Several security solutions have been proposed to detect network abnormal behavior. However, successful attacks is still a big concern in computer society. Lots of security breaches, like Distributed Denial of Service(...Several security solutions have been proposed to detect network abnormal behavior. However, successful attacks is still a big concern in computer society. Lots of security breaches, like Distributed Denial of Service(DDoS),botnets, spam, phishing, and so on, are reported every day, while the number of attacks are still increasing. In this paper, a novel voting-based deep learning framework, called VNN, is proposed to take the advantage of any kinds of deep learning structures. Considering several models created by different aspects of data and various deep learning structures, VNN provides the ability to aggregate the best models in order to create more accurate and robust results. Therefore, VNN helps the security specialists to detect more complicated attacks. Experimental results over KDDCUP'99 and CTU-13, as two well known and more widely employed datasets in computer network area, revealed the voting procedure was highly effective to increase the system performance, where the false alarms were reduced up to 75% in comparison with the original deep learning models, including Deep Neural Network(DNN), Convolutional Neural Network(CNN), Long Short-Term Memory(LSTM), and Gated Recurrent Unit(GRU).展开更多
Few-shot learning has emerged as a crucial technique for coral species classification,addressing the challenge of limited labeled data in underwater environments.This study introduces an optimized few-shot learning mo...Few-shot learning has emerged as a crucial technique for coral species classification,addressing the challenge of limited labeled data in underwater environments.This study introduces an optimized few-shot learning model that enhances classification accuracy while minimizing reliance on extensive data collection.The proposed model integrates a hybrid similarity measure combining Euclidean distance and cosine similarity,effectively capturing both feature magnitude and directional relationships.This approach achieves a notable accuracy of 71.8%under a 5-way 5-shot evaluation,outperforming state-of-the-art models such as Prototypical Networks,FEAT,and ESPT by up to 10%.Notably,the model demonstrates high precision in classifying Siderastreidae(87.52%)and Fungiidae(88.95%),underscoring its effectiveness in distinguishing subtle morphological differences.To further enhance performance,we incorporate a self-supervised learning mechanism based on contrastive learning,enabling the model to extract robust representations by leveraging local structural patterns in corals.This enhancement significantly improves classification accuracy,particularly for species with high intra-class variation,leading to an overall accuracy of 76.52%under a 5-way 10-shot evaluation.Additionally,the model exploits the repetitive structures inherent in corals,introducing a local feature aggregation strategy that refines classification through spatial information integration.Beyond its technical contributions,this study presents a scalable and efficient approach for automated coral reef monitoring,reducing annotation costs while maintaining high classification accuracy.By improving few-shot learning performance in underwater environments,our model enhances monitoring accuracy by up to 15%compared to traditional methods,offering a practical solution for large-scale coral conservation efforts.展开更多
This article presents a novel people-tracking approach to cope with partial occlusions caused by scene objects. Instead of predicting when and where the occlusions will occur, a part-based model is used to model the p...This article presents a novel people-tracking approach to cope with partial occlusions caused by scene objects. Instead of predicting when and where the occlusions will occur, a part-based model is used to model the pixel distribution of the target body under occlusion. The subdivided patches corresponding to a template image will be tracked independently using Markov chain Monte Carlo （MCMC） method. A set of voting-based rules is established for the patch-tracking result to verify if the target is indeed located at the estimated position. Experiments show the effectiveness of the proposed method.展开更多
基金supported by the National Natural Science Foundation of China (No. 61872212)the National Key Research and Development Program of China (No.2016YFB1000102)。
文摘Several security solutions have been proposed to detect network abnormal behavior. However, successful attacks is still a big concern in computer society. Lots of security breaches, like Distributed Denial of Service(DDoS),botnets, spam, phishing, and so on, are reported every day, while the number of attacks are still increasing. In this paper, a novel voting-based deep learning framework, called VNN, is proposed to take the advantage of any kinds of deep learning structures. Considering several models created by different aspects of data and various deep learning structures, VNN provides the ability to aggregate the best models in order to create more accurate and robust results. Therefore, VNN helps the security specialists to detect more complicated attacks. Experimental results over KDDCUP'99 and CTU-13, as two well known and more widely employed datasets in computer network area, revealed the voting procedure was highly effective to increase the system performance, where the false alarms were reduced up to 75% in comparison with the original deep learning models, including Deep Neural Network(DNN), Convolutional Neural Network(CNN), Long Short-Term Memory(LSTM), and Gated Recurrent Unit(GRU).
基金funded by theNational Science and TechnologyCouncil(NSTC),Taiwan,under grant numbers NSTC 112-2634-F-019-001 and NSTC 113-2634-F-A49-007.
文摘Few-shot learning has emerged as a crucial technique for coral species classification,addressing the challenge of limited labeled data in underwater environments.This study introduces an optimized few-shot learning model that enhances classification accuracy while minimizing reliance on extensive data collection.The proposed model integrates a hybrid similarity measure combining Euclidean distance and cosine similarity,effectively capturing both feature magnitude and directional relationships.This approach achieves a notable accuracy of 71.8%under a 5-way 5-shot evaluation,outperforming state-of-the-art models such as Prototypical Networks,FEAT,and ESPT by up to 10%.Notably,the model demonstrates high precision in classifying Siderastreidae(87.52%)and Fungiidae(88.95%),underscoring its effectiveness in distinguishing subtle morphological differences.To further enhance performance,we incorporate a self-supervised learning mechanism based on contrastive learning,enabling the model to extract robust representations by leveraging local structural patterns in corals.This enhancement significantly improves classification accuracy,particularly for species with high intra-class variation,leading to an overall accuracy of 76.52%under a 5-way 10-shot evaluation.Additionally,the model exploits the repetitive structures inherent in corals,introducing a local feature aggregation strategy that refines classification through spatial information integration.Beyond its technical contributions,this study presents a scalable and efficient approach for automated coral reef monitoring,reducing annotation costs while maintaining high classification accuracy.By improving few-shot learning performance in underwater environments,our model enhances monitoring accuracy by up to 15%compared to traditional methods,offering a practical solution for large-scale coral conservation efforts.
基金supported by the National Natural Science Foundation of China(60772114)
文摘This article presents a novel people-tracking approach to cope with partial occlusions caused by scene objects. Instead of predicting when and where the occlusions will occur, a part-based model is used to model the pixel distribution of the target body under occlusion. The subdivided patches corresponding to a template image will be tracked independently using Markov chain Monte Carlo （MCMC） method. A set of voting-based rules is established for the patch-tracking result to verify if the target is indeed located at the estimated position. Experiments show the effectiveness of the proposed method.
