K. Gajewski, Witold Prusak, Jaroslaw Fafara, Aleksander Skrzypiec, T. Turlej
{"title":"ARTIFICIALLY AIDED FUNGI RECOGNITION USING CONVOLUTIONAL NEURAL NETWORKS","authors":"K. Gajewski, Witold Prusak, Jaroslaw Fafara, Aleksander Skrzypiec, T. Turlej","doi":"10.5593/sgem2022v/3.2/s14.33","DOIUrl":null,"url":null,"abstract":"This article presents the concept of using neural networks in the recognition of fungi for use in a mobile forest ecosystem inspection robot. There are many dependencies regarding the occurrence of fungi in the vicinity of specific tree species. The presence of some fungi may be the result of a developing tree disease. The possibility of quick recognition of the fungus species using an autonomous mobile robot will allow for faster detection and prevention of the disease in entire ecosystems. An attempt was made to use neural networks to improve the efficiency of recognizing a specific species of fungus. This paper presents a comparison between our network and the AlexNet method network (created by Alex Krizhevsky) [1] for fungal recognition. This system was designed so that created by our students' science club NewTech AGH mobile inspection robot \"RUMCAJS\" could map the fungal population over time. Based on the comparison of the neural networks used, the possibility of correct use of the proposed solution for the detection of fungi was shown, as well as a more effective method in this application was indicated. The proposed method can be successfully implemented for the inspection of ecosystems using autonomous robots.","PeriodicalId":331146,"journal":{"name":"SGEM International Multidisciplinary Scientific GeoConference� EXPO Proceedings","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2022-12-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"SGEM International Multidisciplinary Scientific GeoConference� EXPO Proceedings","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.5593/sgem2022v/3.2/s14.33","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
This article presents the concept of using neural networks in the recognition of fungi for use in a mobile forest ecosystem inspection robot. There are many dependencies regarding the occurrence of fungi in the vicinity of specific tree species. The presence of some fungi may be the result of a developing tree disease. The possibility of quick recognition of the fungus species using an autonomous mobile robot will allow for faster detection and prevention of the disease in entire ecosystems. An attempt was made to use neural networks to improve the efficiency of recognizing a specific species of fungus. This paper presents a comparison between our network and the AlexNet method network (created by Alex Krizhevsky) [1] for fungal recognition. This system was designed so that created by our students' science club NewTech AGH mobile inspection robot "RUMCAJS" could map the fungal population over time. Based on the comparison of the neural networks used, the possibility of correct use of the proposed solution for the detection of fungi was shown, as well as a more effective method in this application was indicated. The proposed method can be successfully implemented for the inspection of ecosystems using autonomous robots.