{"title":"A Robust Human-Machine Hybrid Visual Tracking System","authors":"Tongtong Zhou, Yadong Liu","doi":"10.1109/ICAICE54393.2021.00146","DOIUrl":"https://doi.org/10.1109/ICAICE54393.2021.00146","url":null,"abstract":"Visual object tracking is a basic task in computer vision, which has been widely used in intelligent transportation, autonomous driving, security systems, military reconnaissance and other fields. Most studies in visual object tracking assume that the target changes smoothly and the target will not disappear for a long time. However, in practical applications, challenges such as complete occlusion, rapid movement and target appearance dramatic change, make it very difficult to track the target consistently for a long time. In this work, we propose a human-machine collaboration method to cope with such challenges. We hope to build a tracking framework that combines the powerful tracking capabilities of human vision with the state-of-the-art tracking methods, so as to get a robust visual tracking system. Humans participation can effectively improve the accuracy and robustness of tracking. In the process of human participation, the tracker can also improve its discrimination ability by recognizing the target of human interest. Compared with the state-of-the-art trackers, our method achieves higher performance on a fairly complex experimental dataset.","PeriodicalId":388444,"journal":{"name":"2021 2nd International Conference on Artificial Intelligence and Computer Engineering (ICAICE)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123863510","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":"Application of Stacking ensemble learning in option implied volatility","authors":"Q. Zhang, Jiapeng Liu, D. Tian, Han Yue","doi":"10.1109/ICAICE54393.2021.00123","DOIUrl":"https://doi.org/10.1109/ICAICE54393.2021.00123","url":null,"abstract":"The change of option implied volatility is the key problem of option risk early warning management. Based on Stacking ensemble learning in machine learning, this paper takes the implied volatility of SSE 50ETF call option as the object, and takes 30 indicators as the characteristics of implied volatility prediction. Random forest (RF), Adaptive Boosting (AdaBoost), Gradient Boosting Decision Tree (GBDT) and Extreme Gradient Boosting (XGBoost), which pertain to the tree-based algorithms, are stacked as base classifiers in the first layer. Combining with cross validation, the output of base classifier is used as the input training of meta classifiers Logistic Regression (LR), Support Vector Machines (SVM) and K-Nearest Neighbor (KNN). The results show that Stacking ensemble learning is better than tree-based algorithms in the prediction of the trend of implied volatility of call options, and the average prediction accuracy can reach 78.09%. At the same time, the Stacking ensemble learning is used for back testing, and the average cumulative return is 70%. The Stacking ensemble learning proposed in this paper improves the prediction performance of implied volatility of call options and provides a new method for investors' investment decision-making.","PeriodicalId":388444,"journal":{"name":"2021 2nd International Conference on Artificial Intelligence and Computer Engineering (ICAICE)","volume":"117 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117301872","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":"An Image-based Measuring Technique for the Prediction of Human Body Size","authors":"Xin Pei, Sitong Wu, Jing Zhao, C. Lin","doi":"10.1109/ICAICE54393.2021.00153","DOIUrl":"https://doi.org/10.1109/ICAICE54393.2021.00153","url":null,"abstract":"To achieve high prediction accuracy of human body keeps an open issue for decades of years, especially when COVID comes and online retail becomes the major consumption channels. The body measurement is the key to solve cloth matching and recommendation in clothing e-commerce. This paper proposes a practical framework of image-based body measurement, by only taking the user's front and side photos. This framework does not require pure background or precise standing position, and supports manual modification of the measurement results. The framework takes people's height, weight and gender as params to initialize a common body size set, and corrects each part of the set by analyzing the body proportion via the front and side images. The prediction accuracy was tested with the 50 digital models and 10 real people. Results showed that the circumference sizes such as chest, waist, hips, have errors less then 5%, while the length sizes such as arm, leg approach to actual length on net body models. For real people, the errors depend on the wearing clothes. In addition to high accuracy, the method has a rapid process speed, reaching 19QPS on a NVIDIA RTX5000 GPU server.","PeriodicalId":388444,"journal":{"name":"2021 2nd International Conference on Artificial Intelligence and Computer Engineering (ICAICE)","volume":"335 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124709850","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":"Classification of Histopathologic Images of Breast Cancer by Multi-teacher Small-sample Knowledge Distillation","authors":"Leiqi Wang, Hui-juan Lu","doi":"10.1109/ICAICE54393.2021.00127","DOIUrl":"https://doi.org/10.1109/ICAICE54393.2021.00127","url":null,"abstract":"Model fusion can effectively improve the effect of model prediction, but it will bring about an increase in time. In this paper, the dual-stage progressive knowledge distillation is improved in combination with multi-teacher knowledge distillation technology. A simple and effective multi-teacher's Softtarget integration method is proposed in multi-teacher network knowledge distillation. Improve the guiding role of excellent models in knowledge distillation. Dual-stage progressive knowledge distillation is a method for small sample knowledge distillation. A progressive network grafting method is used to realize knowledge distillation in a small sample environment. In the first step, the student blocks are grafted one by one onto the teacher network and intertwined with other teacher blocks for training, and the training process only updates the parameters of the grafted blocks. In the second step, the trained student blocks are grafted onto the teacher network in turn, so that the learned student blocks adapt to each other and finally replace the teacher network to obtain a lighter network structure. Using Softtarget acquired by this method in Dual-stage progressive knowledge distillation instead of Hardtarget training, excellent results were obtained on BreakHis data sets.","PeriodicalId":388444,"journal":{"name":"2021 2nd International Conference on Artificial Intelligence and Computer Engineering (ICAICE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128804826","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":"Intelligent Repair Method of Old Movie Speckle Noise Based on AI Deep Learning","authors":"Yu Zheng, Jiandong Cui, H. Zhong, Dong-Hyuk Choi","doi":"10.1109/icaice54393.2021.00019","DOIUrl":"https://doi.org/10.1109/icaice54393.2021.00019","url":null,"abstract":"Speckle noise in old movies is caused by erasure or other reasons, which affects the quality of video images. Therefore, an intelligent repair method of speckle noise in old movies based on AI deep learning is proposed. Based on the analysis of the characteristics of speckle noise in old films, the image noise is filtered by bilateral filtering, and the image features are obtained by combining a convolution neural network to obtain the perceptual loss data. On this basis, the jump connection is added to the deep learning network structure of image restoration. Taking the loss function as the training object, the high-quality restoration of speckle noise is realized by optimizing the loss function. The test results show that the design method can ensure that the average PSNR value of the repaired image can reach more than 40 under lower and shorter training iterations, and the effect is obvious.","PeriodicalId":388444,"journal":{"name":"2021 2nd International Conference on Artificial Intelligence and Computer Engineering (ICAICE)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127628419","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":"Multi-Task Medical Image-to-Images Translation using Transformer for Chest X-Ray Radiography","authors":"Jingyu Xie","doi":"10.1109/icaice54393.2021.00139","DOIUrl":"https://doi.org/10.1109/icaice54393.2021.00139","url":null,"abstract":"Chest X-ray is one of the main methods for screening chest diseases, which has the characteristics of low radiation dose, fast imaging and low cost. In order to better assist doctors in disease diagnosis, usually the X-ray bone suppression and organ segmentation are performed. Many research progress has been made in this field, but the accuracy of the above two tasks is still limited due to the inherent characteristics of medical images. Firstly, the shape of organs of different individuals varies greatly.So there are inevitable segmentation errors if the overall shape is not perceived. Generally, the boundary of organs is fuzzy, so it is prone to misclassification near the boundary. In addition, existing bone suppression methods still can't completely remove bone shadows. In this paper, we propose a deep learning model whose overall architecture is designed based on the pix2pix network. This model generates both bone suppression images and organ segmentation images.Aiming at the above three issues, we make some improvements. We innovatively use the Transformer structure to enhance the attention to the global context and enhance the perception of the overall shape of the organ in the feature extraction process. Also, we design a new loss function, which gives larger weight to the position error near the organ boundary in the later stage of network training. This loss function pays more attention to edge information and helps determine the position of organ boundaries. We evaluate the effectiveness of the model on the X-ray image dataset, and compare it with the latest algorithm comprehensively. We also evaluate the effectiveness of our improvements through ablation study which shows that our improvements are effective.","PeriodicalId":388444,"journal":{"name":"2021 2nd International Conference on Artificial Intelligence and Computer Engineering (ICAICE)","volume":"200 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121318854","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 Evacuation of People in Quasi-Rectangular Subway Tunnel Fire Environment Based on Pathfinder Software","authors":"Peng Wang, Wanfu Liu, Wuqin Qi, Xingbo Li, Shiwei Fang, Zhenfeng Qi","doi":"10.1109/ICAICE54393.2021.00169","DOIUrl":"https://doi.org/10.1109/ICAICE54393.2021.00169","url":null,"abstract":"Based on the evacuation theory, this research proposes a calculation method of evacuation reliability suitable for subway tunnel fire. Use FDS to simulate the fire environment and Pathfinder software to simulate the evacuation plan. The results show that: compared with natural ventilation, the opening of longitudinal ventilation can ensure a safe evacuation environment for occupants upwind from the fire source, and effectively reduce the CO concentration in the tunnel, but shorten the available safety egress time. The evacuation plan is not feasible when the load factor is 100%. When only the side door of the second carriage is opened, the evacuation time of personnel toward the two-sided platform must be the shortest. But the plan is still not feasible. The evacuation plan is feasible when the load factor is 55%. Compared with the evacuation plan toward the platforms on both sides, the evacuation plan toward the platform on the fresh air side has higher evacuation reliability.","PeriodicalId":388444,"journal":{"name":"2021 2nd International Conference on Artificial Intelligence and Computer Engineering (ICAICE)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116448677","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":"A configuration analysis method of high-speed railway train-set equipment based on handle","authors":"P. Sun, Hui Wang","doi":"10.1109/ICAICE54393.2021.00164","DOIUrl":"https://doi.org/10.1109/ICAICE54393.2021.00164","url":null,"abstract":"The industry identification analysis system is the primary technical problem in building the data acquisition capability of the railway Internet of things, and the configuration analysis of complex equipment assets is the difficulty in application. According to the configuration analysis management requirements of complex equipment of the high-speed railway train-set, combined with the technical information management of the high-speed railway train-set, the corresponding handle structure of the high-speed railway train-set is given, and the distributed processing algorithm of configuration analysis of the high-speed railway train-set is designed. The high-speed analysis ability, stability and reliability of the system and method described in this paper are verified by experiments, which can meet the practical requirements of railway identification application innovation.","PeriodicalId":388444,"journal":{"name":"2021 2nd International Conference on Artificial Intelligence and Computer Engineering (ICAICE)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124100961","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":"Analysis of Library loan records based on hybrid Apriori- Genetic algorithm","authors":"Yu-Chieh Shen, Jingwen Wang, Xiaobing Yang, Yunkai Lv","doi":"10.1109/icaice54393.2021.00012","DOIUrl":"https://doi.org/10.1109/icaice54393.2021.00012","url":null,"abstract":"The library database system accumulates lots of information resources in the long-term service process. Apriori algorithm mining association rules has low time and space efficiency, resulting in a large number of association rules. Using hybrid Apriori-Genetic algorithm to analyze records can be effective. Firstly, the records are cleaned and preprocessed, and the algorithm is used to analyze records, so as to find the association rules between different types of books borrowed, and then the two algorithms are compared in the the number of association rules and Algorithm time complexity.","PeriodicalId":388444,"journal":{"name":"2021 2nd International Conference on Artificial Intelligence and Computer Engineering (ICAICE)","volume":"120 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126006793","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":"The Tracking control of Autonomous Underwater Vehicle Based on FA - Model Predictive Control","authors":"W. Zhou, Daqi Zhu, Xiaotong Yan","doi":"10.1109/ICAICE54393.2021.00091","DOIUrl":"https://doi.org/10.1109/ICAICE54393.2021.00091","url":null,"abstract":"Aiming at the trajectory tracking control of the autonomous underwater vehicle (AUV), a new model predictive control (MPC) method based on the firefly algorithm optimization (FA) is proposed. This article firstly gives the concept of trajectory tracking and model predictive control, and then uses FA-MPC to achieve tracking control. The firefly algorithm is used to solve the optimization problem of minimizing the objective function under the condition of satisfying control amount constraints and control incremental constraints. The simulation experiment results illustrate that FA-MPC can effectively solve the speed jump problem caused by the use of backstepping control, and FA-MPC is stable and feasible in the trajectory tracking problem.","PeriodicalId":388444,"journal":{"name":"2021 2nd International Conference on Artificial Intelligence and Computer Engineering (ICAICE)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126674714","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}