Pub Date : 2022-05-20DOI: 10.1109/cvidliccea56201.2022.9824721
Honglin Ji, Zijian Cui
Dim infrared targets often have low recognition accuracy in complex environments, because of its small size and low brightness. Therefore, this paper presents an infrared background suppression method based on low-pass adaptive morphological filtering. In this method, the infrared background noise is connected into multiple regions by low-pass filtering, the adaptive morphological mask is generated by local mean variance ratio and region growth, and then most of the background is suppressed by top-hat operation. The test results show that this method can suppress the background of infrared images and enhance the target signal.
{"title":"Infrared background suppression method based on low-pass adaptive morphological filtering","authors":"Honglin Ji, Zijian Cui","doi":"10.1109/cvidliccea56201.2022.9824721","DOIUrl":"https://doi.org/10.1109/cvidliccea56201.2022.9824721","url":null,"abstract":"Dim infrared targets often have low recognition accuracy in complex environments, because of its small size and low brightness. Therefore, this paper presents an infrared background suppression method based on low-pass adaptive morphological filtering. In this method, the infrared background noise is connected into multiple regions by low-pass filtering, the adaptive morphological mask is generated by local mean variance ratio and region growth, and then most of the background is suppressed by top-hat operation. The test results show that this method can suppress the background of infrared images and enhance the target signal.","PeriodicalId":23649,"journal":{"name":"Vision","volume":"11 1","pages":"574-577"},"PeriodicalIF":0.0,"publicationDate":"2022-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79818296","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 : 2022-05-20DOI: 10.1109/cvidliccea56201.2022.9824243
Yanbing Wang, Fang Lv
In order to solve the problems of complexity and low efficiency of traditional water quality monitoring methods, this paper designs a water quality monitoring system combining STM32F103 microcontroller and Huawei Cloud IoT platform based on narrowband Internet of Things technology. This system can continuously collect the temperature, pH, TDS and ORP data of the automatically monitored waters of the target. The data can be automatically uploaded to the cloud platform in real time, and users can query real-time monitoring information on the HUAWEI CLOUD IoT platform, which overcomes the shortcomings of traditional water quality monitoring systems such as long data collection cycles and poor real-time performance, and has certain innovation and application value.
{"title":"Design of water quality monitoring system based on NB-IoT technolog","authors":"Yanbing Wang, Fang Lv","doi":"10.1109/cvidliccea56201.2022.9824243","DOIUrl":"https://doi.org/10.1109/cvidliccea56201.2022.9824243","url":null,"abstract":"In order to solve the problems of complexity and low efficiency of traditional water quality monitoring methods, this paper designs a water quality monitoring system combining STM32F103 microcontroller and Huawei Cloud IoT platform based on narrowband Internet of Things technology. This system can continuously collect the temperature, pH, TDS and ORP data of the automatically monitored waters of the target. The data can be automatically uploaded to the cloud platform in real time, and users can query real-time monitoring information on the HUAWEI CLOUD IoT platform, which overcomes the shortcomings of traditional water quality monitoring systems such as long data collection cycles and poor real-time performance, and has certain innovation and application value.","PeriodicalId":23649,"journal":{"name":"Vision","volume":"15 1","pages":"825-828"},"PeriodicalIF":0.0,"publicationDate":"2022-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84142520","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 : 2022-05-20DOI: 10.1109/cvidliccea56201.2022.9825036
Xinyu Zhang
Brain tumors have always been one of the common tumors threatening human life safety. At present, there are still relatively few computer-aided diagnostic systems in China specifically for detection of specific conditions of brain tumor, as well as related studies. This study collected a certain number of publicly available datasets of brain magnetic resonance imaging (MRI) images and data preprocessing such as normalization was conducted on it. According to the characteristics of medical image complexity of brain MRI, this study proposed an approach of incorporating attention mechanism with Convolutional Neural Network (CNN) to reduce the influence caused by irrelevant background information features in images. The experiment results based on the proposed method were compared with self-defined classic models such as VGGNet and MobileNet. Through testing on the dataset, the results show that the CNN model's accuracy after adding an attention mechanism improves significantly compared to the other three models, demonstrating that the attention mechanism in the model can reduce the impact of context irrelevant information to the classification outcome to some extent and performed well on the brain tumor recognition classification task. Finally, this paper deploys the trained analysis model on the web page, the interface is simple and friendly, and convenient for medical staff to operate.
{"title":"A Highly Accurate Attention-Based Convolutional Neural Network for Classification of Brain Tumors","authors":"Xinyu Zhang","doi":"10.1109/cvidliccea56201.2022.9825036","DOIUrl":"https://doi.org/10.1109/cvidliccea56201.2022.9825036","url":null,"abstract":"Brain tumors have always been one of the common tumors threatening human life safety. At present, there are still relatively few computer-aided diagnostic systems in China specifically for detection of specific conditions of brain tumor, as well as related studies. This study collected a certain number of publicly available datasets of brain magnetic resonance imaging (MRI) images and data preprocessing such as normalization was conducted on it. According to the characteristics of medical image complexity of brain MRI, this study proposed an approach of incorporating attention mechanism with Convolutional Neural Network (CNN) to reduce the influence caused by irrelevant background information features in images. The experiment results based on the proposed method were compared with self-defined classic models such as VGGNet and MobileNet. Through testing on the dataset, the results show that the CNN model's accuracy after adding an attention mechanism improves significantly compared to the other three models, demonstrating that the attention mechanism in the model can reduce the impact of context irrelevant information to the classification outcome to some extent and performed well on the brain tumor recognition classification task. Finally, this paper deploys the trained analysis model on the web page, the interface is simple and friendly, and convenient for medical staff to operate.","PeriodicalId":23649,"journal":{"name":"Vision","volume":"5 1","pages":"124-128"},"PeriodicalIF":0.0,"publicationDate":"2022-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80941682","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 : 2022-05-20DOI: 10.1109/cvidliccea56201.2022.9823984
Shuai Guo
Online education has developed rapidly since 2020, and the completion of after-school exercises is a part of online education, which plays an important role in improving students’ knowledge. However, the existing question recommendation systems mainly have two problems: (1) The question recommendation is completely based on the parametric theoretical model. The parametric theoretical model parameterizes the questions and the students’ ability to answer the questions, so it cannot provide a personalized question recommendation strategy. (2) The question recommendation strategy depends on the teacher’s formulation, and the efficiency is not high. In order to solve the above two problems, this paper is based on deep knowledge tracing and uses a strategy for recommending questions for students’ weak knowledge points. This method first uses the deep knowledge tracing model to model students’ personal knowledge level, and then finds out students’ weak knowledge points. Recommend questions for students’ weak knowledge points. Under the real experimental data set, this method can recommend personalized questions for students without the participation of experts.
{"title":"Design and implementation of question recommendation system based on deep knowledge tracing","authors":"Shuai Guo","doi":"10.1109/cvidliccea56201.2022.9823984","DOIUrl":"https://doi.org/10.1109/cvidliccea56201.2022.9823984","url":null,"abstract":"Online education has developed rapidly since 2020, and the completion of after-school exercises is a part of online education, which plays an important role in improving students’ knowledge. However, the existing question recommendation systems mainly have two problems: (1) The question recommendation is completely based on the parametric theoretical model. The parametric theoretical model parameterizes the questions and the students’ ability to answer the questions, so it cannot provide a personalized question recommendation strategy. (2) The question recommendation strategy depends on the teacher’s formulation, and the efficiency is not high. In order to solve the above two problems, this paper is based on deep knowledge tracing and uses a strategy for recommending questions for students’ weak knowledge points. This method first uses the deep knowledge tracing model to model students’ personal knowledge level, and then finds out students’ weak knowledge points. Recommend questions for students’ weak knowledge points. Under the real experimental data set, this method can recommend personalized questions for students without the participation of experts.","PeriodicalId":23649,"journal":{"name":"Vision","volume":"20 1","pages":"1041-1046"},"PeriodicalIF":0.0,"publicationDate":"2022-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80994085","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}
2D/3D image registration is one of the key technologies to realize pose estimation in computer-aided surgery. In order to improve the global and local search performance of the model in the pose parameter space, an improved multi-resolution 2D/3D registration method is proposed in this paper. Firstly, aiming at the problem that the intensity-based similarity measure is not sensitive to small offset between images, the gradient information with strong sensitivity to image texture edge is introduced, and the Intensity and Gradient Weighted Correlation (IGWC) coefficient similarity measure is proposed; Secondly, aiming at the problem of slow convergence of global optimization algorithm and small capture range of local optimization algorithm, a global-local combined registration optimization strategy is proposed. The experimental results show that this method improves the registration accuracy and success rate.
{"title":"An improved multi-resolution 2D/3D registration method","authors":"Yipei Cao, Fei He, Feng Qu, Tiejun Wang, Chen Yang, Weili Shi, Zhengang Jiang","doi":"10.1109/cvidliccea56201.2022.9824315","DOIUrl":"https://doi.org/10.1109/cvidliccea56201.2022.9824315","url":null,"abstract":"2D/3D image registration is one of the key technologies to realize pose estimation in computer-aided surgery. In order to improve the global and local search performance of the model in the pose parameter space, an improved multi-resolution 2D/3D registration method is proposed in this paper. Firstly, aiming at the problem that the intensity-based similarity measure is not sensitive to small offset between images, the gradient information with strong sensitivity to image texture edge is introduced, and the Intensity and Gradient Weighted Correlation (IGWC) coefficient similarity measure is proposed; Secondly, aiming at the problem of slow convergence of global optimization algorithm and small capture range of local optimization algorithm, a global-local combined registration optimization strategy is proposed. The experimental results show that this method improves the registration accuracy and success rate.","PeriodicalId":23649,"journal":{"name":"Vision","volume":"28 1","pages":"1-6"},"PeriodicalIF":0.0,"publicationDate":"2022-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78749349","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 : 2022-05-20DOI: 10.1109/cvidliccea56201.2022.9824089
S. Lou, Gang Liu, Zhiyu Chen, Jianwei Guo, Peng Liu
Mobile crowdsensing (MCS) is a new crowdsourcing model. With the continuous development of MCS, more and more task requesters and workers participate in the MCS, and how to design a reasonable task allocation scheme hasbecome a hot topic of research. In this paper, we investigate the spatiotemporal task allocation problem considering task time constraints and workers’ execution capabilities, and proposean efficient task allocation algorithm based on the discrete particle swarm optimization to maximise social welfare. In order to further optimise the task allocation scheme, a greedy algorithm is introduced to reduce the distance workers have to travel to perform the task and hence the cost of performing the task. Simulation results show that the algorithm is effective in improving social welfare.
{"title":"Mobile Crowdsensing Task Allocation Model Based on Discrete Particle Swarm Optimization","authors":"S. Lou, Gang Liu, Zhiyu Chen, Jianwei Guo, Peng Liu","doi":"10.1109/cvidliccea56201.2022.9824089","DOIUrl":"https://doi.org/10.1109/cvidliccea56201.2022.9824089","url":null,"abstract":"Mobile crowdsensing (MCS) is a new crowdsourcing model. With the continuous development of MCS, more and more task requesters and workers participate in the MCS, and how to design a reasonable task allocation scheme hasbecome a hot topic of research. In this paper, we investigate the spatiotemporal task allocation problem considering task time constraints and workers’ execution capabilities, and proposean efficient task allocation algorithm based on the discrete particle swarm optimization to maximise social welfare. In order to further optimise the task allocation scheme, a greedy algorithm is introduced to reduce the distance workers have to travel to perform the task and hence the cost of performing the task. Simulation results show that the algorithm is effective in improving social welfare.","PeriodicalId":23649,"journal":{"name":"Vision","volume":"10 1","pages":"1-7"},"PeriodicalIF":0.0,"publicationDate":"2022-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82862683","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 : 2022-05-20DOI: 10.1109/cvidliccea56201.2022.9824933
Zhen He, Liang Ye
Traditional service methods can not guarantee the development needs of IP network. Aiming at this problem, this paper designs a transmission mechanism based on virtual network slicing and time sensitivity differentiation. By distinguishing the QoS requirements of different services, the time sensitive flow can be forwarded first, so as to ensure that the time sensitive flow can obtain the maximum QoS guarantee through the optimal forwarding path. The experimental results show that the transmission method used in this paper can better balance the network load and improve the resource utilization of the network.
{"title":"Research and implementation of real-time transmission technology for industrial interconnection","authors":"Zhen He, Liang Ye","doi":"10.1109/cvidliccea56201.2022.9824933","DOIUrl":"https://doi.org/10.1109/cvidliccea56201.2022.9824933","url":null,"abstract":"Traditional service methods can not guarantee the development needs of IP network. Aiming at this problem, this paper designs a transmission mechanism based on virtual network slicing and time sensitivity differentiation. By distinguishing the QoS requirements of different services, the time sensitive flow can be forwarded first, so as to ensure that the time sensitive flow can obtain the maximum QoS guarantee through the optimal forwarding path. The experimental results show that the transmission method used in this paper can better balance the network load and improve the resource utilization of the network.","PeriodicalId":23649,"journal":{"name":"Vision","volume":"52 1","pages":"1133-1136"},"PeriodicalIF":0.0,"publicationDate":"2022-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90974783","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 : 2022-05-20DOI: 10.1109/cvidliccea56201.2022.9825138
Lei Zhang, Haoying Wu
To solve the problems of mode collapse and training instability in generative adversarial networks (GANs), a framework simulating the complementary structure of DNA is proposed, in which a complementary unit and a generalization unit are added. Four latent vectors representing four bases of A, T,C and G are obtained from the complementary unit. Through the combination of latent vectors, the generalization unit avoids the fitting of high-dimensional data distribution and obtains a more comprehensive vector space. Experimental results show that the problems of model collapse and training instability are effectively solved, compared with state-of-the-art VAE-GAN, the FID score increases 52.2%, indicating that the quality and diversity of images generated by the model are improved.
{"title":"A Novel Generative Adversarial Network simulating the complementary structure of DNA genetic information","authors":"Lei Zhang, Haoying Wu","doi":"10.1109/cvidliccea56201.2022.9825138","DOIUrl":"https://doi.org/10.1109/cvidliccea56201.2022.9825138","url":null,"abstract":"To solve the problems of mode collapse and training instability in generative adversarial networks (GANs), a framework simulating the complementary structure of DNA is proposed, in which a complementary unit and a generalization unit are added. Four latent vectors representing four bases of A, T,C and G are obtained from the complementary unit. Through the combination of latent vectors, the generalization unit avoids the fitting of high-dimensional data distribution and obtains a more comprehensive vector space. Experimental results show that the problems of model collapse and training instability are effectively solved, compared with state-of-the-art VAE-GAN, the FID score increases 52.2%, indicating that the quality and diversity of images generated by the model are improved.","PeriodicalId":23649,"journal":{"name":"Vision","volume":"1 1","pages":"9-14"},"PeriodicalIF":0.0,"publicationDate":"2022-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90463070","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 : 2022-05-20DOI: 10.1109/cvidliccea56201.2022.9825271
Cheng Jiang, JiaQi Sun, Kexin Qi, Cheng Jin, GangTie Jin
The key technique of sit-and-reach distance measurement in this study mainly consists of two stages: the first stage is the initial calibration of the test site, including camera calibration, identification and calibration of the test site identification points and finger key points; the second stage is the distance measurement stage, including the calculation of the coordinates of the finger tip to the scale projection point of the suspended distance measurement and the calculation of the conversion distance of the projection point. After the validity test of ranging, the error was 0. 148cm with standard deviation of 0.118, maximum value of 0.495, and minimum value of 0.002 for 90 experiments, which proved that the research results had high ranging accuracy. Since this study uses a common webcam, the method is easy to be widely used.
{"title":"Exploring a Computer Vision and Artificial Intelligence-based Approach to Sit-and-reach Distance Measurement","authors":"Cheng Jiang, JiaQi Sun, Kexin Qi, Cheng Jin, GangTie Jin","doi":"10.1109/cvidliccea56201.2022.9825271","DOIUrl":"https://doi.org/10.1109/cvidliccea56201.2022.9825271","url":null,"abstract":"The key technique of sit-and-reach distance measurement in this study mainly consists of two stages: the first stage is the initial calibration of the test site, including camera calibration, identification and calibration of the test site identification points and finger key points; the second stage is the distance measurement stage, including the calculation of the coordinates of the finger tip to the scale projection point of the suspended distance measurement and the calculation of the conversion distance of the projection point. After the validity test of ranging, the error was 0. 148cm with standard deviation of 0.118, maximum value of 0.495, and minimum value of 0.002 for 90 experiments, which proved that the research results had high ranging accuracy. Since this study uses a common webcam, the method is easy to be widely used.","PeriodicalId":23649,"journal":{"name":"Vision","volume":"1 1","pages":"225-229"},"PeriodicalIF":0.0,"publicationDate":"2022-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89766635","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 : 2022-05-20DOI: 10.1109/cvidliccea56201.2022.9824846
Xintao Xu, Zhelong Jiang, Gang Chen, Zhigang Li, Guoliang Gong, Huaxiang Lu
This paper proposes a generic neural network fitting algorithm based on CNN for nonlinear functions that overcomes the challenges of a large number of nonlinear functions in terms of hardware deployment and computing circuit generality in diverse neural network models. The model takes advantage of the principle that functions have varying degrees of difficulty fitting in different spaces, mapping the input to high-dimensional space with 1x1 convolution, and utilizing CNN to extract features of nonlinear functions with its strong feature extraction ability in high-dimensional space. Furthermore, MaxPool and ReLU improve the ability of nonlinear fitting. When fitting Tanh, Sigmoid, and ELU activation functions with 16bit accuracy, the proposed algorithm has an average error of less than 0.0006, with a parameter size of 5.793 k.
{"title":"General nonlinear function neural network fitting algorithm based on CNN","authors":"Xintao Xu, Zhelong Jiang, Gang Chen, Zhigang Li, Guoliang Gong, Huaxiang Lu","doi":"10.1109/cvidliccea56201.2022.9824846","DOIUrl":"https://doi.org/10.1109/cvidliccea56201.2022.9824846","url":null,"abstract":"This paper proposes a generic neural network fitting algorithm based on CNN for nonlinear functions that overcomes the challenges of a large number of nonlinear functions in terms of hardware deployment and computing circuit generality in diverse neural network models. The model takes advantage of the principle that functions have varying degrees of difficulty fitting in different spaces, mapping the input to high-dimensional space with 1x1 convolution, and utilizing CNN to extract features of nonlinear functions with its strong feature extraction ability in high-dimensional space. Furthermore, MaxPool and ReLU improve the ability of nonlinear fitting. When fitting Tanh, Sigmoid, and ELU activation functions with 16bit accuracy, the proposed algorithm has an average error of less than 0.0006, with a parameter size of 5.793 k.","PeriodicalId":23649,"journal":{"name":"Vision","volume":"65 1","pages":"1079-1082"},"PeriodicalIF":0.0,"publicationDate":"2022-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86519190","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：