A new approach to detect mildew disease on cucumber (Pseudoperonospora cubensis) leaves with image processing

Ozguven M. M., Altas Z.

JOURNAL OF PLANT PATHOLOGY, cilt.104, sa.4, ss.1397-1406, 2022 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 104 Sayı: 4
  • Basım Tarihi: 2022
  • Doi Numarası: 10.1007/s42161-022-01178-z
  • Dergi Adı: JOURNAL OF PLANT PATHOLOGY
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, BIOSIS, CAB Abstracts, Environment Index
  • Sayfa Sayıları: ss.1397-1406
  • Anahtar Kelimeler: Image processing, Cucumber, Mildew disease, Pseudoperonospora cubensis, ABSOLUTE ERROR MAE, LEAF, IDENTIFICATION, RECOGNITION, CLASSIFICATION, SEGMENTATION, SEVERITY, RMSE
  • Ankara Üniversitesi Adresli: Hayır

Özet

Image processing algorithms were employed in present study to determine the level of damage caused by mildew disease on cucumber plants. Fifty of infected plant images were randomly selected and processed with the image processing algorithm developed using image processing toolbox module of MATLAB. Then the results obtained from the image processing algorithm were compared with the assessments of experts. The image processing method predicted the disease levels with 1.90 RMSE and Theil's UII of 0.0312. Kolmogorov-Smirnov test was used to test the normality assumption of the data and test results revealed a normal distribution (p>0.05). Determination coefficient (R-2 = 0.995, p < 0.01) and Pearson's correlation coefficient (r = 0.997, p < 0.01) indicated significant positive relationship between image processing and expert assessments. The study results indicated that present image processing algorithm could successfully be used in place of expert assessment for diagnosis of mildew disease in cucumber plants.