2023 3rd International Symposium on Computer Technology and Information Science (ISCTIS)最新文献

{"title":"Ultra-lightweight Image Compressive Sensing Reconstruction Algorithm Based on Knowledge Distillation","authors":"Yuxin Yang, Wenjie Yuan","doi":"10.1109/ISCTIS58954.2023.10213013","DOIUrl":"https://doi.org/10.1109/ISCTIS58954.2023.10213013","url":null,"abstract":"Deep neural networks have been shown to improve the quality of image compressive sensing reconstruction, but they are often limited in practical applications due to computational complexity. To address this issue, this paper proposes an ultra-lightweight image compressive sensing reconstruction network. In this network, an adaptive bipolar sampling module is used for information extraction, while sub-pixel convolution and depth-separable convolution are employed for reconstruction to reduce network parameters. Additionally, an improved knowledge distillation algorithm is used to train the network, which further enhances its reconstruction performance. Experimental results show that the proposed ultra-lightweight network has the lowesr computational complexity and the faster reconstruction speed.","PeriodicalId":334790,"journal":{"name":"2023 3rd International Symposium on Computer Technology and Information Science (ISCTIS)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123484152","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Novel Graph Neural Network-Based Automated Fault Localization Technique for Novice Programs","authors":"Bin Du, Zhanwen Zhang, Hengyuan Liu, Shumei Wu, Luxi Fan, Zexing Chang","doi":"10.1109/ISCTIS58954.2023.10213100","DOIUrl":"https://doi.org/10.1109/ISCTIS58954.2023.10213100","url":null,"abstract":"In recently years, Massive Open Online Courses have garnered significant attention, becoming a fundamental learning method for programming. However, debugging poses a challenge for novice programmers lacking programming experience due to its considerable effort, time cost, and professional experience. This has spurred interest in automatically localizing faults for novice programs. Regrettably, traditional fault localization techniques are often insufficient due to the inherent attributes of novice programs, such as concise code, straightforward logic, and limited passed tests. In this paper, we propose a novel fault localization technique, GccvFL, utilizing graph representation learning. Specifically, we first collect variable value sequences in novice programs and merge the dynamic and static features of the code. Then we reserve this detailed information of the faulty program into one graph. A gated graph neural network (GGNN) is used to learn useful features from the graph and output the rank list according to their suspiciousness. Ultimately, GccvFL enhances fault localization by integrating these features within the model. Our evaluation of 89 novice programs from the Beijing University of Chemical Technology Online Judge (BUCTOJ) dataset shows GccvFL significantly outperforms existing fault localization techniques on all the observed metrics. GccvFL localizes eight fault statements more than state-of-the-art techniques in TOP-1, showing that GccvFL's superior fault localization performance in novice programs.","PeriodicalId":334790,"journal":{"name":"2023 3rd International Symposium on Computer Technology and Information Science (ISCTIS)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123638341","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multimodal Interactive Supervised Pedestrian Detection Based on yolov5","authors":"Mingyue Li, Lianzhong Wang, Zhe Zheng, WenpengCui Cui, Rui Liu, Yingying Chi","doi":"10.1109/ISCTIS58954.2023.10213155","DOIUrl":"https://doi.org/10.1109/ISCTIS58954.2023.10213155","url":null,"abstract":"Pedestrian detection and recognition is one of the important and fundamental tasks of environmental awareness in autonomous driving. Some existing research methods are based on visual images (RGB) for detection, but detection methods that only use visual images cannot meet harsh detection environments, such as in cloudy, rainy, foggy, and poor lighting environments, where the detection effect is poor. In recent years, fusion detection based on visual images (RGB) and thermal images (Thermal) has received increasing attention, but there are still some problems in fusion strategies. The purpose of RGB Thermal pedestrian detection is to fuse complementary visible light images and thermal infrared information to improve the performance of pedestrian detection in day and night environments [1]. In recent years, many experts and scholars have done a lot of research work on this issue [2]–[6], effectively integrating the two modal data of RGB and thermal infrared images, and have achieved some results. However, these methods directly use fused features for pedestrian detection, without considering the problem that the quality of the resulting fused features may not be high. Therefore, multimodal feature fusion pedestrian detection structures need to improve the quality of fusion features, which plays a crucial role in the research of multimodal data fusion pedestrian detection in autonomous driving. The Cross-Modal Supervision (CMS) model designed and implemented in this article has been experimentally verified on the public Kaist [7] dataset. The experimental results show that the accuracy of the cross modal supervised model on the Kaist dataset reaches 53.68°/°and the miss rate decreases to 46.32°/°.","PeriodicalId":334790,"journal":{"name":"2023 3rd International Symposium on Computer Technology and Information Science (ISCTIS)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114081185","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Local Motion Blurred Image Restoration Based on the Reciprocal of DCT High-Frequency Mean Incremental Prior","authors":"Xiao Han, Zhen Jia","doi":"10.1109/ISCTIS58954.2023.10213064","DOIUrl":"https://doi.org/10.1109/ISCTIS58954.2023.10213064","url":null,"abstract":"Aiming at the blur caused by fast-moving objects in surveillance video, we propose a new method to restore blurred images. First, we use the PiCANet saliency detection algorithm to obtain the saliency map of the blurred area. Then we use the saliency map to guide the soft segmentation algorithm to divide the blurred image into foreground and background layers. Second, we propose a reciprocal of DCT high-frequency mean incremental prior to constraining the solution space of clear latent images under the framework of maximum a posteriori probability. Finally, we also propose an improved bilateral filtering algorithm to enhance the details of the restored image. The experimental results show that our algorithm's deblurring visual effect and objective evaluation index are superior to other algorithms.","PeriodicalId":334790,"journal":{"name":"2023 3rd International Symposium on Computer Technology and Information Science (ISCTIS)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116325640","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Fuel cell fault classification based on long and short-term memory full convolutional neural networks","authors":"Ning Zhou, Hao Chen, Jianxin Zhou","doi":"10.1109/ISCTIS58954.2023.10213112","DOIUrl":"https://doi.org/10.1109/ISCTIS58954.2023.10213112","url":null,"abstract":"The development of modern automotive industry has accelerated the technological development and commercial application of fuel cells due to the challenges of resources and environment. The establishment of a sound failure prediction and management (PHM) system for hydrogen energy vehicles can achieve the goal of improving product quality and saving energy. Proton exchange membrane fuel cell (PEMFC) fault classification is the key to achieve the PHM system. The dataset used in this paper is realtime data collected on a live fuel cell vehicle. Considering the impact of unbalanced fault samples on fault classification accuracy, hybrid sampling is used in the data preprocessing stage to balance the number of samples, and a long and short-term memory full convolutional neural network is proposed to enhance the deep learning-based time series classification method by using global temporal attention and temporal pseudo-Gaussian enhanced self-attention. The experimental results demonstrate that the method in this paper has higher classification accuracy and precision compared with the traditional methods.","PeriodicalId":334790,"journal":{"name":"2023 3rd International Symposium on Computer Technology and Information Science (ISCTIS)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127916783","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Dense false-target jamming Suppression Method Based on the Scene Cognition Network","authors":"Wei Hong, Qianyi Tian, Chun-xu Jiang","doi":"10.1109/ISCTIS58954.2023.10213065","DOIUrl":"https://doi.org/10.1109/ISCTIS58954.2023.10213065","url":null,"abstract":"Recognition and suppression of dense false-target jamming has been a hotspot in research of radar anti-jamming technique. This essay proposes a method to suppress the dense false-target jamming according to scene cognition. Model of scene cognition network is obtained off-line through training real and imaginary data after pulse compression of echo data. Real targets and jamming will be recognized intelligently with the full-trained scene cognition network. Dense false-target jamming is suppressed with multistage spectral peak detection and spectrum removing in the recognized jamming area. Finally, jamming suppression results are imported to the scene cognition network again to evaluate the effect of jamming suppression. The results of experiments of simulated data and actual data prove that proposed method can effectively suppress the dense false-target jamming and detect the real targets. Our method is relatively valuable in engineering application.","PeriodicalId":334790,"journal":{"name":"2023 3rd International Symposium on Computer Technology and Information Science (ISCTIS)","volume":"85 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132076370","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An Improved SSD Network-Based Target Detection Method for Penguin Populations","authors":"Yixin Fan, Chunyan Ma","doi":"10.1109/ISCTIS58954.2023.10213078","DOIUrl":"https://doi.org/10.1109/ISCTIS58954.2023.10213078","url":null,"abstract":"Global warming is seriously threatening the natural environment of the Antarctic region, and penguins, as the iconic Antarctic marine life, are very responsive to climate change. Therefore, the changing pattern of the penguin population can provide a scientific basis for predicting the response of the Antarctic ecosystem to climate change. In this paper, we propose an improved penguin population target detection algorithm based on the SSD (Single Shot MultiBox Detector) network to address the practical problems of the harsh Antarctic climate. The algorithm adds the ECANet attention module to enhance the detection capability of the model. Experiments show that the detection capability is improved by 13.5%, and the ap value is improved by 26% compared with the traditional SSD algorithm.","PeriodicalId":334790,"journal":{"name":"2023 3rd International Symposium on Computer Technology and Information Science (ISCTIS)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131654475","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Digital construction of data traceability based on dynamic recognition algorithm","authors":"Peidong He, XiaoJun Li, Li Xiao, YangFan Zhang, WenQi Shen, ShuYu Deng","doi":"10.1109/ISCTIS58954.2023.10213140","DOIUrl":"https://doi.org/10.1109/ISCTIS58954.2023.10213140","url":null,"abstract":"Traceability result confirmation is an important link to ensure the accuracy and reliability of the standard measurement value. At present, the corresponding certificates/reports issued by superior technical institutions are not digital information, which requires manual comparison of relevant data, resulting in low efficiency and high error rate. Combined with deep learning theory, this paper proposes a dynamic recognition algorithm which can be applied to image, text and other information carriers to realize intelligent recognition of images, symbols and digital content. Based on this algorithm, a digital system for quantitative traceability is developed, and a set of intelligent data extraction, error correction and structured platform is built to improve the efficiency and accuracy of metrological verification.","PeriodicalId":334790,"journal":{"name":"2023 3rd International Symposium on Computer Technology and Information Science (ISCTIS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131766332","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Research on Service Function Chain Orchestration for Intractable Scenarios","authors":"Xiangyu Yin, X. Chen, Tao Wu, Liang Li, Qing Li, Mingfeng Zhao, Bin Xiang, Lingyun Liu","doi":"10.1109/ISCTIS58954.2023.10213001","DOIUrl":"https://doi.org/10.1109/ISCTIS58954.2023.10213001","url":null,"abstract":"Service Function Chaining (SFC) has become a typical network service delivery approach in software-defined networking/ network function virtualization (SDN/NFV) network architectures. The complex network topology and dynamic nature of network resources in NFV networks have made the orchestration of service function chains a research hotspot and challenge. Existing studies consider mapping various types of virtual network function (VNF) nodes in the virtual service function chain to the underlying physical network, assuming that physical nodes can accommodate all types of VNF services, which imposes high demands on the performance of the physical network. In response to the aforementioned issues, this paper proposes a novel network architecture for scenarios where the physical network is intractable. To validate the effectiveness of the proposed network architecture, mathematical programming algorithms and genetic algorithms are employed to conduct experiments on the orchestration of SFC.","PeriodicalId":334790,"journal":{"name":"2023 3rd International Symposium on Computer Technology and Information Science (ISCTIS)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132619528","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A study of EEG classification based on attention mechanism and EEGNet Motor Imagination","authors":"Mengyuan Liu, Fucheng Cao, Xinzui Wang, Yanli Yang","doi":"10.1109/ISCTIS58954.2023.10213202","DOIUrl":"https://doi.org/10.1109/ISCTIS58954.2023.10213202","url":null,"abstract":"In recent years, with the development of deep learning (DL), deep learning has made great contributions to the improvement of motor imagination brain-computer interface (MI-BCI). At the same time, the proposed attention mechanism also provides a new method for the feature extraction and classification of EEG. CNN has limitations in perceiting global dependencies, and the Attention mechanism can make up for this limitation. Based on this, a method combining convolution and Attention is proposed. The Attention mechanism was added based on the EEGNet model, and the BCI competition-IV 2a data set was used for the experiment. The experimental results show that the accuracy of EEGNet in the four categories of motor imagination can reach 54.50%, and the accuracy can reach 70.49% by adding the multi-head attention mechanism to EEGNet. Due to the characteristics of parallel computing of Attention mechanism, on the server with NVIDIA A100-PCIE 40GB graphics card and CUDA version 11.6, the training time of each epoch is reduced by 3-5s compared with the original time. In recent years, with the development of deep learning (DL), Deep learning has made great contributions to the improvement of the motor imagination brain computer interface (MI-BCI). At the same time, the proposed attention mechanism also provides a new method for the feature extraction and classification of EEG. CNN has limitations in perceiting global dependencies, and the Attention mechanism can make up for this limitation. Based on this, a method combining convolution and Attention is proposed. The Attention mechanism was added based on the EEGNet model, and the BCI competition-IV 2a data set was used for the experiment. The experimental results show that the accuracy of EEGNet in the four categories of motor imagination can reach 54.50%, and the accuracy can reach 70.49% by adding the multi-head attention mechanism to EEGNet. Due to the parallel computing features of the Attention mechanism, on the NVIDIA A100-PCIE 40GB graphics card and CUDA 11.6 server, the training time of each epoch is reduced by 3-5s compared to the original time, reducing the training time.","PeriodicalId":334790,"journal":{"name":"2023 3rd International Symposium on Computer Technology and Information Science (ISCTIS)","volume":"16 2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132723049","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}