{"title":"Improving Radiology Report Generation via Object Dropout Strategy and MLP-based Captioner","authors":"Zhixian Chen, Yuhao Tang","doi":"10.1109/IMCEC55388.2022.10019809","DOIUrl":"https://doi.org/10.1109/IMCEC55388.2022.10019809","url":null,"abstract":"Recently, automatically generating radiology reports has been addressed since it can not only relieve the pressure on doctors but also avoid misdiagnosis. Radiology report generation is a fundamental and critical step of auxiliary diagnosis. Due to the COVID-19 pandemic, a more accurate and robust structure for radiology report generation is urgently needed. Although radiology report generation is achieving remarkable progress, existing methods still face two main shortcomings. On the one hand, the strong noise in medical images usually interferes with the diagnosis process. On the other hand, these methods usually require complex structure while ignoring that efficiency is also an important metric for this task. To solve the two aforementioned problems, we introduce a novel method for medical report generation, the termed attention-guided object dropout MLP(ODM) model. In brief, ODM first incorporates a tailored pre-trained model to pre-align medical regions and corresponding language reports to capture text-related image features. Then, a fine-grained dropout strategy based on the attention matrix is proposed to relieve training pressure by dropping content-irrelevant information. Finally, inspired by the lightweight structure of Multilayer Perceptron(MLP), ODM adopts an MLP-based structure as an encoder to simplify the entire framework. Extensive experiments demonstrate the effectiveness of our ODM. More remarkably, ODM achieves state-of-the-art performance on IU X-Ray, MIMIC-CXR, and ROCO datasets, with the CIDEr-D score being increased from 26.8% to 41.4%, 21.1% to 30.2%, and 9.1% to 19.3%, respectively.","PeriodicalId":423151,"journal":{"name":"2022 IEEE 5th Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115228422","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}
Chen Ruiqiong, L. Ya, Zhao Zhixiong, Fan Duosheng, Qin Weijin, Xu Longxia
{"title":"Research on Remote Reproduction of Standard Time Based on Different Time Comparison Methods","authors":"Chen Ruiqiong, L. Ya, Zhao Zhixiong, Fan Duosheng, Qin Weijin, Xu Longxia","doi":"10.1109/IMCEC55388.2022.10019948","DOIUrl":"https://doi.org/10.1109/IMCEC55388.2022.10019948","url":null,"abstract":"This paper studies various high-precision time comparison technologies such as GNSS common-view, all-view, precise point position(PPP), etc., and based on the world's top four national time service center standard time UTC (NTSC), proposes real-time common-view, real-time all-view and real-time PPP complementary and integrated remote time comparison technology, combined with equipment delay calibration technology and oscillator dynamic taming technology, it has established a standard time remote reproduction system, which consists of a base station and user stations. The base station is maintained by the National Time Service Center, and the users only need to install a standard time reproduction device on the client side, and automatically and continuously obtain base station data from the server side through the internet, and the national standard time can be restored. It can provide global users indiscriminately with a 1PPS signal with a deviation of less than 5ns from the national standard time UTC (NTSC), and the timing uncertainty is better than 2ns. At present, it has provided services for many industries.","PeriodicalId":423151,"journal":{"name":"2022 IEEE 5th Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC)","volume":"72 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121823092","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Exploring the Possibility of Higher-Order Fault Attacks","authors":"Tong Wu, Dawei Zhou, Qingyu Ou, Weiyu Chu","doi":"10.1109/IMCEC55388.2022.10019960","DOIUrl":"https://doi.org/10.1109/IMCEC55388.2022.10019960","url":null,"abstract":"Higher-order side-channel attacks have been widely studied. In this paper, the principles and preprocessing methods of higher-order side-channel attacks are analyzed, and the essential consistency between higher-order side-channel attacks and higher-order fault attacks is defined at the algorithmic and physical levels. Higher-order leakage point in fault attacks and the possibility of higher-order attacks is confirmed, and several possible attack methods are proposed. Finally, the current problems of higher-order side-channel attacks are analyzed, and hot issues worth further research in the field of higher-order fault attacks, such as promoting first-order attack methods, constructing a unified physical leakage model, carrying out combined attacks, and designing a common defense strategy, are proposed.","PeriodicalId":423151,"journal":{"name":"2022 IEEE 5th Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC)","volume":"114 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127343504","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Research On Deep Learning Based End-To-End Wireless Communication System Technology","authors":"R. Lu, Han Jiang, Lin Hu","doi":"10.1109/IMCEC55388.2022.10019964","DOIUrl":"https://doi.org/10.1109/IMCEC55388.2022.10019964","url":null,"abstract":"As an important part of AI, deep learning is widely used in image classification and speech recognition. In particular, the application of deep neural networks with powerful digital signal processing capabilities in communication systems has become a research hotspot in recent years. As a common unsupervised learning model in deep learning, the autoencoder is similar to the traditional wireless communication system and can be used as a design scheme for the physical layer of the wireless communication system. From the perspective of “global optimization” of the wireless communication system, this paper is based on deep learning autoencoder and uses the MATLAB deep learning toolbox to design an end-to-end wireless communication system and conducts simulation under AWGN. The results show that the proposed system is similar to the traditional wireless communication system in terms of performance and has certain “generalization” capability for coding rate, EbNo and other parameters.","PeriodicalId":423151,"journal":{"name":"2022 IEEE 5th Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124953194","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Implementation and research of recommendation system based on trust mechanism","authors":"Jinhui Yang, Ting Shi, Shijin Li","doi":"10.1109/IMCEC55388.2022.10020104","DOIUrl":"https://doi.org/10.1109/IMCEC55388.2022.10020104","url":null,"abstract":"This paper analyzes and studies the traditional algorithm of recommendation system. At present, user based collaborative filtering algorithms and project-based collaborative filtering algorithms have been widely recognized, but they also have shortcomings. Such as the calculation error of user similarity is large, and there is a user preference problem in user rating. The collaborative filtering algorithm considers the relationship between project resources and user ratings, but does not consider the trust between users. To solve these problems, we changed the cosine similarity used in the traditional recommendation algorithm to the modified cosine similarity by comparing the similarity algorithm to avoid user preference problems. By using a hybrid collaborative filtering algorithm, combining user score similarity and user item attribute similarity, the nearest neighbor set is more accurate. .","PeriodicalId":423151,"journal":{"name":"2022 IEEE 5th Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125172767","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Research on Surface Defect Detection System of Brake Disc","authors":"Zheng Bin, Wang Cong, Li Xue","doi":"10.1109/IMCEC55388.2022.10019860","DOIUrl":"https://doi.org/10.1109/IMCEC55388.2022.10019860","url":null,"abstract":"Brake disc is one of the key parts in the automobile braking system, which is assembled together with the wheel hub. Although dry friction on the brake lining, it produces large friction force and reduces the wheel speed to achieve the effect of deceleration braking. When running at high speed, the emergency braking car bears large wear and tear, and its quality directly affects the braking performance and safety performance of the car. Therefore, this paper takes the automobile brake disc as the detection object and develops a brake disc surface defect detection and identification system based on machine vision combined with HALCON software to detect and identify the brake disc defects. According to the defects within the surface of the brake disc, the following work is done for the machine vision inspection system. To meet the requirements of detection accuracy and detect the errors in natural collection, a set of image acquisition platform is designed and it can realize X, Y and Z axis direction adjustment. A power regulator with adjustable brightness is provided based on the configuration of the light source. To solve the problem of image processing and analysis of brake disc surface defects, the image processing flow based on HALCON was designed. It includes gray-scale transformation, pretreatment, binarization, morphological processing, threshold segmentation, and other operations that were used to extract target features. The results show that the system has high detection accuracy and detection efficiency and can meet the requirements of practical production.","PeriodicalId":423151,"journal":{"name":"2022 IEEE 5th Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC)","volume":"136 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123256458","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Identification of binary file compilation information","authors":"Zulie Pan, Yintong Yan, Lu Yu, Taiyan Wang","doi":"10.1109/IMCEC55388.2022.10019958","DOIUrl":"https://doi.org/10.1109/IMCEC55388.2022.10019958","url":null,"abstract":"Cross-architecture binary code similarity detection technology has been widely used in vulnerability discovery, reverse engineering and patch detection. The identification of binary file compilation information is conducive to the improvement of the accuracy in binary code similarity detection. The compilation information of binary files includes compilation architecture, compiler, optimization option and obfuscation strategy. For the compilation architecture, we build a compiling architecture feature library based on the ELF header information of the binary file for identification; for the compiler, we use Linux system commands to identify; for the optimization option and obfuscation strategy, we extract 70 static features of binary file function-level assembly language and establish a genetic neural network model for identification. In addition, we set up five experimental tasks to learn more about the compilation architecture and compiler impact on model identification optimization options and obfuscation strategies. The final experimental results show that the accuracy of the binary file compilation information identification model designed by us is 100% for both compilation architectures and compilers identification, and the F-Score for optimization options identification can reach 89.46%. The F-Score for obfuscation strategies identification can reach 88.74%, and the F-Score for simultaneous identification of optimization options and obfuscation strategies can reach 84.28%, which is significantly better than previous works.","PeriodicalId":423151,"journal":{"name":"2022 IEEE 5th Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125343391","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}
Junting Yang, Chen Liang, Xiping Ma, Xiaoyang Dong, Yaxin Li
{"title":"Research on Reactive Power Optimization of High Proportion New Energy Access Distribution Network Based on NSGA-III Algorithm","authors":"Junting Yang, Chen Liang, Xiping Ma, Xiaoyang Dong, Yaxin Li","doi":"10.1109/IMCEC55388.2022.10020128","DOIUrl":"https://doi.org/10.1109/IMCEC55388.2022.10020128","url":null,"abstract":"A reactive power optimization model with the objective of minimizing voltage offset and network loss is established from a comprehensive consideration of the safety and stability of the operation of high-occupancy wind power and PV connected to the grid. MATLAB software is used to simulate and analyze the reactive power optimization of wind power and PV power plants connected to the IEEE33 node system, and the multi- NSGA-III algorithm is used to solve the reactive power optimization scheduling scheme. The NSGA-III algorithm based on the Pareto front solution can provide solutions for decision makers according to different demands, and simultaneously achieve the objectives of improving the voltage stability of the tested system and reducing the network loss, which verifies the effectiveness of the proposed optimization strategy.","PeriodicalId":423151,"journal":{"name":"2022 IEEE 5th Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125391137","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Research on Occlusion Face Detection Method in Complex Environment","authors":"Yong Luo","doi":"10.1109/IMCEC55388.2022.10020030","DOIUrl":"https://doi.org/10.1109/IMCEC55388.2022.10020030","url":null,"abstract":"In recent years, wearing masks has become a habit of some people. In this context, the requirements for masking face recognition technology are getting higher and higher, and people's requirements for identity recognition intelligence and security are getting higher and higher. In practical applications, the face images obtained by the face recognition system are easily blocked by other people or objects. These interference factors make the collected face data incomplete and affect the accuracy of the entire face recognition system. For example, Apple's FaceID is difficult to identify Faces with masks. Therefore, it is urgent to improve the effect of existing face recognition systems in occluding face recognition. At present, most of the mainstream face recognition methods are based on deep learning. This type of method uses large-scale face data sets for training and learning, and the deep learning method for occlusion face recognition is also based on this. However, this type of data set is very limited. Based on this paper It mainly uses a generated occlusion face data set, and adds the occlusion face data set in the real environment to train the Yolov5 deep learning model to improve the detection ability of identifying occluded faces. The experimental results show that in the real environment, the occluded face detection proposed in this paper has a high accuracy, close to 97%.","PeriodicalId":423151,"journal":{"name":"2022 IEEE 5th Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116116998","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}
Xiangyu Zhang, Jun Fang, Jingfan Zou, Quan Chen, Shuo Chen, Jianfeng Hong, Shuang Wang
{"title":"Coordinated Scheduling Algorithm for Grid-connected Photovoltaic Power Generation and Energy Storage Systems","authors":"Xiangyu Zhang, Jun Fang, Jingfan Zou, Quan Chen, Shuo Chen, Jianfeng Hong, Shuang Wang","doi":"10.1109/IMCEC55388.2022.10019973","DOIUrl":"https://doi.org/10.1109/IMCEC55388.2022.10019973","url":null,"abstract":"With the development of the economy, the utilization technology of clean energy has become the key technology to obtain maximum benefits. Among them, distributed photovoltaic power generation directly converts solar energy into electric energy, following the principles of adapting to local conditions, clean and efficient, and decentralized layout. Local solar energy can be made full use of. However, photovoltaic power generation is affected by many uncertain factors, and the power is not stable. In this paper, a coordinated scheduling model of distributed photovoltaic power generation and grid-connected energy storage is established, combined with the principle of time-of-use electricity price, with economic benefits as the objective function. A dynamic programming algorithm is proposed to solve the charging and discharging scheduling problem of energy storage, and an example is simulated, which has a good effect.","PeriodicalId":423151,"journal":{"name":"2022 IEEE 5th Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116386663","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}