Analysis Of Banana Plant Disease Characterization Using Thermal Camera With Tressolding Method

Abstract

Banana is a fruit plant that is widely produced in Indonesia. Unfortunately, this plant is very susceptible to diseases which can reduce the quality and quantity of the crop. This paper proposes disease detection in banana plants using a thermal camera. The detection is carried out using image processing techniques with multilevel thresholding methods. The image is captured using a thermal camera, then the image is preprocessed to suit what is desired. After that, so that the position is the same as the image taken using a digital camera, the image produced by the thermal camera is carried out by an image registration process. The image processing result is compared with the ground truth image obtained from a digital camera to determine the effectiveness of the proposed method. The effectiveness of the proposed method is measured using the parameters Recall, Precision, F-measure, and Accuracy. The effectiveness of the proposed method is quite effective because it produces parameter values above 80%, namely the recall value of 86,59%, the Precision of 99,1%, the F-measure of 92%, and the accuracy of 89,78%.

Plant phenotyping relevance

熱画像とマルチレベル閾値処理によりバナナ植物の病徴を検出・評価する手法を提案し、デジタル画像を基準に性能検証しているため、植物フェノタイピング手法が中心である。

Code and data availability

The paper describes thermal/digital image acquisition of banana plants and MATLAB-based multilevel thresholding analysis, but contains no data availability statement, no public dataset or image repository, and no author code release. Only example figures are embedded in the article itself, which does not qualify as a分享

No evidence-backed public reproduction asset is currently recorded.