Pub Date : 2021-11-01DOI: 10.1109/ITME53901.2021.00117
Sen Wang, Fu Xie, Yiming Jia
With the continuous development of China's economic structure, vocational education has received unprecedented attention. More and more students begin to choose secondary vocational education. What teaching methods can be used to teach secondary vocational students more pertinently, which is a problem that needs to be discussed. In order to explore the influence of collaborative learning in the teaching process of secondary vocational schools, this article takes the course of “Computer Network Technology” as an example to explore the application of collaborative learning in the course of “Computer Network Technology” in secondary. The results show that compared with the traditional teaching model, collaborative learning can pay more attention to students' problem-solving ability and team cooperation ability. Through collaborative learning, students' autonomy is improved, and the sense of collaboration is enhanced. Therefore, it is necessary to introduce collaborative learning in the teaching process of secondary vocational schools to stimulate learning motivation and improve students' literacy.
{"title":"Application of Collaborative Learning in the Course of “Computer Network Technology” in Secondary Vocational School","authors":"Sen Wang, Fu Xie, Yiming Jia","doi":"10.1109/ITME53901.2021.00117","DOIUrl":"https://doi.org/10.1109/ITME53901.2021.00117","url":null,"abstract":"With the continuous development of China's economic structure, vocational education has received unprecedented attention. More and more students begin to choose secondary vocational education. What teaching methods can be used to teach secondary vocational students more pertinently, which is a problem that needs to be discussed. In order to explore the influence of collaborative learning in the teaching process of secondary vocational schools, this article takes the course of “Computer Network Technology” as an example to explore the application of collaborative learning in the course of “Computer Network Technology” in secondary. The results show that compared with the traditional teaching model, collaborative learning can pay more attention to students' problem-solving ability and team cooperation ability. Through collaborative learning, students' autonomy is improved, and the sense of collaboration is enhanced. Therefore, it is necessary to introduce collaborative learning in the teaching process of secondary vocational schools to stimulate learning motivation and improve students' literacy.","PeriodicalId":6774,"journal":{"name":"2021 11th International Conference on Information Technology in Medicine and Education (ITME)","volume":"20 1","pages":"552-555"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84645866","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 : 2021-11-01DOI: 10.1109/ITME53901.2021.00114
Ruihu Chen, Fengting Jia, Yafei Sun, R. Luo
With the rapid development of science and technology, the drawbacks of talent training based on traditional industries gradually emerge. Due to the slow development of the profession, it often leads to low professional identity, which brings new challenges to the cultivation of talents in universities. The new generation of information technology brings new vitality to the development of traditional industries, and the traditional production modes are transformed to intelligent production to improve the market competitiveness of products and industries. In this paper, the traditional leather industry is the object of study, and the current situation of leather talents training in colleges and universities in recent years is investigated. The impact of information technology on the leather industry and the development of the direction of leather professional higher education are analyzed. Finally, the reform strategy of leather professional talent training under the opportunity of new generation of information technology is proposed.
{"title":"Research on the necessity of accelerating the embedding of information technology in the education of traditional leather manufacturing industry","authors":"Ruihu Chen, Fengting Jia, Yafei Sun, R. Luo","doi":"10.1109/ITME53901.2021.00114","DOIUrl":"https://doi.org/10.1109/ITME53901.2021.00114","url":null,"abstract":"With the rapid development of science and technology, the drawbacks of talent training based on traditional industries gradually emerge. Due to the slow development of the profession, it often leads to low professional identity, which brings new challenges to the cultivation of talents in universities. The new generation of information technology brings new vitality to the development of traditional industries, and the traditional production modes are transformed to intelligent production to improve the market competitiveness of products and industries. In this paper, the traditional leather industry is the object of study, and the current situation of leather talents training in colleges and universities in recent years is investigated. The impact of information technology on the leather industry and the development of the direction of leather professional higher education are analyzed. Finally, the reform strategy of leather professional talent training under the opportunity of new generation of information technology is proposed.","PeriodicalId":6774,"journal":{"name":"2021 11th International Conference on Information Technology in Medicine and Education (ITME)","volume":"53 1","pages":"537-540"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89475831","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 : 2021-11-01DOI: 10.1109/ITME53901.2021.00103
Xueyu Che, Xiaomei Yu, Wenqian Sun, Shuang Ma, Xiangwei Zheng
Secondary vocational education is an important component in modern education. Recently, the research on teaching methods in secondary vocational education has attracted great attention and achieved remarkable results. However, there are still some problems such as the lack of initiative in students. In this paper, the visualization software of CiteSpace is used to analyze the teaching methods in secondary vocational school in China in recent ten years. The analysis results show that the task-driven method and micro-class teaching are two hot teaching methods in secondary vocational school. Therefore, these two methods are applied to the actual teaching of computer basic course in secondary vocational school. The practice results show that the hybrid learning method of task-driven method and micro-class teaching method significantly improves the teaching effect and raises students' learning enthusiasm.
{"title":"Research on Teaching Methods in Secondary Vocational School based on CiteSpace","authors":"Xueyu Che, Xiaomei Yu, Wenqian Sun, Shuang Ma, Xiangwei Zheng","doi":"10.1109/ITME53901.2021.00103","DOIUrl":"https://doi.org/10.1109/ITME53901.2021.00103","url":null,"abstract":"Secondary vocational education is an important component in modern education. Recently, the research on teaching methods in secondary vocational education has attracted great attention and achieved remarkable results. However, there are still some problems such as the lack of initiative in students. In this paper, the visualization software of CiteSpace is used to analyze the teaching methods in secondary vocational school in China in recent ten years. The analysis results show that the task-driven method and micro-class teaching are two hot teaching methods in secondary vocational school. Therefore, these two methods are applied to the actual teaching of computer basic course in secondary vocational school. The practice results show that the hybrid learning method of task-driven method and micro-class teaching method significantly improves the teaching effect and raises students' learning enthusiasm.","PeriodicalId":6774,"journal":{"name":"2021 11th International Conference on Information Technology in Medicine and Education (ITME)","volume":"45 1","pages":"485-489"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88628688","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 : 2021-11-01DOI: 10.1109/ITME53901.2021.00104
Wenxiang Fu, Xiaomei Yu, Zhaokun Gong, Xueyu Che, Qian Mao, Yan Li
Nowadays, the computer network course becomes one of the core compulsory courses for computer related majors. This paper analyzes the problems existing in computer network course teaching. And then, aiming at the problems of low utilization of curriculum resources and poor classroom learning effect, this paper presents the method of improving teaching mode based on personalized micro-videos from the perspective of educational equity. Specially, a personalized micro-videos teaching model based on autonomous learning and collaborative learning is constructed. Taking “home network construction” in the course of computer network in vocational education as an example, the course is designed and the teaching practice is carried out. The results of teaching practice have proved that the novel teaching model improves the teaching effect and raises the learning enthusiasm of students greatly.
{"title":"The Applications of Personalized Micro-videos in Computer Network Course Teaching","authors":"Wenxiang Fu, Xiaomei Yu, Zhaokun Gong, Xueyu Che, Qian Mao, Yan Li","doi":"10.1109/ITME53901.2021.00104","DOIUrl":"https://doi.org/10.1109/ITME53901.2021.00104","url":null,"abstract":"Nowadays, the computer network course becomes one of the core compulsory courses for computer related majors. This paper analyzes the problems existing in computer network course teaching. And then, aiming at the problems of low utilization of curriculum resources and poor classroom learning effect, this paper presents the method of improving teaching mode based on personalized micro-videos from the perspective of educational equity. Specially, a personalized micro-videos teaching model based on autonomous learning and collaborative learning is constructed. Taking “home network construction” in the course of computer network in vocational education as an example, the course is designed and the teaching practice is carried out. The results of teaching practice have proved that the novel teaching model improves the teaching effect and raises the learning enthusiasm of students greatly.","PeriodicalId":6774,"journal":{"name":"2021 11th International Conference on Information Technology in Medicine and Education (ITME)","volume":"37 1","pages":"490-494"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88934453","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 : 2021-11-01DOI: 10.1109/ITME53901.2021.00083
Wu Qing-song, Su Song-zhi, Wu Chang-wen
TCM(Traditional Chinese Medicine) physical classi-fication system based on “Five Elements” is the foundation and core material of physical study, and it is a classification system appropriate for a group's physical characteristics. Obtaining appropriate physical classification can aid in disease diagnosis efficiency. We believe that the original problem cannot be simply described as five separate tasks, like independent score inference using five different models for each element category. So we propose an approach based on the Insightface algorithm and Multi-Task Gaussian Process Regression (MTGPR) model to classify faces. MTGPR is a model that attempts to learn inter-task dependencies based solely on the task identities and the observed data for each task. It uses a parameterized covariance function over the input features x to develop a “free-form” task-similarity matrix. In MTGPR model, this is achieved by having a common covariance function over the features $x$ of the input observations. The experimental results show that our proposed method has improved results compared to the traditional Resnet-based classification method.
{"title":"Face Classification Based on Multi-task Gaussian Process Regression and Chinese Medicine Five Element System","authors":"Wu Qing-song, Su Song-zhi, Wu Chang-wen","doi":"10.1109/ITME53901.2021.00083","DOIUrl":"https://doi.org/10.1109/ITME53901.2021.00083","url":null,"abstract":"TCM(Traditional Chinese Medicine) physical classi-fication system based on “Five Elements” is the foundation and core material of physical study, and it is a classification system appropriate for a group's physical characteristics. Obtaining appropriate physical classification can aid in disease diagnosis efficiency. We believe that the original problem cannot be simply described as five separate tasks, like independent score inference using five different models for each element category. So we propose an approach based on the Insightface algorithm and Multi-Task Gaussian Process Regression (MTGPR) model to classify faces. MTGPR is a model that attempts to learn inter-task dependencies based solely on the task identities and the observed data for each task. It uses a parameterized covariance function over the input features x to develop a “free-form” task-similarity matrix. In MTGPR model, this is achieved by having a common covariance function over the features $x$ of the input observations. The experimental results show that our proposed method has improved results compared to the traditional Resnet-based classification method.","PeriodicalId":6774,"journal":{"name":"2021 11th International Conference on Information Technology in Medicine and Education (ITME)","volume":"41 1","pages":"385-389"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79276414","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 : 2021-11-01DOI: 10.1109/ITME53901.2021.00118
Minna Xia, Xinxin Peng, Yang Liu, Sihuang Liu
During the epidemic period, in response to the national education policy of “continuous suspension of classes”, the teaching work of universities was carried out online with the help of various online teaching platforms such as Superplatform, rain classroom, smart tree, university MODC, school-based platform smart classroom, as well as well as software with the function of teaching live broadcast. In order to improve the teaching effect, build students ‘cognitive system, cultivate students’ ability of independent learning and collaborative learning, and help students to achieve the knowledge objectives, skills objectives and quality objectives required by the curriculum. The author from the current situation of the post-epidemic era, combined with hunan automobile engineering vocational college online mixed teaching reality, with research results at home and abroad as the theoretical support, develop granular micro class video, system design orderly teaching link, in the actual practice of offline teaching mode, and reversed transmission teachers, teaching materials, teaching method reform.
{"title":"Research on Mixed Teaching Mode under the Background of the reform of teachers, teaching materials and teaching methods","authors":"Minna Xia, Xinxin Peng, Yang Liu, Sihuang Liu","doi":"10.1109/ITME53901.2021.00118","DOIUrl":"https://doi.org/10.1109/ITME53901.2021.00118","url":null,"abstract":"During the epidemic period, in response to the national education policy of “continuous suspension of classes”, the teaching work of universities was carried out online with the help of various online teaching platforms such as Superplatform, rain classroom, smart tree, university MODC, school-based platform smart classroom, as well as well as software with the function of teaching live broadcast. In order to improve the teaching effect, build students ‘cognitive system, cultivate students’ ability of independent learning and collaborative learning, and help students to achieve the knowledge objectives, skills objectives and quality objectives required by the curriculum. The author from the current situation of the post-epidemic era, combined with hunan automobile engineering vocational college online mixed teaching reality, with research results at home and abroad as the theoretical support, develop granular micro class video, system design orderly teaching link, in the actual practice of offline teaching mode, and reversed transmission teachers, teaching materials, teaching method reform.","PeriodicalId":6774,"journal":{"name":"2021 11th International Conference on Information Technology in Medicine and Education (ITME)","volume":"4 1","pages":"526-531"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79202878","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 : 2021-11-01DOI: 10.1109/ITME53901.2021.00023
Teng Xu, Jian Chen, Zuoyong Li, Yuanzheng Cai
Recently, official statistics reported that the Chinese population aged 60 and above has been 26.402 million, which accounts for 18.70% of total population. It is urgent to develop fall detection technologies for alleviating the risk causing by falling of elder person. In this paper, we propose a real-time, high-precision, and deep learning-based fall detection method with automatic person detection and tracking. Specifically, the proposed method first improves the YOLOv3 network to more efficiently detect person and extract feature maps of the object. Then, it inputs the extracted feature maps from the YOLOv3 into a multi-target tracking network for cascade matching and IOU matching in a Deep SORT algorithm, respectively. Next, it improves YOLOv5 network to detect posture anomalies. Finally, it refines the detected posture anomalies for obtaining the final fall detection result. Experimental results show that the proposed method simultaneously improves accuracy and efficiency of the fall detection.
{"title":"Fall Detection Based on Person Detection and Multi-target Tracking","authors":"Teng Xu, Jian Chen, Zuoyong Li, Yuanzheng Cai","doi":"10.1109/ITME53901.2021.00023","DOIUrl":"https://doi.org/10.1109/ITME53901.2021.00023","url":null,"abstract":"Recently, official statistics reported that the Chinese population aged 60 and above has been 26.402 million, which accounts for 18.70% of total population. It is urgent to develop fall detection technologies for alleviating the risk causing by falling of elder person. In this paper, we propose a real-time, high-precision, and deep learning-based fall detection method with automatic person detection and tracking. Specifically, the proposed method first improves the YOLOv3 network to more efficiently detect person and extract feature maps of the object. Then, it inputs the extracted feature maps from the YOLOv3 into a multi-target tracking network for cascade matching and IOU matching in a Deep SORT algorithm, respectively. Next, it improves YOLOv5 network to detect posture anomalies. Finally, it refines the detected posture anomalies for obtaining the final fall detection result. Experimental results show that the proposed method simultaneously improves accuracy and efficiency of the fall detection.","PeriodicalId":6774,"journal":{"name":"2021 11th International Conference on Information Technology in Medicine and Education (ITME)","volume":"29 1","pages":"60-65"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85833574","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 : 2021-11-01DOI: 10.1109/ITME53901.2021.00062
Min Huang, Lizhen Ye, Junze Chen, Rurui Fu, Changle Zhou
Meditation has been shown as an efficient way to promote human well-being. Most studies focused on meditation in sitting posture. However, meditation in walking posture was rarely studied. In order to identify these two meditation states (i.e., sitting and walking), we proposed a classification framework by leveraging different features extracted from the EEG signals and the random forest classifier. This study first investigated different single-modal features, including original power, power ratio, and non-linear dynamics. Further, we also concatenated all the single-modal features into a multi-modal feature. The experimental results show that the original power feature is better than the non-linear dynamics feature in meditation state classification. Moreover, the multi-modal feature outperforms all the single-modal features and can identify sitting and walking meditation with high accuracy.
{"title":"Feature Representation for Meditation State Classification in EEG Signal","authors":"Min Huang, Lizhen Ye, Junze Chen, Rurui Fu, Changle Zhou","doi":"10.1109/ITME53901.2021.00062","DOIUrl":"https://doi.org/10.1109/ITME53901.2021.00062","url":null,"abstract":"Meditation has been shown as an efficient way to promote human well-being. Most studies focused on meditation in sitting posture. However, meditation in walking posture was rarely studied. In order to identify these two meditation states (i.e., sitting and walking), we proposed a classification framework by leveraging different features extracted from the EEG signals and the random forest classifier. This study first investigated different single-modal features, including original power, power ratio, and non-linear dynamics. Further, we also concatenated all the single-modal features into a multi-modal feature. The experimental results show that the original power feature is better than the non-linear dynamics feature in meditation state classification. Moreover, the multi-modal feature outperforms all the single-modal features and can identify sitting and walking meditation with high accuracy.","PeriodicalId":6774,"journal":{"name":"2021 11th International Conference on Information Technology in Medicine and Education (ITME)","volume":"10 1","pages":"267-270"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89695723","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 : 2021-11-01DOI: 10.1109/ITME53901.2021.00026
Zhaoyang Zhao, Jianwei Ma, Chao Ma, Yuzhu Wang
Pedestrian detection is an important branch of computer vision and has been the focus of research due to its wide range of applications. Although commonly used object detection model Faster R-CNN has achieved good results. However, there are still some shortcomings in the specific task of detecting pedestrians. This paper made three improvements to the Faster R-CNN to better adapt it to the pedestrian detection task. First, we did a lot of experiments and finally chose MobileNetv2 as our backbone network. Second, we designed a multi-branch feature pyramid network (M-FPN), which is used to better integrate the model's shallow feature information with the deep feature information improved the model's ability to detect pedestrians. Finally, an attention region proposal network SE-RPN is used to improve the model's ability to focus on pedestrian features and suppress attention to background interference features. The experimental results show that the improvement strategy proposed in this paper has achieved better results. These strategies improve the average accuracy of Faster R-CNN on our self-built dataset by 6.14% and the detection speed by 27fps. The AP on Caltech dataset reaches 87.01%, and the detection speed can achieve 39.4fps.
{"title":"An Improved Faster R-CNN Algorithm for Pedestrian Detection","authors":"Zhaoyang Zhao, Jianwei Ma, Chao Ma, Yuzhu Wang","doi":"10.1109/ITME53901.2021.00026","DOIUrl":"https://doi.org/10.1109/ITME53901.2021.00026","url":null,"abstract":"Pedestrian detection is an important branch of computer vision and has been the focus of research due to its wide range of applications. Although commonly used object detection model Faster R-CNN has achieved good results. However, there are still some shortcomings in the specific task of detecting pedestrians. This paper made three improvements to the Faster R-CNN to better adapt it to the pedestrian detection task. First, we did a lot of experiments and finally chose MobileNetv2 as our backbone network. Second, we designed a multi-branch feature pyramid network (M-FPN), which is used to better integrate the model's shallow feature information with the deep feature information improved the model's ability to detect pedestrians. Finally, an attention region proposal network SE-RPN is used to improve the model's ability to focus on pedestrian features and suppress attention to background interference features. The experimental results show that the improvement strategy proposed in this paper has achieved better results. These strategies improve the average accuracy of Faster R-CNN on our self-built dataset by 6.14% and the detection speed by 27fps. The AP on Caltech dataset reaches 87.01%, and the detection speed can achieve 39.4fps.","PeriodicalId":6774,"journal":{"name":"2021 11th International Conference on Information Technology in Medicine and Education (ITME)","volume":"28 1","pages":"76-80"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87974102","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 : 2021-11-01DOI: 10.1109/ITME53901.2021.00070
Xinquan Yang, Xuechen Li, Linlin Shen, Min Cao, Changen Zhou
Lung disease screening using Chest x-ray (CXR) radiographs can obviously decrease the incidence of lung cancer. Using computer-aided diagnosis system to assist doctors in lung disease screening can greatly improve the diagnosis efficiency. In this paper, a coarse feature reuse deep neural network for CXR lesion detection is proposed. Firstly, we design a coarse feature reuse (CFR) block that can reuse low-level semantic features and extract high-level semantic information, which is used to replace the max-pooling layer in the shallow part of the network to achieve better feature extraction. A novel backbone network - RRCNet, which combines RepVGG block and Resblock, is proposed. The RepVggblock is used for better feature extraction at shallow layers and the Resblock is used for better feature fusion at deep layers. Extensive experiments on VinDr-CXR dataset demonstrate that our RRCNet-based detection network outperformes other classic detectors on both mAP (17.67%) and inference speed (0.1426s).
{"title":"A Coarse Feature Reuse Deep Neural Network for CXR Lesion Detection","authors":"Xinquan Yang, Xuechen Li, Linlin Shen, Min Cao, Changen Zhou","doi":"10.1109/ITME53901.2021.00070","DOIUrl":"https://doi.org/10.1109/ITME53901.2021.00070","url":null,"abstract":"Lung disease screening using Chest x-ray (CXR) radiographs can obviously decrease the incidence of lung cancer. Using computer-aided diagnosis system to assist doctors in lung disease screening can greatly improve the diagnosis efficiency. In this paper, a coarse feature reuse deep neural network for CXR lesion detection is proposed. Firstly, we design a coarse feature reuse (CFR) block that can reuse low-level semantic features and extract high-level semantic information, which is used to replace the max-pooling layer in the shallow part of the network to achieve better feature extraction. A novel backbone network - RRCNet, which combines RepVGG block and Resblock, is proposed. The RepVggblock is used for better feature extraction at shallow layers and the Resblock is used for better feature fusion at deep layers. Extensive experiments on VinDr-CXR dataset demonstrate that our RRCNet-based detection network outperformes other classic detectors on both mAP (17.67%) and inference speed (0.1426s).","PeriodicalId":6774,"journal":{"name":"2021 11th International Conference on Information Technology in Medicine and Education (ITME)","volume":"12 1","pages":"307-313"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87217973","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：