{"title":"解决了摄像机陷阱视频流中鸟类检测与分类的生物多样性分析问题","authors":"Mikhail G. Dorrer, A. Alekhina","doi":"10.1051/e3sconf/202339003011","DOIUrl":null,"url":null,"abstract":"The work is devoted to solving the problem of assessing the comparative efficiency of several common architectures of convolutional neural networks for monitoring birds in a natural environment. The problem was solved by detecting birds recorded by video traps installed on feeders in several regions of Panama by different architectures. Then a comparison was made between the recognition quality metrics – IoU and mAP, and based on the values of the metrics, a conclusion was made about the effectiveness of the architectures. Experiments have shown that the YOLO architecture of the Tiny version with comparative modules wins in the accuracy table. In the future, it is planned to improve the application of neural network architectures by finalizing the dataset with the involvement of expert bird watchers and open ornithological ontologies.","PeriodicalId":11445,"journal":{"name":"E3S Web of Conferences","volume":"77 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":"{\"title\":\"Solving the problem of biodiversity analysis of bird detection and classification in the video stream of camera traps\",\"authors\":\"Mikhail G. Dorrer, A. Alekhina\",\"doi\":\"10.1051/e3sconf/202339003011\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The work is devoted to solving the problem of assessing the comparative efficiency of several common architectures of convolutional neural networks for monitoring birds in a natural environment. The problem was solved by detecting birds recorded by video traps installed on feeders in several regions of Panama by different architectures. Then a comparison was made between the recognition quality metrics – IoU and mAP, and based on the values of the metrics, a conclusion was made about the effectiveness of the architectures. Experiments have shown that the YOLO architecture of the Tiny version with comparative modules wins in the accuracy table. In the future, it is planned to improve the application of neural network architectures by finalizing the dataset with the involvement of expert bird watchers and open ornithological ontologies.\",\"PeriodicalId\":11445,\"journal\":{\"name\":\"E3S Web of Conferences\",\"volume\":\"77 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2023-01-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"E3S Web of Conferences\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1051/e3sconf/202339003011\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"E3S Web of Conferences","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1051/e3sconf/202339003011","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Solving the problem of biodiversity analysis of bird detection and classification in the video stream of camera traps
The work is devoted to solving the problem of assessing the comparative efficiency of several common architectures of convolutional neural networks for monitoring birds in a natural environment. The problem was solved by detecting birds recorded by video traps installed on feeders in several regions of Panama by different architectures. Then a comparison was made between the recognition quality metrics – IoU and mAP, and based on the values of the metrics, a conclusion was made about the effectiveness of the architectures. Experiments have shown that the YOLO architecture of the Tiny version with comparative modules wins in the accuracy table. In the future, it is planned to improve the application of neural network architectures by finalizing the dataset with the involvement of expert bird watchers and open ornithological ontologies.
期刊介绍:
E3S Web of Conferences is an Open Access publication series dedicated to archiving conference proceedings in all areas related to Environment, Energy and Earth Sciences. The journal covers the technological and scientific aspects as well as social and economic matters. Major disciplines include: soil sciences, hydrology, oceanography, climatology, geology, geography, energy engineering (production, distribution and storage), renewable energy, sustainable development, natural resources management… E3S Web of Conferences offers a wide range of services from the organization of the submission of conference proceedings to the worldwide dissemination of the conference papers. It provides an efficient archiving solution, ensuring maximum exposure and wide indexing of scientific conference proceedings. Proceedings are published under the scientific responsibility of the conference editors.