{"title":"利用深度学习实例分割方法确定番茄潜叶蝇:利用深度学习实例分割方法确定番茄叶蝉:Tuta absoluta (Meyrick) (Lepidoptera: Gelechiidae) 的危害情况","authors":"Tahsin Uygun, Mehmet Metin Ozguven","doi":"10.1007/s00217-024-04516-w","DOIUrl":null,"url":null,"abstract":"<div><p>Pests significantly negatively affect product yield and quality in agricultural production. Agricultural producers may not accurately identify pests and signs of pest damage. Thus, incorrect or excessive insecticides may be used. Excessive use of insecticides not only causes human health and environmental pollution, but also increases input costs. Therefore, early detection and diagnosis of pests is extremely important. In this study, the effectiveness of the instance segmentation method, a deep learning-based method, was investigated for the early detection of the damage caused by the <i>T. absoluta</i> pest in the leaf part of the tomato plant under greenhouse conditions. An original dataset was created by acquiring 800 healthy and damaged images under greenhouse conditions. The acquired images were labelled as bounding box and automatically converted to a mask label with the Segment Anything Model (SAM) model. The created dataset was trained with YOLOv8(n/s/m/l/x)-Seg models. As a result of the training, the box performance of the proposed YOLOv8l-Seg model was measured as 0.924 in the mAP<sub>0.5</sub> metric. The YOLOv8l-Seg model mask values are, respectively: mAP<sub>0.5</sub>, mAP<sub>0.5–0.95</sub>, Precision, Recall showed the best performance with values of 0.935, 0.806, 0.956 and 0.859. Then, the YOLOv8l-Seg model, trained with different data input sizes, showed the best performance at 640 × 640 size and the lowest performance with a value of 0.699 in the mAP<sub>0.5</sub> metric in the 80 × 80 size. The same dataset was trained with YOLOv7, YOLOv5l, YOLACT and Mask R-CNN instance segmentation models and performance comparisons were made with the YOLOv8l-Seg model. As a result, it was determined that the model that best detected <i>T. absoluta</i> damage in tomato plants was the YOLOv8l-Seg model. The Mask R-CNN model showed the lowest performance with a metric of 0.806 mAP<sub>0.5</sub>. The results obtained from this study revealed that the proposed model and method can be used effectively in detecting the damage caused by the <i>T. absoluta</i> pest.</p></div>","PeriodicalId":549,"journal":{"name":"European Food Research and Technology","volume":null,"pages":null},"PeriodicalIF":3.0000,"publicationDate":"2024-04-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s00217-024-04516-w.pdf","citationCount":"0","resultStr":"{\"title\":\"Determination of tomato leafminer: Tuta absoluta (Meyrick) (Lepidoptera: Gelechiidae) damage on tomato using deep learning instance segmentation method\",\"authors\":\"Tahsin Uygun, Mehmet Metin Ozguven\",\"doi\":\"10.1007/s00217-024-04516-w\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><p>Pests significantly negatively affect product yield and quality in agricultural production. Agricultural producers may not accurately identify pests and signs of pest damage. Thus, incorrect or excessive insecticides may be used. Excessive use of insecticides not only causes human health and environmental pollution, but also increases input costs. Therefore, early detection and diagnosis of pests is extremely important. In this study, the effectiveness of the instance segmentation method, a deep learning-based method, was investigated for the early detection of the damage caused by the <i>T. absoluta</i> pest in the leaf part of the tomato plant under greenhouse conditions. An original dataset was created by acquiring 800 healthy and damaged images under greenhouse conditions. The acquired images were labelled as bounding box and automatically converted to a mask label with the Segment Anything Model (SAM) model. The created dataset was trained with YOLOv8(n/s/m/l/x)-Seg models. As a result of the training, the box performance of the proposed YOLOv8l-Seg model was measured as 0.924 in the mAP<sub>0.5</sub> metric. The YOLOv8l-Seg model mask values are, respectively: mAP<sub>0.5</sub>, mAP<sub>0.5–0.95</sub>, Precision, Recall showed the best performance with values of 0.935, 0.806, 0.956 and 0.859. Then, the YOLOv8l-Seg model, trained with different data input sizes, showed the best performance at 640 × 640 size and the lowest performance with a value of 0.699 in the mAP<sub>0.5</sub> metric in the 80 × 80 size. The same dataset was trained with YOLOv7, YOLOv5l, YOLACT and Mask R-CNN instance segmentation models and performance comparisons were made with the YOLOv8l-Seg model. As a result, it was determined that the model that best detected <i>T. absoluta</i> damage in tomato plants was the YOLOv8l-Seg model. The Mask R-CNN model showed the lowest performance with a metric of 0.806 mAP<sub>0.5</sub>. The results obtained from this study revealed that the proposed model and method can be used effectively in detecting the damage caused by the <i>T. absoluta</i> pest.</p></div>\",\"PeriodicalId\":549,\"journal\":{\"name\":\"European Food Research and Technology\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":3.0000,\"publicationDate\":\"2024-04-04\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://link.springer.com/content/pdf/10.1007/s00217-024-04516-w.pdf\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"European Food Research and Technology\",\"FirstCategoryId\":\"97\",\"ListUrlMain\":\"https://link.springer.com/article/10.1007/s00217-024-04516-w\",\"RegionNum\":3,\"RegionCategory\":\"农林科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"FOOD SCIENCE & TECHNOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"European Food Research and Technology","FirstCategoryId":"97","ListUrlMain":"https://link.springer.com/article/10.1007/s00217-024-04516-w","RegionNum":3,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"FOOD SCIENCE & TECHNOLOGY","Score":null,"Total":0}
Determination of tomato leafminer: Tuta absoluta (Meyrick) (Lepidoptera: Gelechiidae) damage on tomato using deep learning instance segmentation method
Pests significantly negatively affect product yield and quality in agricultural production. Agricultural producers may not accurately identify pests and signs of pest damage. Thus, incorrect or excessive insecticides may be used. Excessive use of insecticides not only causes human health and environmental pollution, but also increases input costs. Therefore, early detection and diagnosis of pests is extremely important. In this study, the effectiveness of the instance segmentation method, a deep learning-based method, was investigated for the early detection of the damage caused by the T. absoluta pest in the leaf part of the tomato plant under greenhouse conditions. An original dataset was created by acquiring 800 healthy and damaged images under greenhouse conditions. The acquired images were labelled as bounding box and automatically converted to a mask label with the Segment Anything Model (SAM) model. The created dataset was trained with YOLOv8(n/s/m/l/x)-Seg models. As a result of the training, the box performance of the proposed YOLOv8l-Seg model was measured as 0.924 in the mAP0.5 metric. The YOLOv8l-Seg model mask values are, respectively: mAP0.5, mAP0.5–0.95, Precision, Recall showed the best performance with values of 0.935, 0.806, 0.956 and 0.859. Then, the YOLOv8l-Seg model, trained with different data input sizes, showed the best performance at 640 × 640 size and the lowest performance with a value of 0.699 in the mAP0.5 metric in the 80 × 80 size. The same dataset was trained with YOLOv7, YOLOv5l, YOLACT and Mask R-CNN instance segmentation models and performance comparisons were made with the YOLOv8l-Seg model. As a result, it was determined that the model that best detected T. absoluta damage in tomato plants was the YOLOv8l-Seg model. The Mask R-CNN model showed the lowest performance with a metric of 0.806 mAP0.5. The results obtained from this study revealed that the proposed model and method can be used effectively in detecting the damage caused by the T. absoluta pest.
期刊介绍:
The journal European Food Research and Technology publishes state-of-the-art research papers and review articles on fundamental and applied food research. The journal''s mission is the fast publication of high quality papers on front-line research, newest techniques and on developing trends in the following sections:
-chemistry and biochemistry-
technology and molecular biotechnology-
nutritional chemistry and toxicology-
analytical and sensory methodologies-
food physics.
Out of the scope of the journal are:
- contributions which are not of international interest or do not have a substantial impact on food sciences,
- submissions which comprise merely data collections, based on the use of routine analytical or bacteriological methods,
- contributions reporting biological or functional effects without profound chemical and/or physical structure characterization of the compound(s) under research.