Ning Yang1,2, Minfeng Yuan1, Pan Wang1, Rongbiao Zhang1, Jun Sun1, Hanping Mao2. 1. School of Electrical and Information Engineering, Jiangsu University, Zhenjiang, PR China. 2. Institute of Agricultural Engineering, Jiangsu university, Zhenjiang, PR China.
Abstract
BACKGROUND: As one of China's important economic crops, tea is economically damaged due to its large yield. The overall goal of this study is to develop an effective, simple, apt computer vision algorithm to detect tea disease area using infrared thermal image processing techniques and to estimate tea disease. RESULTS: This paper finds that the area of tea disease has certain regularity with its infrared image gray distribution. Using this rule, we extracted two characteristic parameters into a classifier to help achieve rapid tea disease detection, which increases the accuracy of detection a small amount. The tea disease detection algorithm consisted of the following steps: classify canopy infrared thermal image; convert red, green and blue image to hue, saturation and value; thresholding; color identification; noise filtering; binarization; closed operation; and counting. A correlation coefficient R2 of 0.97 was obtained between the tea disease detection algorithm and counting performed through human observation, which is 2% higher than traditional algorithms without classifiers. CONCLUSIONS: This article provides guidance for monitoring the condition of tea gardens with airborne thermal imaging cameras.
BACKGROUND: As one of China's important economic crops, tea is economically damaged due to its large yield. The overall goal of this study is to develop an effective, simple, apt computer vision algorithm to detect tea disease area using infrared thermal image processing techniques and to estimate tea disease. RESULTS: This paper finds that the area of tea disease has certain regularity with its infrared image gray distribution. Using this rule, we extracted two characteristic parameters into a classifier to help achieve rapid tea disease detection, which increases the accuracy of detection a small amount. The tea disease detection algorithm consisted of the following steps: classify canopy infrared thermal image; convert red, green and blue image to hue, saturation and value; thresholding; color identification; noise filtering; binarization; closed operation; and counting. A correlation coefficient R2 of 0.97 was obtained between the tea disease detection algorithm and counting performed through human observation, which is 2% higher than traditional algorithms without classifiers. CONCLUSIONS: This article provides guidance for monitoring the condition of tea gardens with airborne thermal imaging cameras.