IDENTIFICATION AND RECOGNITION OF LEAF DISEASE USING ENHANCED SEGMENTATION TECHNIQUES

Abstract

Segmenting refers to the technique of breaking up an image into its component parts one by one. When it comes to the process of segmenting photos, there is a plethora of choice available at current point in time. These options range from the easy thresholding approach to the complicated color image segmentation techniques. The bulk of the time, the parts that go into making up these sub-assemblies are items that individuals are able to easily identify and categorize as being distinct from one another. As a result of the limitation of computer lack of intelligence to differentiate between distinct items, a wide variety of techniques have been devised and utilized in the process of segmenting photographs. In order to complete its tasks, the image segmentation algorithm requires a wide range of image characteristics to be provided as input. This could be referring to the colors that are contained within an image, the borders that are included within the image, or a particular region that is contained within the image. In order to break down color images into their component elements, we make use of an algorithm that is inspired by natural selection. The research uses enhanced segmentation techniques to identify and recognize the leaf disease in plants. The study conducts extensive simulation to test the efficacy of the model. The results show that the proposed method achieves higher segmentation accuracy than other methods.

Keywords

• segmentation
• images