{"title":"Object Detection in Remote Sensing Images Based on Feature Fusion and Multi-Branch Attention","authors":"Li Zhou, Min Wang, Jianyu Chen","doi":"10.1109/AINIT59027.2023.10212672","DOIUrl":"https://doi.org/10.1109/AINIT59027.2023.10212672","url":null,"abstract":"Object detection in remote-sensing images is an important and challenging task. With the development of deep learning technology, the method based on convolutional neural network has made considerable progress. However, due to the problems of remote-sensing images, such as dense arrangement, arbitrary direction and complex background, traditional detection networks are difficult to use adequately the semantic information in images. We design a novel single-stage detector based on feature fusion and three-branch attention. The feature map extracted by the backbone network is fully fused with the semantic information of different levels through the balanced pyramid structure, and then the critical foreground features are captured through the angle parameters decoupled three-branch attention network to improve the detection performance. Experimental results show that our method achieves better detection performance than many state-of-the-art methods.","PeriodicalId":276778,"journal":{"name":"2023 4th International Seminar on Artificial Intelligence, Networking and Information Technology (AINIT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124404967","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":"Copyright Page","authors":"","doi":"10.1109/ainit59027.2023.10212520","DOIUrl":"https://doi.org/10.1109/ainit59027.2023.10212520","url":null,"abstract":"","PeriodicalId":276778,"journal":{"name":"2023 4th International Seminar on Artificial Intelligence, Networking and Information Technology (AINIT)","volume":"590 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116176564","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":"Measuring Sleep Stages and Quality Based on K-Means and Random Forest Algorithms","authors":"Jiahui Li, X. Ge, Lixiang Hu, Qiuhua Zhu, Zhiwen Zhang, Fuheng Lv","doi":"10.1109/AINIT59027.2023.10212725","DOIUrl":"https://doi.org/10.1109/AINIT59027.2023.10212725","url":null,"abstract":"The purpose of this study is to measure sleep quality based on k-means and random forest algorithms in a big data environment. Firstly, the sleeping data of users is clustered into different stages of sleep using the k-means algorithm. Then, the random forest algorithm is used to predict sleep health indicators based on the clustering results. Health prediction based on sleeping stage is of high accuracy and reliability. Our study provides a new method for users to better understand their sleep quality and health status.","PeriodicalId":276778,"journal":{"name":"2023 4th International Seminar on Artificial Intelligence, Networking and Information Technology (AINIT)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124493114","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 BA-RRT-Based Indoor Geomagnetic Positioning Algorithm","authors":"Yudi Sun, Hongfei Yang","doi":"10.1109/AINIT59027.2023.10212873","DOIUrl":"https://doi.org/10.1109/AINIT59027.2023.10212873","url":null,"abstract":"With the development of the navigation technology, the geomagnetic positioning method is widely used due to its superior characteristics, such as the non-accumulating error, high positioning accuracy, and all-weather applications. Currently, most of the geomagnetic positioning methods need to be combined with external sensors to obtain positioning results, which leads to the limitation of the application environment of traditional geomagnetic positioning methods according to their combined sensors, so it is necessary to implement independent geomagnetic positioning. However, without external sensors providing path information, the process of geomagnetic matching will be more complex, making it more difficult to locate. To solve this problem, a geomagnetic independent positioning method based on the Bat Algorithm combined with the improved Rapidly-exploring Random Tree (BA-RRT) algorithm is proposed in this paper, which can locate with geomagnetic measurement sequence and a priori geomagnetic map in the absence of path information. Each bat position in the Bat Algorithm represents the path starting point, the improved Rapidly-exploring Random Tree is used to match the geomagnetic sequences. The motion path with the best adaptation is obtained by iterative meritocracy, and the localization results are obtained. Positioning experiments were conducted by indoor measurement of geomagnetic data, and the localization accuracy exceeds 90% with accurate geomagnetic map and no obvious interference, verifying the effectiveness of BA-RRT. The method proposed in this paper can provide a new approach for future research on geomagnetic independent positioning.","PeriodicalId":276778,"journal":{"name":"2023 4th International Seminar on Artificial Intelligence, Networking and Information Technology (AINIT)","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125877001","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":"EF2Net: Better Extracting, Fusing and Focusing Text Features for Scene Text Detection","authors":"Xiangyang Qu, Chongyang Zhang","doi":"10.1109/AINIT59027.2023.10212684","DOIUrl":"https://doi.org/10.1109/AINIT59027.2023.10212684","url":null,"abstract":"Text detection in natural scene images is a chal-lenging task that requires localization and fitting of text regions. Currently, existing natural scene methods use fixed-size convolutional kernels to extract text instance features and have achieved good results. However, due to the extremely large aspect ratio of text regions in natural scenes, extracting features using fixed-size convolutional kernels introduces background noise, which affects the accuracy of text detection. In addition, complex backgrounds in natural scenes may cause text features in existing methods to be incorrectly detected as text, while small and ambiguous text may be missed in the detection. To address these challenges, first, we use a new backbone with multi-branch depth band convolution to better capture text features in large aspect ratios and multi-scale backgrounds. Then, we propose a novel FPN that can obtain detailed information and scale sequence features to enhance the feature information of small texts. Finally, we design a dynamic text detection head that combines a text detection head with three attention mechanisms. We perceive from three dimensions: scale, space, and channel, enhance multi-scale text region features, focus on foreground targets, and accurately locate text regions, finally achieving the effect of reducing false and missed detections. In conclusion, the method proposed in this paper achieves good performance in text detection tasks in natural scenes and solves some problems in existing methods. Experimental results show that our proposed model achieves a comprehensive surpass compared with the text detection baseline.","PeriodicalId":276778,"journal":{"name":"2023 4th International Seminar on Artificial Intelligence, Networking and Information Technology (AINIT)","volume":"49 5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126121393","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":"Relationship Prediction based Anomaly Detection in Heterogeneous Information Networks","authors":"Wenyu Chen","doi":"10.1109/AINIT59027.2023.10212939","DOIUrl":"https://doi.org/10.1109/AINIT59027.2023.10212939","url":null,"abstract":"In heterogeneous information networks, there are many different types of nodes and different types of edges. Some homogeneous anomaly detection methods can't be directly used in heterogeneous information networks. It is very meaningful and challenging to study anomaly detection in heterogeneous information networks. In this paper, we introduce a framework which can detect anomaly in heterogeneous information network. We design the Relationship Prediction Neural Network Model (RPNN) to predict the relationship between nodes to learn the representation vector. Then the representation vector of the node type that we focus on is applied to the anomaly detection algorithm for detection. The experiments conducted on two real-world datasets show that our proposed model is effective compared with the state-of-the-art methods.","PeriodicalId":276778,"journal":{"name":"2023 4th International Seminar on Artificial Intelligence, Networking and Information Technology (AINIT)","volume":"69 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126190941","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 Mobile Terminal Authentication Method Based on Certificateless Signature","authors":"Jie Shi, XiaoHu Chen, XinJian Li, Shijun Wang","doi":"10.1109/AINIT59027.2023.10212463","DOIUrl":"https://doi.org/10.1109/AINIT59027.2023.10212463","url":null,"abstract":"With the rapid development of information science and technology and the promotion of network applications, especially the development of mobile networks, intelligent mobile terminal, and mobile applications, mobile terminal should be applied in various aspects of life. However, due to the characteristics of mobile networks and terminal, various network attacks have caused increasingly serious information security issues. Therefore, researching secure access solutions for mobile terminal is also an urgent need. So, ensuring that confidential data is not compromised and achieving authentication of mobile access objects is also a top priority in secure access solutions. This article proposes a certificateless aggregation authentication scheme for mobile terminal, which can provide conditional privacy protection for mobile terminal and resist various attacks with strong security. The performance analysis of this scheme and similar schemes shows that this scheme has higher authentication efficiency.","PeriodicalId":276778,"journal":{"name":"2023 4th International Seminar on Artificial Intelligence, Networking and Information Technology (AINIT)","volume":"284 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127395075","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 new ciphertext based OT protocol in cloud computing","authors":"Xiaopeng Zhu, Yong Wu, Xiaodong Li, Jianyi Zhang","doi":"10.1109/AINIT59027.2023.10212517","DOIUrl":"https://doi.org/10.1109/AINIT59027.2023.10212517","url":null,"abstract":"Cloud computing is the on-demand availability of computer system resources, especially computing power. Cloud computing relies on sharing resources to achieve coherence, and it needs to communicate frequently with the user. In addition, the user's information must be kept secret from the cloud, and additional computing information in the cloud should also be kept private from the user. The problem can be solved with oblivious transfer. Oblivious transfer (OT) protocols are an essential part of secure multi-party computation in cryptography. However, the previous scheme of OT is inefficient. When messages need to be transmitted multiple times, it becomes challenging to use in protecting privacy data transfer for cloud computing. In this paper, we analyze the features of cloud computing and the former OT protocols based on public key encryption (PKE) and Diffie-Hellman (DH). We propose a new OT protocol strategy, which uses a function in the ciphertext. We map the obtained symmetric public key and reduce computation for the protocol based on DH. We provide an implementation on cloud computing and compare our protocols with well-known OTs. In cloud computing, it shows that the time for each OT is almost negligible. In addition, the experimental results show that our protocol is at least 24% to 40% more efficient than other protocols. On the premise of ensuring user privacy, our strategy can make cloud data transmission more efficient.","PeriodicalId":276778,"journal":{"name":"2023 4th International Seminar on Artificial Intelligence, Networking and Information Technology (AINIT)","volume":"443 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123052681","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 Load-Balanced Deployment Algorithm for Wireless Mesh Network Gateways in Open Campus areas using Improved Particle Swarm Optimization","authors":"Youwu Liu, R. Parthasarathy","doi":"10.1109/AINIT59027.2023.10212964","DOIUrl":"https://doi.org/10.1109/AINIT59027.2023.10212964","url":null,"abstract":"The gateway deployment of a Wireless Mesh Network (WMN) is an important indicator of its performance, and selecting and deploying gateway nodes in complex environments such as open areas on campuses is a crucial issue. In this regard, an improved algorithm based on firefly and particle swarm optimization is proposed. Firstly, the topology structure of the campus open area (COA) is minimized using a 3D-Tabu algorithm to reduce the network node numbers. Secondly, the global network is optimized using the Firefly algorithm in the multi-objective environment. Finally, the particle swarm optimization algorithm is used to minimize the number of gateways and optimize load balancing. Simulated experimental results show that the improved algorithm effectively optimizes load balancing and gateway numbers, and has a better balance between gateway numbers and load balancing compared to the two kinds of heuristic algorithms. The algorithm can effectively improve network throughput with a QOS (Quality of Service) guarantee.","PeriodicalId":276778,"journal":{"name":"2023 4th International Seminar on Artificial Intelligence, Networking and Information Technology (AINIT)","volume":"91 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126461636","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":"Case Retrieval Method for Crop Diseases and Pests Control Based on Knowledge Graph","authors":"Wenkang Tang, Han Wang, Jingwen Li","doi":"10.1109/AINIT59027.2023.10212860","DOIUrl":"https://doi.org/10.1109/AINIT59027.2023.10212860","url":null,"abstract":"Traditional case retrieval methods cannot reflect the internal connections between cases, resulting in inaccurate and comprehensive retrieval results. According to the characteristics of crop pest control cases, a case retrieval method combining Knowledge graph and BERT model is proposed to improve the retrieval effect. Comprehensively consider the relationship structure and entity attribute characteristics of the Knowledge graph of crop disease and pest control cases to conduct case retrieval, represent the crop disease and pest control cases in the form of triple groups and build a Knowledge graph. On the one hand, use the Jaccard similarity coefficient to calculate the relationship similarity of cases; On the other hand, the BERT model is used to vectorize attribute features and calculate case attribute similarity. Weighted sum of the two parts is used to obtain the total similarity of the case, and case retrieval is performed. Multiple experiments have verified the effectiveness of this method, and the case retrieval results are more accurate and comprehensive.","PeriodicalId":276778,"journal":{"name":"2023 4th International Seminar on Artificial Intelligence, Networking and Information Technology (AINIT)","volume":"76 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127552623","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}