.202143)the Open Fund ofKey Laboratory of Safety Control of Bridge Engineering,Ministry of Education(Changsha University of Science Technology)under Grant 21KB07.
文摘Log anomaly detection is an important paradigm for system troubleshooting.Existing log anomaly detection based on Long Short-Term Memory(LSTM)networks is time-consuming to handle long sequences.Transformer model is introduced to promote efficiency.However,most existing Transformer-based log anomaly detection methods convert unstructured log messages into structured templates by log parsing,which introduces parsing errors.They only extract simple semantic feature,which ignores other features,and are generally supervised,relying on the amount of labeled data.To overcome the limitations of existing methods,this paper proposes a novel unsupervised log anomaly detection method based on multi-feature(UMFLog).UMFLog includes two sub-models to consider two kinds of features:semantic feature and statistical feature,respectively.UMFLog applies the log original content with detailed parameters instead of templates or template IDs to avoid log parsing errors.In the first sub-model,UMFLog uses Bidirectional Encoder Representations from Transformers(BERT)instead of random initialization to extract effective semantic feature,and an unsupervised hypersphere-based Transformer model to learn compact log sequence representations and obtain anomaly candidates.In the second sub-model,UMFLog exploits a statistical feature-based Variational Autoencoder(VAE)about word occurrence times to identify the final anomaly from anomaly candidates.Extensive experiments and evaluations are conducted on three real public log datasets.The results show that UMFLog significantly improves F1-scores compared to the state-of-the-art(SOTA)methods because of the multi-feature.
基金the National Natural Science Foundation of China(No.62063006)to the Natural Science Foundation of Guangxi Province(No.2023GXNS-FAA026025)+3 种基金to the Innovation Fund of Chinese Universities Industry-University-Research(ID:2021RYC06005)to the Research Project for Young and Middle-aged Teachers in Guangxi Universities(ID:2020KY15013)to the Special Research Project of Hechi University(ID:2021GCC028)supported by the Project of Outstanding Thousand Young Teachers’Training in Higher Education Institutions of Guangxi,Guangxi Colleges and Universities Key Laboratory of AI and Information Processing(Hechi University),Education Department of Guangxi Zhuang Autonomous Region.
文摘Visual simultaneous localization and mapping(SLAM)is crucial in robotics and autonomous driving.However,traditional visual SLAM faces challenges in dynamic environments.To address this issue,researchers have proposed semantic SLAM,which combines object detection,semantic segmentation,instance segmentation,and visual SLAM.Despite the growing body of literature on semantic SLAM,there is currently a lack of comprehensive research on the integration of object detection and visual SLAM.Therefore,this study aims to gather information from multiple databases and review relevant literature using specific keywords.It focuses on visual SLAM based on object detection,covering different aspects.Firstly,it discusses the current research status and challenges in this field,highlighting methods for incorporating semantic information from object detection networks into mileage measurement,closed-loop detection,and map construction.It also compares the characteristics and performance of various visual SLAM object detection algorithms.Lastly,it provides an outlook on future research directions and emerging trends in visual SLAM.Research has shown that visual SLAM based on object detection has significant improvements compared to traditional SLAM in dynamic point removal,data association,point cloud segmentation,and other technologies.It can improve the robustness and accuracy of the entire SLAM system and can run in real time.With the continuous optimization of algorithms and the improvement of hardware level,object visual SLAM has great potential for development.
基金the National Natural Science Foundation of China(61772386)Joint fund project(nsfc-guangdong big data science center project),project number:U1611262,Hubei University of Science and Technology,Master of Engineering,special construction,project number:2018-19GZ01,Hubei University of Science and Technology Teaching Reform Project,project number:2018-XB-023,S201910927028.
文摘This paper proposes a novel method of lane detection,which adopts VGG16 as the basis of convolutional neural network to extract lane line features by cavity convolution,wherein the lane lines are divided into dotted lines and solid lines.Expanding the field of experience through hollow convolution,the full connection layer of the network is discarded,the last largest pooling layer of the VGG16 network is removed,and the processing of the last three convolution layers is replaced by hole convolution.At the same time,CNN adopts the encoder and decoder structure mode,and uses the index function of the maximum pooling layer in the decoder part to upsample the encoder in a counter-pooling manner,realizing semantic segmentation.And combined with the instance segmentation,and finally through the fitting to achieve the detection of the lane line.In addition,the currently disclosed lane line data sets are relatively small,and there is no distinction between lane solid lines and dashed lines.To this end,our work made a lane line data set for the lane virtual and real identification,and based on the proposed algorithm effective verification of the data set achieved by the increased segmentation.The final test shows that the proposed method has a good balance between lane detection speed and accuracy,which has good robustness.
基金This work was supported by National Natural Science Foundation of China(No.62172308,No.U1626107,No.61972297,No.62172144,and No.62062019).
文摘Power Shell has been widely deployed in fileless malware and advanced persistent threat(APT)attacks due to its high stealthiness and live-off-theland technique.However,existing works mainly focus on deobfuscation and malicious detection,lacking the malicious Power Shell families classification and behavior analysis.Moreover,the state-of-the-art methods fail to capture fine-grained features and semantic relationships,resulting in low robustness and accuracy.To this end,we propose Power Detector,a novel malicious Power Shell script detector based on multimodal semantic fusion and deep learning.Specifically,we design four feature extraction methods to extract key features from character,token,abstract syntax tree(AST),and semantic knowledge graph.Then,we intelligently design four embeddings(i.e.,Char2Vec,Token2Vec,AST2Vec,and Rela2Vec) and construct a multi-modal fusion algorithm to concatenate feature vectors from different views.Finally,we propose a combined model based on transformer and CNN-Bi LSTM to implement Power Shell family detection.Our experiments with five types of Power Shell attacks show that PowerDetector can accurately detect various obfuscated and stealth PowerShell scripts,with a 0.9402 precision,a 0.9358 recall,and a 0.9374 F1-score.Furthermore,through singlemodal and multi-modal comparison experiments,we demonstrate that PowerDetector’s multi-modal embedding and deep learning model can achieve better accuracy and even identify more unknown attacks.
文摘In recent years,computer visionfinds wide applications in maritime surveillance with its sophisticated algorithms and advanced architecture.Auto-matic ship detection with computer vision techniques provide an efficient means to monitor as well as track ships in water bodies.Waterways being an important medium of transport require continuous monitoring for protection of national security.The remote sensing satellite images of ships in harbours and water bodies are the image data that aid the neural network models to localize ships and to facilitate early identification of possible threats at sea.This paper proposes a deep learning based model capable enough to classify between ships and no-ships as well as to localize ships in the original images using bounding box tech-nique.Furthermore,classified ships are again segmented with deep learning based auto-encoder model.The proposed model,in terms of classification,provides suc-cessful results generating 99.5%and 99.2%validation and training accuracy respectively.The auto-encoder model also produces 85.1%and 84.2%validation and training accuracies.Moreover the IoU metric of the segmented images is found to be of 0.77 value.The experimental results reveal that the model is accu-rate and can be implemented for automatic ship detection in water bodies consid-ering remote sensing satellite images as input to the computer vision system.
文摘Data fusion is usually an important process in multi-sensor remotely sensed imagery integration environments with the aim of enriching features lacking in the sensors involved in the fusion process. This technique has attracted much interest in many researches especially in the field of agriculture. On the other hand, deep learning (DL) based semantic segmentation shows high performance in remote sensing classification, and it requires large datasets in a supervised learning way. In the paper, a method of fusing multi-source remote sensing images with convolution neural networks (CNN) for semantic segmentation is proposed and applied to identify crops. Venezuelan Remote Sensing Satellite-2 (VRSS-2) and the high-resolution of Google Earth (GE) imageries have been used and more than 1000 sample sets have been collected for supervised learning process. The experiment results show that the crops extraction with an average overall accuracy more than 93% has been obtained, which demonstrates that data fusion combined with DL is highly feasible to crops extraction from satellite images and GE imagery, and it shows that deep learning techniques can serve as an invaluable tools for larger remote sensing data fusion frameworks, specifically for the applications in precision farming.
基金973 Program,Grant/Award Number:2014CB340504The State Key Program of National Natural Science of China,Grant/Award Number:61533018+3 种基金National Natural Science Foundation of China,Grant/Award Number:61402220The Philosophy and Social Science Foundation of Hunan Province,Grant/Award Number:16YBA323Natural Science Foundation of Hunan Province,Grant/Award Number:2020JJ4525Scientific Research Fund of Hunan Provincial Education Department,Grant/Award Number:18B279,19A439。
文摘Current Chinese event detection methods commonly use word embedding to capture semantic representation,but these methods find it difficult to capture the dependence relationship between the trigger words and other words in the same sentence.Based on the simple evaluation,it is known that a dependency parser can effectively capture dependency relationships and improve the accuracy of event categorisation.This study proposes a novel architecture that models a hybrid representation to summarise semantic and structural information from both characters and words.This model can capture rich semantic features for the event detection task by incorporating the semantic representation generated from the dependency parser.The authors evaluate different models on kbp 2017 corpus.The experimental results show that the proposed method can significantly improve performance in Chinese event detection.
文摘A semantic unit based event detection scheme in soccer videos is proposed in this paper.The scheme can be characterized as a three-layer framework. At the lowest layer, low-level featuresincluding color, texture, edge, shape, and motion are extracted. High-level semantic events aredefined at the highest layer. In order to connect low-level features and high-level semantics, wedesign and define some semantic units at the intermediate layer. A semantic unit is composed of asequence of consecutives frames with the same cue that is deduced from low-level features. Based onsemantic units, a Bayesian network is used to reason the probabilities of events. The experiments forshoot and card event detection in soccer videos show that the proposed method has an encouragingperformance.
基金Supported by the National Natural Science Foundation of China (No. 61072110)the Industrial Tackling Project of Shaanxi Province (2010K06-20)the Natural Science Foundation of Shaanxi Province (SJ08F15)
文摘Focusing on the problem of goal event detection in soccer videos,a novel method based on Hidden Markov Model(HMM) and the semantic rule is proposed.Firstly,a HMM for a goal event is constructed.Then a Normalized Semantic Weighted Sum(NSWS) rule is established by defining a new feature of shots,semantic observation weight.The test video is detected based on the HMM and the NSWS rule,respectively.Finally,a fusion scheme based on logic distance is proposed and the detection results of the HMM and the NSWS rule are fused by optimal weights in the decision level,obtaining the final result.Experimental results indicate that the proposed method achieves 96.43% precision and 100% recall,which shows the effectiveness of this letter.
文摘1. IntroductionHumans have the ability (or competence) to think logically, and this is an undeniable fact. However,what this ability consists in is a difficult question. It might be said that logical ability consists in theknowledge of a set of logic rules. But what are those logic rules? For centuries logicians have devel-
文摘Semantic duplicates in databases represent today an important data quality challenge which leads to bad decisions. In large databases, we sometimes find ourselves with tens of thousands of duplicates, which necessitates an automatic deduplication. For this, it is necessary to detect duplicates, with a fairly reliable method to find as many duplicates as possible and powerful enough to run in a reasonable time. This paper proposes and compares on real data effective duplicates detection methods for automatic deduplication of files based on names, working with French texts or English texts, and the names of people or places, in Africa or in the West. After conducting a more complete classification of semantic duplicates than the usual classifications, we introduce several methods for detecting duplicates whose average complexity observed is less than O(2n). Through a simple model, we highlight a global efficacy rate, combining precision and recall. We propose a new metric distance between records, as well as rules for automatic duplicate detection. Analyses made on a database containing real data for an administration in Central Africa, and on a known standard database containing names of restaurants in the USA, have shown better results than those of known methods, with a lesser complexity.
文摘The detection of phishing and legitimate websites is considered a great challenge for web service providers because the users of such websites are indistinguishable.Phishing websites also create traffic in the entire network.Another phishing issue is the broadening malware of the entire network,thus highlighting the demand for their detection while massive datasets(i.e.,big data)are processed.Despite the application of boosting mechanisms in phishing detection,these methods are prone to significant errors in their output,specifically due to the combination of all website features in the training state.The upcoming big data system requires MapReduce,a popular parallel programming,to process massive datasets.To address these issues,a probabilistic latent semantic and greedy levy gradient boosting(PLS-GLGB)algorithm for website phishing detection using MapReduce is proposed.A feature selection-based model is provided using a probabilistic intersective latent semantic preprocessing model to minimize errors in website phishing detection.Here,the missing data in each URL are identified and discarded for further processing to ensure data quality.Subsequently,with the preprocessed features(URLs),feature vectors are updated by the greedy levy divergence gradient(model)that selects the optimal features in the URL and accurately detects the websites.Thus,greedy levy efficiently differentiates between phishing websites and legitimate websites.Experiments are conducted using one of the largest public corpora of a website phish tank dataset.Results show that the PLS-GLGB algorithm for website phishing detection outperforms stateof-the-art phishing detection methods.Significant amounts of phishing detection time and errors are also saved during the detection of website phishing.
文摘Semantic change detection is extension of change detection task in which it is not only used to identify the changed regions but also to analyze the land area semantic(labels/categories)details before and after the timelines are analyzed.Periodical land change analysis is used for many real time applications for valuation purposes.Majority of the research works are focused on Convolutional Neural Networks(CNN)which tries to analyze changes alone.Semantic information of changes appears to be missing,there by absence of communication between the different semantic timelines and changes detected over the region happens.To overcome this limitation,a CNN network is proposed incorporating the Resnet-34 pre-trained model on Fully Convolutional Network(FCN)blocks for exploring the temporal data of satellite images in different timelines and change map between these two timelines are analyzed.Further this model achieves better results by analyzing the semantic information between the timelines and based on localized information collected from skip connections which help in generating a better change map with the categories that might have changed over a land area across timelines.Proposed model effectively examines the semantic changes such as from-to changes on land over time period.The experimental results on SECOND(Semantic Change detectiON Dataset)indicates that the proposed model yields notable improvement in performance when it is compared with the existing approaches and this also improves the semantic segmentation task on images over different timelines and the changed areas of land area across timelines.
基金supported by the National Natural Science Foundation of China(Grant No.:71203164)
文摘Purpose: This paper aims to provide a method to detect research communities based on research interest in researcher network, which combines the topological structure and vertex attributes in a unified manner.Design/methodology/approach: A heterogeneous researcher network has been constructed by combining multiple relations of academic researchers. Vertex attributes and their similarities were considered and calculated. An approach has been proposed and tested to detect research community in research organizations based on this multi-relation researcher network.Findings: Detection of topologically well-connected, semantically coherent and meaningful research community was achieved.Research limitations: The sample size of evaluation experiments was relatively small. In the present study, a limited number of 72 researchers were analyzed for constructing researcher network and detecting research community. Therefore, a large sample size is required to give more information and reliable results.Practical implications: The proposed multi-relation researcher network and approaches for discovering research communities of similar research interests will contribute to collective innovation behavior such as brainstorming and to promote interdisciplinary cooperation.Originality/value: Recent researches on community detection devote most efforts to singlerelation researcher networks and put the main focus on the topological structure of networks.In reality, there exist multi-relation social networks. Vertex attribute also plays an important role in community detection. The present study combined multiple single-relational researcher networks into a multi-relational network and proposed a structure-attribute clustering method for detecting research community in research organizations.