A.F. Cheshkova
Siberian Federal Research Center of Agricultural Biotechnology, Russian Academy of Sciences, pos. Krasnoobsk, Novosibirskii r-n, Novosibirskaya obl., 630501, Russian Federation
Abstract. White spot (Ramularia tulasnei) is one of the most common diseases that has a significant impact on the yield of garden strawberries. Its detection using hyperspectral measurements and analysis is a possible alternative to traditional methods. In the study, strawberry leaves with visible symptoms of the disease were used for spectral analysis. The reflection spectrum of the leaves was recorded with a Photonfocus hyperspectral camera (wavelength range 475-900 nm, 150 channels with a spectral resolution of 3 nm and a spatial resolution of 2048×1088 pixels) under laboratory conditions using the linear scanning method. Differentiation of healthy and affected areas was performed using the support vector machine, k-nearest neighbours algorithm, spectral angle method and 9 vegetation indices. To compare the efficiency of classification algorithms, the error matrices of the corresponding methods were calculated. The highest classification accuracy (98%) had the support vector machine using vegetation indices as informative features. Sufficiently high classification accuracy was shown by the methods of k-nearest neighbours (97%) and spectral angle (96%). The least accuracy was observed for the classification method according to the threshold value of vegetation indices (82%). The results of the study confirm the possibility of using hyperspectral technologies to detect the white spot of garden strawberries.
Keywords: hyperspectral technologies; strawberry fungal diseases; image analysis; machine learning methods
Author Details: A.F. Cheshkova, Cand. Sc. (Phys.-Mat.) leading research fellow (e-mail: Этот адрес электронной почты защищён от спам-ботов. У вас должен быть включен JavaScript для просмотра.).
For citation: Cheshkova AF [Detection of white spot of garden strawberry using hyperspectral imaging] Dostizheniya nauki i tekhniki APK. 2022;36(9):84-8. Russian. doi: 10.53859/02352451_2022_36_9_84.