Pub Date : 2022-09-23DOI: 10.1109/ICACTE55855.2022.9943629
Jhanika F. Fanlo, Joyce Anne H. Lanceta, Janea Patrizia R. Pascua, Alfonso Louis Philip R. Salas, John Patrick C. To, Ramon L. Rodriguez
The world was put in disarray when the novel coronavirus first began. Furthermore, when the World Health Organization (WHO) declared the novel coronavirus outbreak a public health emergency of international concern (PHEIC), people prepared safety protocols to minimize the effect of the virus. One of these is the implementation of e-learning in countries, including the Philippines. As this contactless learning began, students’ motivation decreased due to a lack of private space/classroom and face-to-face communication with their teachers. Learners’ motivation is as crucial as this influences their pace to learn. The researchers developed a tool to help students with their studies and motivate them. LINYA is a web-based text annotation tool in machine learning. The tool was developed using an NLP method in machine learning. The researchers used automated Agile testing with four phases in testing the web tool. It began with component testing and progressed to integration, system, and acceptance testing. Based on the results from simulated data, the tests showed favorable results, with mean scores ranging from 3.8 to 4.6, for all areas of a usability test. It further shows that the developed system is ready for implementation.
{"title":"LINYA: Name Entity Recognition Web-based Text Annotation","authors":"Jhanika F. Fanlo, Joyce Anne H. Lanceta, Janea Patrizia R. Pascua, Alfonso Louis Philip R. Salas, John Patrick C. To, Ramon L. Rodriguez","doi":"10.1109/ICACTE55855.2022.9943629","DOIUrl":"https://doi.org/10.1109/ICACTE55855.2022.9943629","url":null,"abstract":"The world was put in disarray when the novel coronavirus first began. Furthermore, when the World Health Organization (WHO) declared the novel coronavirus outbreak a public health emergency of international concern (PHEIC), people prepared safety protocols to minimize the effect of the virus. One of these is the implementation of e-learning in countries, including the Philippines. As this contactless learning began, students’ motivation decreased due to a lack of private space/classroom and face-to-face communication with their teachers. Learners’ motivation is as crucial as this influences their pace to learn. The researchers developed a tool to help students with their studies and motivate them. LINYA is a web-based text annotation tool in machine learning. The tool was developed using an NLP method in machine learning. The researchers used automated Agile testing with four phases in testing the web tool. It began with component testing and progressed to integration, system, and acceptance testing. Based on the results from simulated data, the tests showed favorable results, with mean scores ranging from 3.8 to 4.6, for all areas of a usability test. It further shows that the developed system is ready for implementation.","PeriodicalId":165068,"journal":{"name":"2022 15th International Conference on Advanced Computer Theory and Engineering (ICACTE)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127918616","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}
Pub Date : 2022-09-23DOI: 10.1109/ICACTE55855.2022.9943732
Kanyangzi Xu, Zhe Wu
Theory of games is an important discipline of applied mathematics, which has found numerous applications in psychology, ecology, and social science. While a well-established framework has been established for classical game theory, extending, and generalizing it towards the quantum domain still has some open questions. In this project, we focus on the prisoner’s dilemma-a standard example of a game analyzed in game theory-in a quantum world. Here, we start with the quantum version of prisoner’s dilemma, explain the quantum circuit implementation for describing the behavior of quantum prisoners, and investigate how quantum entanglement can change their strategies. By using the IBM quantum computer, we experimentally study the quantum prisoner dilemma, and we conclude that the best quantum strategies will break the prisoner’s dilemma.
{"title":"Experimental Implementation of Quantum Prisoner Dilemma on IBM Quantum Computers","authors":"Kanyangzi Xu, Zhe Wu","doi":"10.1109/ICACTE55855.2022.9943732","DOIUrl":"https://doi.org/10.1109/ICACTE55855.2022.9943732","url":null,"abstract":"Theory of games is an important discipline of applied mathematics, which has found numerous applications in psychology, ecology, and social science. While a well-established framework has been established for classical game theory, extending, and generalizing it towards the quantum domain still has some open questions. In this project, we focus on the prisoner’s dilemma-a standard example of a game analyzed in game theory-in a quantum world. Here, we start with the quantum version of prisoner’s dilemma, explain the quantum circuit implementation for describing the behavior of quantum prisoners, and investigate how quantum entanglement can change their strategies. By using the IBM quantum computer, we experimentally study the quantum prisoner dilemma, and we conclude that the best quantum strategies will break the prisoner’s dilemma.","PeriodicalId":165068,"journal":{"name":"2022 15th International Conference on Advanced Computer Theory and Engineering (ICACTE)","volume":"140 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128617777","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}
Pub Date : 2022-09-23DOI: 10.1109/ICACTE55855.2022.9943768
Ying Yuan, Runyu Liu, F. Deng
With the end of the Second National Pollution Source Census (SNPSC), a large amount of census results data has been obtained. As basic data, these data are rich in value and can provide important support for various businesses related to ecology and environment. However, in order to obtain the value contained in the pollution source census results data, it is necessary to mine the results data, which is difficult to be done by general business personnel in the field of ecology and environment, and this requires professionals to use professional tools to complete this work. To address this situation, this paper develops an analysis and sharing system for the results data of the SNPSC based on the collation of the results data of the SNPSC. This system uses Apache Kylin and Web GIS to build a data analysis toolset, which achieves the analysis of the SNPSC results data from the dimension of business data analysis and the dimension of spatial distribution of pollution sources. and achieves the sharing of the analysis results. The system is designed with an easy-to-use interface, so that even non-professionals can realize the data analysis of the SNPSC results data through this system and make the second NPSC results data more valuable.
{"title":"Analysis and Sharing System of the Second Pollution Source Census Results Data Based on Apache Kylin and WebGIS","authors":"Ying Yuan, Runyu Liu, F. Deng","doi":"10.1109/ICACTE55855.2022.9943768","DOIUrl":"https://doi.org/10.1109/ICACTE55855.2022.9943768","url":null,"abstract":"With the end of the Second National Pollution Source Census (SNPSC), a large amount of census results data has been obtained. As basic data, these data are rich in value and can provide important support for various businesses related to ecology and environment. However, in order to obtain the value contained in the pollution source census results data, it is necessary to mine the results data, which is difficult to be done by general business personnel in the field of ecology and environment, and this requires professionals to use professional tools to complete this work. To address this situation, this paper develops an analysis and sharing system for the results data of the SNPSC based on the collation of the results data of the SNPSC. This system uses Apache Kylin and Web GIS to build a data analysis toolset, which achieves the analysis of the SNPSC results data from the dimension of business data analysis and the dimension of spatial distribution of pollution sources. and achieves the sharing of the analysis results. The system is designed with an easy-to-use interface, so that even non-professionals can realize the data analysis of the SNPSC results data through this system and make the second NPSC results data more valuable.","PeriodicalId":165068,"journal":{"name":"2022 15th International Conference on Advanced Computer Theory and Engineering (ICACTE)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114442008","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}
Pub Date : 2022-09-23DOI: 10.1109/ICACTE55855.2022.9943760
Changlong Wang, Siyun Bi, Rong Zhang, Qibin Fu, Tingting Gan
The construction and application of Knowledge Graph require effective reasoning support. However, the standard reasoning engines can not effectively deal with large-scale Knowledge Graphs because they load and compute Knowledge Graphs as a whole. This paper proposes a modular reasoning approach to Knowledge Graph. Firstly, the facts in the Knowledge Graph are partitioned into modules according to the predicate type and entity. Then the concepts and attributes involved in the fact module are used as seed signatures to extract the ontology module from the schema. During the reasoning procedure, the reasoning engine partially loads fact modules and the related ontology modules. Experiments show that the proposed approach can deal with large-scale Knowledge Graphs in a modular way with less time and memory.
{"title":"A Modular Reasoning Approach to Knowledge Graph","authors":"Changlong Wang, Siyun Bi, Rong Zhang, Qibin Fu, Tingting Gan","doi":"10.1109/ICACTE55855.2022.9943760","DOIUrl":"https://doi.org/10.1109/ICACTE55855.2022.9943760","url":null,"abstract":"The construction and application of Knowledge Graph require effective reasoning support. However, the standard reasoning engines can not effectively deal with large-scale Knowledge Graphs because they load and compute Knowledge Graphs as a whole. This paper proposes a modular reasoning approach to Knowledge Graph. Firstly, the facts in the Knowledge Graph are partitioned into modules according to the predicate type and entity. Then the concepts and attributes involved in the fact module are used as seed signatures to extract the ontology module from the schema. During the reasoning procedure, the reasoning engine partially loads fact modules and the related ontology modules. Experiments show that the proposed approach can deal with large-scale Knowledge Graphs in a modular way with less time and memory.","PeriodicalId":165068,"journal":{"name":"2022 15th International Conference on Advanced Computer Theory and Engineering (ICACTE)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122792601","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}
Pub Date : 2022-09-23DOI: 10.1109/ICACTE55855.2022.9943674
Miaomiao Li, Jie Yu, Shasha Li, Jun Ma, Huijun Liu
Named entity recognition is a key task in the field of natural language processing, which plays a key role in many downstream tasks. Adversarial examples attack based on hard label black box is to generate adversarial examples which make the model classification wrong under the condition that only the decision results of the model are obtained. However, at present, there is little research on adversarial examples attack in hard-label black box setting for named entity recognition task. Influenced by adversarial examples attacks in hard-label black box settings in text classification task, we apply genetic algorithm to adversarial examples attacks in named entity recognition task. In this paper, we first randomly generate the initial adversarial examples, and shorten the search space to a certain extent, and then use genetic algorithm to continuously optimize the examples, and finally generate high quality adversarial examples. Experiments and analysis show that the adversarial examples generated in the hard label black box setting can effectively reduce the accuracy of the model.
{"title":"Textual Adversarial Attacks on Named Entity Recognition in a Hard Label Black Box Setting","authors":"Miaomiao Li, Jie Yu, Shasha Li, Jun Ma, Huijun Liu","doi":"10.1109/ICACTE55855.2022.9943674","DOIUrl":"https://doi.org/10.1109/ICACTE55855.2022.9943674","url":null,"abstract":"Named entity recognition is a key task in the field of natural language processing, which plays a key role in many downstream tasks. Adversarial examples attack based on hard label black box is to generate adversarial examples which make the model classification wrong under the condition that only the decision results of the model are obtained. However, at present, there is little research on adversarial examples attack in hard-label black box setting for named entity recognition task. Influenced by adversarial examples attacks in hard-label black box settings in text classification task, we apply genetic algorithm to adversarial examples attacks in named entity recognition task. In this paper, we first randomly generate the initial adversarial examples, and shorten the search space to a certain extent, and then use genetic algorithm to continuously optimize the examples, and finally generate high quality adversarial examples. Experiments and analysis show that the adversarial examples generated in the hard label black box setting can effectively reduce the accuracy of the model.","PeriodicalId":165068,"journal":{"name":"2022 15th International Conference on Advanced Computer Theory and Engineering (ICACTE)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129232370","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}
Pub Date : 2022-09-23DOI: 10.1109/ICACTE55855.2022.9943666
Jingyang Wang, Tao Zou, Zhijia Yang, Hongrui Wang
This paper illustrates the application of an autonomous decentralized system introducing edge nodes in a smart factory. The nodes include atomic nodes, the most fundamental autonomic units constituting an autonomous decentralized system that are capable of independently extracting information from the data field and processing that information internally and simultaneously transmitting the processing results and other internal information to the data field proactively in a broadcast mode, where the transmitted information circulates in the data field. The group-based data field enables the sharing of information by atomic nodes. Grouped management nodes are the managers of the whole group, and edge nodes are the terminal intelligent control units of an intelligent autonomous decentralized system. Improvements in system decentralization, fault tolerance characteristics and flexibility of the system can lead to homogenization, autonomous control and autonomous coordination in complex manufacturing environments, as well as excellent online fault tolerance, online expansion and online maintenance of the system. In addition, it enhances the ever-changing and evolving control requirements by deconstructing the complexity in the system.
{"title":"Smart Plant Application of Autonomous Decentralized Systems with the Introduction of Edge Nodes","authors":"Jingyang Wang, Tao Zou, Zhijia Yang, Hongrui Wang","doi":"10.1109/ICACTE55855.2022.9943666","DOIUrl":"https://doi.org/10.1109/ICACTE55855.2022.9943666","url":null,"abstract":"This paper illustrates the application of an autonomous decentralized system introducing edge nodes in a smart factory. The nodes include atomic nodes, the most fundamental autonomic units constituting an autonomous decentralized system that are capable of independently extracting information from the data field and processing that information internally and simultaneously transmitting the processing results and other internal information to the data field proactively in a broadcast mode, where the transmitted information circulates in the data field. The group-based data field enables the sharing of information by atomic nodes. Grouped management nodes are the managers of the whole group, and edge nodes are the terminal intelligent control units of an intelligent autonomous decentralized system. Improvements in system decentralization, fault tolerance characteristics and flexibility of the system can lead to homogenization, autonomous control and autonomous coordination in complex manufacturing environments, as well as excellent online fault tolerance, online expansion and online maintenance of the system. In addition, it enhances the ever-changing and evolving control requirements by deconstructing the complexity in the system.","PeriodicalId":165068,"journal":{"name":"2022 15th International Conference on Advanced Computer Theory and Engineering (ICACTE)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133093795","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}
Pub Date : 2022-09-23DOI: 10.1109/ICACTE55855.2022.9943626
Donghui Shi, Wenrui Zhu, Rui Cheng, Yuchen Yang
The existing research shows that falls account for a significant proportion of safety accidents. At the same time, as many countries enter an aging society, falls have increasingly become a non-negligible safety issue affecting the lives and health of the elderly. To address the current problems of human fall detection, we propose to extract a human skeleton model based on YoloX-s in combination with Lightweight OpenPose. This model can identify human fall by the difference values of angle change’s rate between the key points of the neck and knees. The results demonstrate that the accuracy rate for fall detection is 97.92% and that for normal behavior detection is 96.46%. The computing speed of the method satisfies the need for real-time processing with satisfactory robustness.
{"title":"Human Fall Detection Algorithm Based on YoloX-s and Lightweight OpenPose","authors":"Donghui Shi, Wenrui Zhu, Rui Cheng, Yuchen Yang","doi":"10.1109/ICACTE55855.2022.9943626","DOIUrl":"https://doi.org/10.1109/ICACTE55855.2022.9943626","url":null,"abstract":"The existing research shows that falls account for a significant proportion of safety accidents. At the same time, as many countries enter an aging society, falls have increasingly become a non-negligible safety issue affecting the lives and health of the elderly. To address the current problems of human fall detection, we propose to extract a human skeleton model based on YoloX-s in combination with Lightweight OpenPose. This model can identify human fall by the difference values of angle change’s rate between the key points of the neck and knees. The results demonstrate that the accuracy rate for fall detection is 97.92% and that for normal behavior detection is 96.46%. The computing speed of the method satisfies the need for real-time processing with satisfactory robustness.","PeriodicalId":165068,"journal":{"name":"2022 15th International Conference on Advanced Computer Theory and Engineering (ICACTE)","volume":"41 13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131130661","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}
Pub Date : 2022-09-23DOI: 10.1109/ICACTE55855.2022.9943636
Wenzhu Wang, Xiaodong Liu, Jie Yu, Jianfeng Li, Z. Mao, Zhuoheng Li, Chenguang Ding, Chao Zhang
The RISC-V instruction set architecture (ISA) stimulates the expeditious development of novel hardware platforms. Consequently, the need for an efficient and easy-to-use operating system on RISC-V architecture emerges. However, new challenges such as system building, hardware adaptation, and application ecosystem should be addressed as the hardware podium develops. This article explores the design and building of openKylin, an open-source operating system, on the RISC-V hardware platform to address these issues, including kernel optimization, UKUI (Ultimate Kylin User Interface) package compilation, and application compatibility. The test results show that the x86 benchmark can run in the openKylin operating system correctly and efficiently on the RISC-V platform.
RISC-V指令集架构(ISA)促进了新型硬件平台的快速发展。因此，对RISC-V架构上高效且易于使用的操作系统的需求应运而生。然而，随着硬件平台的发展，系统构建、硬件适应和应用程序生态系统等新挑战也应该得到解决。本文探讨了openKylin(一个开源操作系统)在RISC-V硬件平台上的设计和构建，以解决这些问题，包括内核优化、UKUI (Ultimate Kylin User Interface)包编译和应用程序兼容性。测试结果表明，x86基准测试可以在RISC-V平台上正确、高效地运行在open麒麟操作系统上。
{"title":"The Design and Building of openKylin on RISC-V Architecture","authors":"Wenzhu Wang, Xiaodong Liu, Jie Yu, Jianfeng Li, Z. Mao, Zhuoheng Li, Chenguang Ding, Chao Zhang","doi":"10.1109/ICACTE55855.2022.9943636","DOIUrl":"https://doi.org/10.1109/ICACTE55855.2022.9943636","url":null,"abstract":"The RISC-V instruction set architecture (ISA) stimulates the expeditious development of novel hardware platforms. Consequently, the need for an efficient and easy-to-use operating system on RISC-V architecture emerges. However, new challenges such as system building, hardware adaptation, and application ecosystem should be addressed as the hardware podium develops. This article explores the design and building of openKylin, an open-source operating system, on the RISC-V hardware platform to address these issues, including kernel optimization, UKUI (Ultimate Kylin User Interface) package compilation, and application compatibility. The test results show that the x86 benchmark can run in the openKylin operating system correctly and efficiently on the RISC-V platform.","PeriodicalId":165068,"journal":{"name":"2022 15th International Conference on Advanced Computer Theory and Engineering (ICACTE)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133991892","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}
Pub Date : 2022-09-23DOI: 10.1109/ICACTE55855.2022.9943557
Jun-Hong Huang, Tao Liu, Ya Wang, Zhibo Chen
Image dehazing is a technique used for repairing blurry images which can effectively reduce the impact of haze on visual tasks. Most of the existing dehazing methods rely on atmospheric models or perform supervised learning based on paired images to obtain haze-free images. However, problems such as relying on prior knowledge of a specific scene and difficulty in collecting paired hazy and haze-free images have hindered the development of image dehazing techniques. In response to the above problems, we are inspired by the CycleGAN algorithm and propose the DAM-CCGAN algorithm, which uses an unsupervised method to dehaze unpaired images. For the blur and color distortion problems which can occur in image dehazing, the DAM-CCGAN algorithm adds a skip connection method and an attention mechanism module (DAM) to the generator. To preserve more image information, we add a detailed perception loss function. Meanwhile, to reduce the complexity of the algorithm, we improve the convolution group structure in the generator. Experiments show that our model achieves a good dehazing effect on both indoor and outdoor hazy images.
{"title":"Unsupervised Image Dehazing Based on Improved Generative Adversarial Networks","authors":"Jun-Hong Huang, Tao Liu, Ya Wang, Zhibo Chen","doi":"10.1109/ICACTE55855.2022.9943557","DOIUrl":"https://doi.org/10.1109/ICACTE55855.2022.9943557","url":null,"abstract":"Image dehazing is a technique used for repairing blurry images which can effectively reduce the impact of haze on visual tasks. Most of the existing dehazing methods rely on atmospheric models or perform supervised learning based on paired images to obtain haze-free images. However, problems such as relying on prior knowledge of a specific scene and difficulty in collecting paired hazy and haze-free images have hindered the development of image dehazing techniques. In response to the above problems, we are inspired by the CycleGAN algorithm and propose the DAM-CCGAN algorithm, which uses an unsupervised method to dehaze unpaired images. For the blur and color distortion problems which can occur in image dehazing, the DAM-CCGAN algorithm adds a skip connection method and an attention mechanism module (DAM) to the generator. To preserve more image information, we add a detailed perception loss function. Meanwhile, to reduce the complexity of the algorithm, we improve the convolution group structure in the generator. Experiments show that our model achieves a good dehazing effect on both indoor and outdoor hazy images.","PeriodicalId":165068,"journal":{"name":"2022 15th International Conference on Advanced Computer Theory and Engineering (ICACTE)","volume":"140 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124441697","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}
Pub Date : 2022-09-23DOI: 10.1109/ICACTE55855.2022.9943600
Minghui Fan, Lei Xiao, Xiang-zhen He, Yawei Chen
The classification and recycling of garbage can greatly improve the utilization of garbage resources. This paper proposes a new convolutional neural network that fuses a multi-branch Xception network with an attention mechanism module. The effective feature information is emphasized and the invalid information is suppressed to overcome the problem caused by the small data set. To verify the usefulness of this network structure in the field of garbage images, this paper uses a widely used data set in the field of garbage image classification. For any network without pre-trained weights, the network proposed in this paper outperforms all other methods by 94.4%.
{"title":"Trash Classification Network Based on Attention Mechanism","authors":"Minghui Fan, Lei Xiao, Xiang-zhen He, Yawei Chen","doi":"10.1109/ICACTE55855.2022.9943600","DOIUrl":"https://doi.org/10.1109/ICACTE55855.2022.9943600","url":null,"abstract":"The classification and recycling of garbage can greatly improve the utilization of garbage resources. This paper proposes a new convolutional neural network that fuses a multi-branch Xception network with an attention mechanism module. The effective feature information is emphasized and the invalid information is suppressed to overcome the problem caused by the small data set. To verify the usefulness of this network structure in the field of garbage images, this paper uses a widely used data set in the field of garbage image classification. For any network without pre-trained weights, the network proposed in this paper outperforms all other methods by 94.4%.","PeriodicalId":165068,"journal":{"name":"2022 15th International Conference on Advanced Computer Theory and Engineering (ICACTE)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128576021","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：