Segmentation Method Based on Artificial Bee Colony for Recognizing Leaf Lesion

Abstract

Many studies on object detection have been initiated but these methods have some limitation. A segmentation method was proposed to recognize a leaf lesion in leaf images and overcome the limitation of existing object detection method in terms of accuracy and processing time. The method includes steps based on Artificial Bee Colony, Otsu, and geometry. The method was conducted in three phases, image preparation, lesion recognition and measurement, and evaluation. The paper shows results of the evaluation phase. The results show that the proposed segmentation method achieved better percentage of accuracy and produced a shorter processing time.