THE AUTOMATIC DETECTION OF TOMATOES LEAF DISEASES

ORAL O., BİLGİN G.

FRESENIUS ENVIRONMENTAL BULLETIN, cilt.30, sa.4, ss.3303-3309, 2021 (SCI-Expanded) identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 30 Sayı: 4
  • Basım Tarihi: 2021
  • Dergi Adı: FRESENIUS ENVIRONMENTAL BULLETIN
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Aerospace Database, Aqualine, Aquatic Science & Fisheries Abstracts (ASFA), CAB Abstracts, Chemical Abstracts Core, Communication Abstracts, Environment Index, Geobase, Greenfile, Metadex, Pollution Abstracts, Veterinary Science Database, Civil Engineering Abstracts
  • Sayfa Sayıları: ss.3303-3309
  • Anahtar Kelimeler: Image Processing, K-means Clustering, Machine Learning, Tomatoes Leaf Diseases
  • Akdeniz Üniversitesi Adresli: Evet

Özet

Automatic perception of plant diseases from symptoms of plant leaves is an important research topic. In this study, an algorithm based on image processing is proposed for the automatic detection and classification of tomatoes leaf diseases Diseased and healthy tomato leaf digital images taken from the field were classified in three different colour fields with K-means clustering and the diseased areas on the leaf were separated from healthy sections. Afterward, the diseased areas obtained from the leaves were classified by NBC and LDA methods. The advantage of using this method is that tomatoes diseases can be identified early or at the first stage. The algorithm developed in the study provides a high accuracy scores till 99% percentages between the diseased leaf and healthy leaf groups. The algorithm has better ability especially in order to discriminate Cladosporium fulvum diseased leaf from healthy leaves. On the other hand, the mean accuracy rate for the classification are 86.74% accuracy with NBC and 86.55% with LDA.