Automatic Early Leaf Spot Disease Segmentation on Cotton Plant Leaf
Khushal Khairnar1, Sajidullah Khan2

1Khushal Khairnar, Department of Computer Science & Engineering, Sandip University, Nashik, India.
2Sajidullah Khan, Professor, Department of Computer Science & Engineering, Sandip University, Nashik, India.

Manuscript received on May 25, 2020. | Revised Manuscript received on June 29, 2020. | Manuscript published on July 30, 2020. | PP: 1161-1164 | Volume-9 Issue-2, July 2020. | Retrieval Number: B4157079220/2020©BEIESP | DOI: 10.35940/ijrte.B4157.079220
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (

Abstract: Diseases are decreasing production of plants. At present, farmers are identifying, diagnosing diseases and monitoring health in plants by their own knowledge and experience. Naked eye observation by farmers and experts on big plantation areas cannot be possible each time and it can be expensive. Accurate identification of visually observed diseases, symptoms and controls has not studied yet. Therefore a fast automatic, economical and accurate system is an essential research topic that may improve in leaf disease detection of plant disease. The proposed automatic early leaf spot disease segmentation on leaf of cotton plant system is based on image processing and machine learning where segmenting the three major diseases such as Bacterial Blight, Alternaria leaf spot and Cercospora leaf spot. Initially, the infected leaf images are captured from cotton plant fields by using a digital camera. Scaling, background removing and color conversion are done in the preprocessing phase. After preprocessing, the infected region is obtained by using K-means clustering algorithm. The infected region can be applied for detecting the diseases on cotton plant. 
Keywords: Clustering, feature extraction, image processing, segmentation.