Edge Detection Based Segmentation of Wheat Plants Diseases

Wheat is mankind’s primary food grain and the fourth most produced cereal; it is the primary food source for a lot of people, particularly in Asian and European countries. In addition to providing proteins, minerals, and carbohydrates, it also creates jobs for millions of people. Wheat plants, on the other hand, are subject to a variety of diseases that can result in significant quality and yield losses. Wheat diseases such as Leaf or Brown Rust, Stem Rust, Yellow Rust, Loose Smut, and Powdery Mildew are now contributing to these losses. To address this issue, a reliable and accurate automated disease identification system must be developed. In this study, we demonstrate the Sobel and Canny edge detection methods for segmentation, data extraction, and identifying the main properties of objects in wheat diseases. We also assess their effectiveness in terms of several established performance metrics, such as histogram, Mean Squared Error (MSE) and structural similarity index measure (SSIM) which are based on edges, slopes, and levels of ambient noise. It is possible to categorize the various edge detection systems and gain a better understanding of the value and constraints of the performance measures employed by combining the findings of the evaluation with those from other related publications.