卷 6 ， 期 20 ， 2023
- 摘要：Aiming at the difficulty of feature extraction caused by the drastic changes in object direction, scale and aspect ratio, and the sometimes dense and sometimes sparse arrangement in remote sensing images, and the boundary discontinuity problem in the regression-based rotating detector, we propose the following solutions: Firstly, the YOLOv7 detector is used for feature extraction and fusion. Secondly, in combination with densely coded label (DCL), angle-classification codecs can be constructed to realize rotation detection. Then, the adaptive sample matching strategy of SimOTA is improved to R-SimOTA, and the accuracy of sample allocation is improved by adding the angle classification loss guidance to the cost function. Finally, angle distance and aspect ratio sensitive weighting (ADARSW) is introduced by adding angle classification loss (DCL Loss) and weight (theta) into the loss function. Guide the model to converge and fit the correct angle prediction value. A large number of experiments and visual analysis have been carried out on a large remote sensing image dataset DOTA, and the results show that the scheme is effective.关键词：object detection;remote sensing image;rotation detection;dense coding label;Angle classification;R-SimOTA adaptive sample matching149|318|0更新时间：2025-03-26
- 摘要：The stability and reliability of the power communication network has a greater impact on the development of society, so it needs technological innovation to maintain the power communication network and ensure the stability of its communication quality. In order to solve the drawbacks of power communication network operation and maintenance of many points and lines, it is necessary to join the advanced big data technology for its operation and management, so as to guarantee the normal operation of power communication network. Therefore, the study of the important role of electric power communication network under big data, and the analysis of the current status of the application of big data in the current electric power communication network, put forward the operation and maintenance strategy of electric power communication network under big data, and finally analyze the operation practice. Indicating that power companies need to strengthen the construction of power communication networks under big data, actively integrate, and achieve cooperation in more fields, truly promoting the establishment and development of power communication networks.关键词：big data;Electric power communication network;data security;Operation and maintenance;strategy37|19|0更新时间：2025-03-26
- 摘要：To construct a survival prediction model for pediatric patients undergoing bone marrow transplantation. A prediction model was built based on the Kaggle dataset for children's bone marrow transplantation using a PSO-BP (Particle Swarm Optimized Back Propagation) neural network. The prediction model has an AUC (Area Under the Curve) value of 0.98, indicating excellent performance of the algorithm model. Feature importance analysis revealed that survival time, recurrence, and CD34+ cell dose per kilogram of recipient weight are the most important factors affecting the survival prognosis of pediatric bone marrow transplantation patients.关键词：Particle Swarm Optimization;BP Neural Network;Bone Marrow Transplantation;Survival Prediction;Pediatric Patients69|73|0更新时间：2025-03-26
- 摘要：The development of data security industry is not only an important foundation for the construction of national data security governance system, but also an important guarantee for the construction of data basic system and the release of the potential of data elements. As the top-level design of the industry is gradually clarified, policy support is continuously enhanced, market vitality is rapidly burst, and the ecological environment is continuously optimized, the overall development potential of the industry is huge and the prospect is broad. This paper reviews the development trend of data security industry at home and abroad, summarizes the current situation and opportunities of China's data security industry in terms of government policies, product and service development, and industrial ecological construction, and puts forward relevant suggestions to promote the high-quality development of data security industry.关键词：data;data security;data security industry98|79|0更新时间：2025-03-26
RESEARCH PAPER
- 摘要：In the current context of environmental changes, wetlands play an increasingly important role, and their position in national ecological maintenance, ensuring ecological security, promoting ecological civilization construction, and implementing sustainable development strategies is becoming increasingly prominent. Their influence and value are constantly increasing. At present, the Nansihu Nature Reserve mainly relies on traditional manual supervision, which is difficult, inefficient, and costly. There is an urgent need for effective data fusion, management, and analysis planning tools. With the rapid development of computer technology, network technology, video sensing and monitoring technology, remote sensing technology, and virtual reality technology, combined with high-end hardware equipment and information management software, linking multi-source data through a three-dimensional geographic information platform can effectively improve the management and decision-making level of protected areas. Through practical applications, it has been shown that the deployment of the Smart Nansihu Information System effectively solves the problems of unclear, non real-time, non intuitive, and non propaganda in the current regulatory process of the protected area, providing scientific and efficient support for the management of the protected area.关键词：informatization;wetland protection;one map76|106|0更新时间：2025-03-26
CORRESPONDENCE
- 摘要：Since OpenAI launched the ChatGPT big model product at the end of 2022, generative artificial intelligence has sparked a wave of innovation worldwide. With the joint efforts of the scientific research and industry sectors, China’s generative artificial intelligence innovation is active and showing a good trend of vigorous development. This article analyzes and summarizes the product innovation, innovation ecology, and application scenario innovation of China’s generative artificial intelligence development. Through enterprise research and expert discussions, it determines the technological gap between China and internationally leading products, and deeply analyzes the problems faced by domestic generative artificial intelligence, including difficulties in breaking through technological barriers, lack of high-quality Chinese data, lack of industry cooperation, and lack of evaluation standards, propose relevant countermeasures and suggestions to provide reference for the decision-making of the party and the government.关键词：generative artificial intelligence;large model;application scenarios;evaluating131|639|0更新时间：2025-03-26
- 摘要：The drone self-organizing network has an extremely wide range of applications in the fields of commerce, scientific research, military, etc. In order to ensure the reliability of network communication services, there are clear requirements for routing protocols. The use of artificial intelligence to optimize the routing of drone self-organizing networks is a hot topic in recent years. For this reason, this paper focuses on the drone Ad Hoc network routing algorithm based on artificial intelligence, analyzes, concludes and compares the routing protocols using heuristics and machine learning, describes its representative algorithms, summaries the advantages of the algorithm, and finally discusses future technical challenges.关键词：artificial intelligence;machine learning;deep reinforcement learning;drones;routing245|1511|0更新时间：2025-03-26
SURVEYS & TUTORIALS
- Technical support is provided by Beijing Founder electronics co., LTD 京ICP备09064830号-19
京公网安备11010802024621 - It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
- Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.
- Total Visits：68219 Visits Today：1
0