Recovering a clear image from single hazy image has been widely investigated in recent researches. Due to the lack of the real hazed image dataset, most studies use artificially synthesized dataset to train the models. Nonetheless, the real word foggy image is far different from the synthesized image. As a result, the existing methods could not defog the real foggy image well, when inputting the real foggy images. In this paper, we introduce a new dehazing algorithm, which adds cycle consistency constraints to the generative adversarial network (GAN). It implements the translation from foggy images to clean images without supervised learning, that is, the model does not need paired data to training. We assume that clear and foggy images come from different domains. There are two generators that act as domain translators, one from foggy image domain to clean image domain, and the other from foggy image to clean image. Two discriminators in the GAN are used for assessing each domain translator. The GAN loss, combined with the cycle consistency loss are used to regularize the model. We carried out experiments to evaluate the proposed method, and the results demonstrate the effectiveness in dehazing and there is indeed difference between the real-fog images and the synthetic images.
{"title":"Image Dehazing Via Cycle Generative Adversarial Network","authors":"Changyou Shi, Jianping Lu, Qian Sun, Shiliang Cheng, Xin Feng, Wei Huang","doi":"10.1145/3503047.3503135","DOIUrl":"https://doi.org/10.1145/3503047.3503135","url":null,"abstract":"Recovering a clear image from single hazy image has been widely investigated in recent researches. Due to the lack of the real hazed image dataset, most studies use artificially synthesized dataset to train the models. Nonetheless, the real word foggy image is far different from the synthesized image. As a result, the existing methods could not defog the real foggy image well, when inputting the real foggy images. In this paper, we introduce a new dehazing algorithm, which adds cycle consistency constraints to the generative adversarial network (GAN). It implements the translation from foggy images to clean images without supervised learning, that is, the model does not need paired data to training. We assume that clear and foggy images come from different domains. There are two generators that act as domain translators, one from foggy image domain to clean image domain, and the other from foggy image to clean image. Two discriminators in the GAN are used for assessing each domain translator. The GAN loss, combined with the cycle consistency loss are used to regularize the model. We carried out experiments to evaluate the proposed method, and the results demonstrate the effectiveness in dehazing and there is indeed difference between the real-fog images and the synthetic images.","PeriodicalId":190604,"journal":{"name":"Proceedings of the 3rd International Conference on Advanced Information Science and System","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125628176","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}
To address the unsatisfactory recognition of distorted text in images with complicated background, a neural-network-based approach was proposed to detect and recognize text. For detection of distorted text, the improved CRAFT model was applied, and deformable convolution was introduced to replace conventional convolution to sufficiently extract the features with irregular background. On this basis, the CRAFT-DCN text detection model was proposed to improve the accuracy of text detection. In order to reduce the interference of distorted text on the recognition model, images of separated texts were tailored according to the coordinates obtained by the detection model. Meanwhile, the Dense-CRNN model was designed, and the dense convolutional layer was introduced in the text recognition model to enhance the reuse of the features, thereby reducing interference of complicated background and recognizing separated distorted text correctly. The experiment results show that, compared with traditional approaches, the improved method introduced in this paper has better detection and recognition rates. And specifically, its text detection accuracy and the text recognition accuracy in actual scenario reach 86.3% and 95.3% respectively.
{"title":"Neural-network-based Approach to Detect and Recognize Distorted Text in Images with Complicated Background","authors":"Yuanyuan Qu, Wenxue Wei, Jiajia Jiang, Yufeng Liang","doi":"10.1145/3503047.3503118","DOIUrl":"https://doi.org/10.1145/3503047.3503118","url":null,"abstract":"To address the unsatisfactory recognition of distorted text in images with complicated background, a neural-network-based approach was proposed to detect and recognize text. For detection of distorted text, the improved CRAFT model was applied, and deformable convolution was introduced to replace conventional convolution to sufficiently extract the features with irregular background. On this basis, the CRAFT-DCN text detection model was proposed to improve the accuracy of text detection. In order to reduce the interference of distorted text on the recognition model, images of separated texts were tailored according to the coordinates obtained by the detection model. Meanwhile, the Dense-CRNN model was designed, and the dense convolutional layer was introduced in the text recognition model to enhance the reuse of the features, thereby reducing interference of complicated background and recognizing separated distorted text correctly. The experiment results show that, compared with traditional approaches, the improved method introduced in this paper has better detection and recognition rates. And specifically, its text detection accuracy and the text recognition accuracy in actual scenario reach 86.3% and 95.3% respectively.","PeriodicalId":190604,"journal":{"name":"Proceedings of the 3rd International Conference on Advanced Information Science and System","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130050530","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}
Gang Chen, Su-jun Wang, Y. Ping, Yi Jin, Changzhi Xu, Ying-zhao Shao, Zhao Han
The external commercial broadcast illuminators are not designed for radar, and the illuminators of opportunity usually have Dopplervarying structures. These structures usually cause ambiguity sidelobes in Doppler dimension. To solve the ambiguity sidelobe problem, a method of mismatched filtering that deals with the ambiguity Doppler sidelobes is proposed. In this new algorithm, the mismatched filtering factor is acquired based on minimizing the signal energy loss and the total energy of the ambiguity Doppler sidelobes. The experimental result shows the effectiveness of the proposed algorithm. CCS CONCEPTS • Computing methodologies; • Distributed computing methodologies;
{"title":"Mismatched filtering for Doppler ambiguity sidelobe suppression in passive bistatic radar","authors":"Gang Chen, Su-jun Wang, Y. Ping, Yi Jin, Changzhi Xu, Ying-zhao Shao, Zhao Han","doi":"10.1145/3503047.3503115","DOIUrl":"https://doi.org/10.1145/3503047.3503115","url":null,"abstract":"The external commercial broadcast illuminators are not designed for radar, and the illuminators of opportunity usually have Dopplervarying structures. These structures usually cause ambiguity sidelobes in Doppler dimension. To solve the ambiguity sidelobe problem, a method of mismatched filtering that deals with the ambiguity Doppler sidelobes is proposed. In this new algorithm, the mismatched filtering factor is acquired based on minimizing the signal energy loss and the total energy of the ambiguity Doppler sidelobes. The experimental result shows the effectiveness of the proposed algorithm. CCS CONCEPTS • Computing methodologies; • Distributed computing methodologies;","PeriodicalId":190604,"journal":{"name":"Proceedings of the 3rd International Conference on Advanced Information Science and System","volume":"61 8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131059080","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}
With the popularization of higher education in my country and the number of college students increasing year by year, the number of students in colleges and universities that need to be subsidized is also increasing day by day. At present, colleges and universities mostly use offline material collection and manual operation. Not only the workload and the amount of tasks are heavy, but also the current situation of irregular funding management is prone to cause unnecessary troubles to the funding work. There is an urgent need to establish a complete management system for student assistance. This paper presents a design method for the information management system of the financial aid for poor students in colleges and universities based on the Web platform. This method is based on the construction of campus network and adds data analysis modules to realize an efficient informatization management of university funding work.
{"title":"Method design based on the national scholarship system","authors":"D. Shi, Jun-Xian Ji","doi":"10.1145/3503047.3503137","DOIUrl":"https://doi.org/10.1145/3503047.3503137","url":null,"abstract":"With the popularization of higher education in my country and the number of college students increasing year by year, the number of students in colleges and universities that need to be subsidized is also increasing day by day. At present, colleges and universities mostly use offline material collection and manual operation. Not only the workload and the amount of tasks are heavy, but also the current situation of irregular funding management is prone to cause unnecessary troubles to the funding work. There is an urgent need to establish a complete management system for student assistance. This paper presents a design method for the information management system of the financial aid for poor students in colleges and universities based on the Web platform. This method is based on the construction of campus network and adds data analysis modules to realize an efficient informatization management of university funding work.","PeriodicalId":190604,"journal":{"name":"Proceedings of the 3rd International Conference on Advanced Information Science and System","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134352091","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}
Modeling the process of information dissemination in online social networks is an important task that helps understand the information diffusion mechanism and analyze the influencing factors. However, most of the existing models focus on the diffusion process of general information, rumors or public sentiments, instead of concerning specific information such as sports news. In this paper, we analyze the diffusion characteristics of sports news in Sina-Weibo. An information dissemination model for sports news is constructed based on the SEIR epidemic model, and the node transition probabilities are set dynamically according to three information factors and user interest attenuation. The experimental results show that the proposed model accurately reflects the dissemination mechanism of sports news. Among four influencing factors, the information value and the published time obviously affect the speed and the range of the diffusion process.
{"title":"Modeling of Sports News Information Dissemination in Social Networks","authors":"Yujia Fu, Jingling Wang","doi":"10.1145/3503047.3503065","DOIUrl":"https://doi.org/10.1145/3503047.3503065","url":null,"abstract":"Modeling the process of information dissemination in online social networks is an important task that helps understand the information diffusion mechanism and analyze the influencing factors. However, most of the existing models focus on the diffusion process of general information, rumors or public sentiments, instead of concerning specific information such as sports news. In this paper, we analyze the diffusion characteristics of sports news in Sina-Weibo. An information dissemination model for sports news is constructed based on the SEIR epidemic model, and the node transition probabilities are set dynamically according to three information factors and user interest attenuation. The experimental results show that the proposed model accurately reflects the dissemination mechanism of sports news. Among four influencing factors, the information value and the published time obviously affect the speed and the range of the diffusion process.","PeriodicalId":190604,"journal":{"name":"Proceedings of the 3rd International Conference on Advanced Information Science and System","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134478422","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}
The concept of architecture is put forward by the US Army, which refers to the composition structure of the system and its mutual relationship, as well as the principles and guidelines guiding the design and development of the system. Architecture design technology and its reference resources have important theoretical and technical support for the top-level design of information system. Based on the analysis of architecture technology and function, this paper systematically studies the connotation and development history of the architecture design framework of the US army, and sorts out the main reference resources of the architecture design of the US army, such as the common joint task list, information system interoperability level model, joint technology architecture. It provides useful reference for the theoretical and technical research of information system top-level design.
{"title":"Research on the information System architecture design framework and reference resources of American Army","authors":"Ping Jian","doi":"10.1145/3503047.3503094","DOIUrl":"https://doi.org/10.1145/3503047.3503094","url":null,"abstract":"The concept of architecture is put forward by the US Army, which refers to the composition structure of the system and its mutual relationship, as well as the principles and guidelines guiding the design and development of the system. Architecture design technology and its reference resources have important theoretical and technical support for the top-level design of information system. Based on the analysis of architecture technology and function, this paper systematically studies the connotation and development history of the architecture design framework of the US army, and sorts out the main reference resources of the architecture design of the US army, such as the common joint task list, information system interoperability level model, joint technology architecture. It provides useful reference for the theoretical and technical research of information system top-level design.","PeriodicalId":190604,"journal":{"name":"Proceedings of the 3rd International Conference on Advanced Information Science and System","volume":"97 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114581831","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}
Aiming at the problem that it is difficult to recognize the working mode of multi-functional radar without prior information, this paper proposes a radar working mode recognition method based on complex network analysis. Use the simulated reconnaissance data of the same type of radar to build a complex network, by analyzing the importance of network nodes, the radar phrases in search and non-search working modes are recognized. And Gephi is used as a visualization platform to show the regular features of radar phrases in different working modes. On this basis, the functional state of the radar system is analyzed by calculating the network density of the complex network.
{"title":"Radar Working Mode Recognition Method Based on Complex Network Analysis","authors":"Liu Yang, Yan-Lou He, Shouye Lv, Jianfeng Ma","doi":"10.1145/3503047.3503051","DOIUrl":"https://doi.org/10.1145/3503047.3503051","url":null,"abstract":"Aiming at the problem that it is difficult to recognize the working mode of multi-functional radar without prior information, this paper proposes a radar working mode recognition method based on complex network analysis. Use the simulated reconnaissance data of the same type of radar to build a complex network, by analyzing the importance of network nodes, the radar phrases in search and non-search working modes are recognized. And Gephi is used as a visualization platform to show the regular features of radar phrases in different working modes. On this basis, the functional state of the radar system is analyzed by calculating the network density of the complex network.","PeriodicalId":190604,"journal":{"name":"Proceedings of the 3rd International Conference on Advanced Information Science and System","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114250066","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}
The shortest linear program has been proved to be a NP-hard problem. In order to obtain the better approximate solution, a frequency-first heuristic method is proposed, which can optimize the number of XOR gates required by linear components while ensuring the stability of the algorithm. Firstly, we combine the pre-emptive gate strategy with frequency-first (FQCY) in the selection stage to reduce the increase of time complexity caused by exhaustive search, so that the high-density matrix can obtain the optimal result within a reasonable time. Secondly, minimization of vector and appropriate randomization are added to deal with the tie, so as to give full play to the advantages of cancellation-allowed circuit and increase the possibility of obtaining the optimal solution. Finally, compared with Paar, BP, RNBP, RSDF algorithms on random matrices of various sizes and densities, it is proved that the probability of obtaining the optimal solution of the proposed algorithm in circuit depth is more than 30% higher than RNBP and RSDF. In terms of the number of XOR gates, especially for larger matrix, the probability of obtaining the optimal solution increases by more than 10%. The stability of the optimal circuit generated by this algorithm is about 90%.
{"title":"A Frequency-first Heuristic for Shortest Linear Programs","authors":"Hua Jiang, Heng Zhang, Huijiao Wang, Xin Wang","doi":"10.1145/3503047.3503069","DOIUrl":"https://doi.org/10.1145/3503047.3503069","url":null,"abstract":"The shortest linear program has been proved to be a NP-hard problem. In order to obtain the better approximate solution, a frequency-first heuristic method is proposed, which can optimize the number of XOR gates required by linear components while ensuring the stability of the algorithm. Firstly, we combine the pre-emptive gate strategy with frequency-first (FQCY) in the selection stage to reduce the increase of time complexity caused by exhaustive search, so that the high-density matrix can obtain the optimal result within a reasonable time. Secondly, minimization of vector and appropriate randomization are added to deal with the tie, so as to give full play to the advantages of cancellation-allowed circuit and increase the possibility of obtaining the optimal solution. Finally, compared with Paar, BP, RNBP, RSDF algorithms on random matrices of various sizes and densities, it is proved that the probability of obtaining the optimal solution of the proposed algorithm in circuit depth is more than 30% higher than RNBP and RSDF. In terms of the number of XOR gates, especially for larger matrix, the probability of obtaining the optimal solution increases by more than 10%. The stability of the optimal circuit generated by this algorithm is about 90%.","PeriodicalId":190604,"journal":{"name":"Proceedings of the 3rd International Conference on Advanced Information Science and System","volume":"287 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124571831","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}
Malware classification is a major challenge as they have multiple families and its type has been ever increasing. With the involvement of deep learning and the availability of massive data, neural networks can easily address this problem. This experimental work focuses on classifying the malware that are in the form of grayscale images into their respective families with high accuracy and low loss. We used transfer learning in a pretrained VGG16 model obtaining an accuracy of 88.40% of accuracy. Additionally, upon experimenting with the ResNet-18, InceptionV3 model to classify the malware images into their families didn't give better results as compared to our custom model. The custom model based on convolution neural network achieved better accuracy and was able to classify malware with 98.7% validation accuracy. Upon comparing our custom model with VGG16, ResNet-18, InceptionV3 the custom model gave better accuracy with a better f1 score of 0.99. But improper tuning of VGG16 yielded low accuracy and high parameter loss. In overall the approach of classifying malware by converting them into images and classifying thus obtained images makes the malware classification problem easier.
{"title":"Image-based Malware Classification using Deep Convolutional Neural Network and Transfer Learning","authors":"Dipendra Pant, Rabindra Bista","doi":"10.1145/3503047.3503081","DOIUrl":"https://doi.org/10.1145/3503047.3503081","url":null,"abstract":"Malware classification is a major challenge as they have multiple families and its type has been ever increasing. With the involvement of deep learning and the availability of massive data, neural networks can easily address this problem. This experimental work focuses on classifying the malware that are in the form of grayscale images into their respective families with high accuracy and low loss. We used transfer learning in a pretrained VGG16 model obtaining an accuracy of 88.40% of accuracy. Additionally, upon experimenting with the ResNet-18, InceptionV3 model to classify the malware images into their families didn't give better results as compared to our custom model. The custom model based on convolution neural network achieved better accuracy and was able to classify malware with 98.7% validation accuracy. Upon comparing our custom model with VGG16, ResNet-18, InceptionV3 the custom model gave better accuracy with a better f1 score of 0.99. But improper tuning of VGG16 yielded low accuracy and high parameter loss. In overall the approach of classifying malware by converting them into images and classifying thus obtained images makes the malware classification problem easier.","PeriodicalId":190604,"journal":{"name":"Proceedings of the 3rd International Conference on Advanced Information Science and System","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125109689","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}
Unmanned Aerial Vehicles (UAVs) have been attracting more and more attention in research and education. Specifically, Swarm intelligence is a promising future technology of UAVs and the frontier of multi-agent system research. It has the characteristics of low individual cost, strong system flexibility and robustness, and has great potential in many tasks. However, due to the constraints of research conditions and cost, most of the current researches on large-scale swarm UAVs are carried out in the simulation environment. Building a low-cost open-source software and hardware platform for swarm UAVs is an important basis for promoting researches on swarm UAVs and multi-agent systems. In this paper, we propose a design of a UAV platform with common cost-efficient hardware and a rich open-source software ecosystem, and provide a software solution for swarm robots based on the open-source robot operating system ROS. These software packages support the rapid programming development of swarm behaviors and different communication topology. Experiments have been conducted for typical UAV tasks like flocking and formation, indicating the effectiveness of the proposed platform.
{"title":"A Cost-Efficient Platform Design for Distributed UAV Swarm Research","authors":"Zhongxuan Cai, Xuefeng Chang, Minglong Li","doi":"10.1145/3503047.3503070","DOIUrl":"https://doi.org/10.1145/3503047.3503070","url":null,"abstract":"Unmanned Aerial Vehicles (UAVs) have been attracting more and more attention in research and education. Specifically, Swarm intelligence is a promising future technology of UAVs and the frontier of multi-agent system research. It has the characteristics of low individual cost, strong system flexibility and robustness, and has great potential in many tasks. However, due to the constraints of research conditions and cost, most of the current researches on large-scale swarm UAVs are carried out in the simulation environment. Building a low-cost open-source software and hardware platform for swarm UAVs is an important basis for promoting researches on swarm UAVs and multi-agent systems. In this paper, we propose a design of a UAV platform with common cost-efficient hardware and a rich open-source software ecosystem, and provide a software solution for swarm robots based on the open-source robot operating system ROS. These software packages support the rapid programming development of swarm behaviors and different communication topology. Experiments have been conducted for typical UAV tasks like flocking and formation, indicating the effectiveness of the proposed platform.","PeriodicalId":190604,"journal":{"name":"Proceedings of the 3rd International Conference on Advanced Information Science and System","volume":"219 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116383850","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：