2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML)最新文献_第2页

Object-Based Perspective Transformation Data Augmentation for Object Detection 面向目标检测的基于对象的透视变换数据增强

2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML) Pub Date : 2022-06-01 DOI: 10.1109/FAIML57028.2022.00043

Zibo Nie, Jianjun Cao, Nianfeng Weng, Xu Yu, Mengda Wang

{"title":"Object-Based Perspective Transformation Data Augmentation for Object Detection","authors":"Zibo Nie, Jianjun Cao, Nianfeng Weng, Xu Yu, Mengda Wang","doi":"10.1109/FAIML57028.2022.00043","DOIUrl":"https://doi.org/10.1109/FAIML57028.2022.00043","url":null,"abstract":"Perspective transformation can mimic the perspective phenomenon in captured images, while few researches focused on utilizing perspective transformation to augment image data. To well exploit its potential in data augmentation, a novel object-based perspective transformation data augmentation framework is proposed in this paper. First, this framework automatically cuts out objects from images based on bounding boxes provided by corresponding annotations. Next, a simulated random perspective transformation is performed on only those cut objects rather than the whole images. The final step consists of the combination of transformed objects and backgrounds, together with calculation of new annotation. Our framework can generate new images by mimicking the perspective phenomenon caused by different orientations of objects and can also introduce new backgrounds at the same time, hence the further augmentation towards insufficient or imbalanced image data without requirements of additional manual manipulations. Experiments are carried out on the Pascal VOC2007 datasets and the results show the effectiveness of our framework on insufficient or imbalanced datasets.","PeriodicalId":307172,"journal":{"name":"2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML)","volume":"74 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116310721","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}

引用次数: 1

Short-term Wireless Load Indicator Forecasting Method Based On Multi-Model Fusion 基于多模型融合的短期无线负荷指标预测方法

2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML) Pub Date : 2022-06-01 DOI: 10.1109/FAIML57028.2022.00014

Wang Xi, Lin Xiaojun

引用次数: 0

Cybersecurity Readiness for Automated Vehicles 自动驾驶汽车的网络安全准备

2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML) Pub Date : 2022-06-01 DOI: 10.1109/FAIML57028.2022.00012

S. Khan, N. Shiwakoti, P. Stasinopoulos, M. Warren

{"title":"Cybersecurity Readiness for Automated Vehicles","authors":"S. Khan, N. Shiwakoti, P. Stasinopoulos, M. Warren","doi":"10.1109/FAIML57028.2022.00012","DOIUrl":"https://doi.org/10.1109/FAIML57028.2022.00012","url":null,"abstract":"Autonomous Vehicle (AV) is a rapidly evolving mobility technology with the potential to drastically alter the future of transportation. Despite the plethora of potential benefits that have prompted their eventual introduction, AVs may also be a source of unprecedented disruption for future travel eco-systems due to their vulnerability to cyber-threats. In this context, this work assesses AVs' cybersecurity readiness. It establishes a Causal Loop Diagram (CLD) based on the System Dynamics approach: a powerful technique inferred from system theory, which can synthesise the behaviour of complicated AV systems. Based on the CLD model, three feedback loops and a system archetype “Fixes-That-Fail” are envisioned, in which the growth in hacker capability, an unforeseen result of technology innovation, demands constant mitigation efforts. The most challenging aspect of this context is determining the trade-off between five components: i) the natural growth of AV technology; ii) stakeholders (communication service providers, road operators, automakers, AV consumers, repairers, and the general public) access to AV technology; iii) the measures to limit hackers' access to AV technology; iv) a pervasive dynamic strategy for circumventing hacker amplification; and v) the efficient usage of AV operating logfiles.","PeriodicalId":307172,"journal":{"name":"2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126045251","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}

引用次数: 0

Text Summarization Clustered Transformer (TSCT) 文本摘要集群转换器(TSCT)

2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML) Pub Date : 2022-06-01 DOI: 10.1109/FAIML57028.2022.00041

R. D. Ahmed, M. Abdulhak, Omar Hesham ELNabrawy

{"title":"Text Summarization Clustered Transformer (TSCT)","authors":"R. D. Ahmed, M. Abdulhak, Omar Hesham ELNabrawy","doi":"10.1109/FAIML57028.2022.00041","DOIUrl":"https://doi.org/10.1109/FAIML57028.2022.00041","url":null,"abstract":"Natural language processing has recently had a considerable reputation due to the quick increase in online and offline data worldwide. The Extractive text summarization grabs the sentence from the corpus using salient related information to produce a concise summary. However, most existing approach to extracting sentence feature engineering has not utilized related contextual information and relation among the sentence. We present clustered Transformer models to mitigate this issue, namely Text summarization using clustered Transformer models. Our proposal has the highest benefit. The utility of our frame-work working on contextual representation is to grab various linguistic context information. We also use surface features to improve our understanding of word and sentence elements. Another utility is that the hierarchical attention mechanism can capture the contextual relation from the word and sentence levels using the transform model. Also, we added clustering after the transformer model to capture the most similar sentence to improve the attentive quality for producing the extractive text summarization.","PeriodicalId":307172,"journal":{"name":"2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML)","volume":"58-60 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123123987","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}

引用次数: 0

Optimization of Cross-border E-commerce Logistics Distribution Network Based on Genetic Neural Network 基于遗传神经网络的跨境电子商务物流配送网络优化

2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML) Pub Date : 2022-06-01 DOI: 10.1109/FAIML57028.2022.00033

Xue-qin Li, B. Wan

{"title":"Optimization of Cross-border E-commerce Logistics Distribution Network Based on Genetic Neural Network","authors":"Xue-qin Li, B. Wan","doi":"10.1109/FAIML57028.2022.00033","DOIUrl":"https://doi.org/10.1109/FAIML57028.2022.00033","url":null,"abstract":"This paper discusses the optimization of cross-border e-commerce logistics distribution network based on genetic neural network. Based on the logistics preferences of different types of customers (that is, high timeliness or low cost of distribution), the design problem of e-commerce logistics distribution network is modeled as a multi-objective optimization problem. The GNN algorithm is improved to solve the multi-objective optimization problem efficiently. Cross-border electronic commerce logistics is an important part of the international trade process. The logistics mode of cross-border e-commerce shortens the value chain. It accelerates the speed of international logistics, but it also complicates the research of cross-border logistics networks. In the aspect of Cross-border electronic commerce logistics distribution optimization: combining the characteristics of Cross-border electronic commerce logistics distribution, establish an optimization model with cost as the goal. Through standard test examples, it is found that the performance of the algorithm needs to be strengthened. Based on analyzing the limitations of the algorithm, the performance of the algorithm is improved, and the validity of the model and the algorithm for Cross-border electronic commerce logistics distribution optimization is verified by numerical examples and actual case data. To solve this complex and dynamic multi-dimensional objective optimization problem, this paper intends to apply genetic neural network to distribution, which logistics cost, customer satisfaction, logistics time cost and other factors.","PeriodicalId":307172,"journal":{"name":"2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML)","volume":"185 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116503717","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}

引用次数: 0

The Impact of Artificial Intelligence Painting on Contemporary Art From Disco Diffusion's Painting Creation Experiment 从Disco Diffusion的绘画创作实验看人工智能绘画对当代艺术的影响

2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML) Pub Date : 2022-06-01 DOI: 10.1109/FAIML57028.2022.00020

Lu Li

引用次数: 2

Algorithm of Basketball Posture Motion Feature Extraction Based on Image Processing Technology 基于图像处理技术的篮球姿态运动特征提取算法

2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML) Pub Date : 2022-06-01 DOI: 10.1109/FAIML57028.2022.00049

Zaima Lu

{"title":"Algorithm of Basketball Posture Motion Feature Extraction Based on Image Processing Technology","authors":"Zaima Lu","doi":"10.1109/FAIML57028.2022.00049","DOIUrl":"https://doi.org/10.1109/FAIML57028.2022.00049","url":null,"abstract":"In the field of basketball, the existing training concept is based on the simulated observation and personal experience of the coaches, and it is subject to subjective judgment of inadequacies. Using image technology to train athletes is mainly to help coaches make decisions and improve the strength of athletes through the identification and recognition of athletes' states and transfer characteristics. The purpose of this paper is to study the basketball pose motion feature extraction algorithm based on image processing technology. This paper firstly builds the basketball posture model, introduces the feature extraction and selection in basketball posture, analyzes the application of image processing algorithm in basketball posture movement feature extraction, and mainly uses filtering algorithm to denoise the moving image. The algorithm in this paper mainly selects the normalized data processing algorithm to perform data induction processing on the feature data extracted in this paper. Experiments show that after image processing technology, the recognition rate of motion features by BP neural network is the highest, and the average recognition rate reaches 96%, which can effectively recognize the motion features of basketball posture.","PeriodicalId":307172,"journal":{"name":"2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117149052","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}

引用次数: 0

System Design of Ground Test Equipment for Star Sensor 星敏感器地面试验设备系统设计

2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML) Pub Date : 2022-06-01 DOI: 10.1109/FAIML57028.2022.00026

Weikang Si, Hengguang Zhang, Ruke Yang, Libin Li, Qingzheng Song, Luwei Yu

{"title":"System Design of Ground Test Equipment for Star Sensor","authors":"Weikang Si, Hengguang Zhang, Ruke Yang, Libin Li, Qingzheng Song, Luwei Yu","doi":"10.1109/FAIML57028.2022.00026","DOIUrl":"https://doi.org/10.1109/FAIML57028.2022.00026","url":null,"abstract":"With the accumulation of high-precision star sensor models year by year, testers have put forward higher requirements for the low degree of automation of the star sensor ground test system, high maintenance costs, and difficulty in centralized management. In this context, it is very important to develop a simple and easy-to-use automated test system for star sensors. This project uses a standard Ethernet interface and a custom communication protocol to study an automated test system to improve test efficiency and quality. Through the python+Flask framework, the React technology of the web front-end and Webpack packaging technology are specific development methods, the Postgresql database storage technology is used, and the front-end interaction method is used to complete the development. The later development efficiency is high, and the maintenance cost is low. This article specifically introduces the principle of the system, functional modules and the design of the database. The results of the system use show that the project basically reached the original design goals and met the basic requirements of the testers. The system has the characteristics of reasonable structure, friendly interface, and easy maintenance.","PeriodicalId":307172,"journal":{"name":"2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131389171","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}

引用次数: 0

Automated segmentation of glands to facilitate quantitative analysis in CD138 whole slide images using a KNet deep learning framework 使用KNet深度学习框架自动分割腺体以促进CD138整张幻灯片图像的定量分析

2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML) Pub Date : 2022-06-01 DOI: 10.1109/FAIML57028.2022.00044

Shun Zou, Feifan Liao

{"title":"Automated segmentation of glands to facilitate quantitative analysis in CD138 whole slide images using a KNet deep learning framework","authors":"Shun Zou, Feifan Liao","doi":"10.1109/FAIML57028.2022.00044","DOIUrl":"https://doi.org/10.1109/FAIML57028.2022.00044","url":null,"abstract":"Segmentation of glands in immunohistochemical (IHC) images is the top priority for automated evaluation of CD138 positive cells, which could exclude irrelevant regions and improve accuracy and effectiveness. In this paper we propose a novel patch-based pipeline with an integrated KNet-like deep learning framework to perform automated gland segmentation for CD138 whole slide images. The patch-based pipeline is composed of patch decomposition, tissue patch extraction, gland segmentation, patch merging, and gland mask padding. The KNet deep learning framework introduces SwinTransformer to extract multiscale patch features, and integrate Uper Net as basic semantic kernels into KNet framework to refine the gland segmentation results. The integrated framework is trained in an end-to-end way with a weighted cross-entropy loss and dice loss. The experimental results show that our proposed framework achieves state-of-the-art performance, and could produce accurate gland masks for real whole slide images, which lays a solid foundation for automated quantitative analysis of CD138 images.","PeriodicalId":307172,"journal":{"name":"2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125328106","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}

引用次数: 0

Student Management Information System Based on Data mining 基于数据挖掘的学生管理信息系统

2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML) Pub Date : 2022-06-01 DOI: 10.1109/FAIML57028.2022.00024

Xinwen Li

{"title":"Student Management Information System Based on Data mining","authors":"Xinwen Li","doi":"10.1109/FAIML57028.2022.00024","DOIUrl":"https://doi.org/10.1109/FAIML57028.2022.00024","url":null,"abstract":"Colleges and universities are places for cultivating talents for the society. How to build a high-quality and high-culture college student team has become the focus of student management in colleges and universities. The basis of managing students lies in the full understanding and understanding of students, although the current college student information system records the information of enrolled students, but there are many drawbacks such as a large number of data, a simple management platform and a lack of intuitive embodiment, in order to solve this problem, give full play to the advantages of massive student information, and provide more meaningful decision-making basis for managers. Based on the data mining algorithm, this paper designs and researches the student management information system. This paper first describes the background significance of the student management information system and the current research status of the system, and builds the system framework according to the related technologies of Data mining algorithms. After the system is completed, the system is debugged. The debugging results can accurately reflect the feasibility of the system.","PeriodicalId":307172,"journal":{"name":"2022 International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML)","volume":"305 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122316910","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}

引用次数: 0