SISTEM DETEKSI PENYAKIT PADA DAUN MANGGA MENGGUNAKAN METODE DETEKSI TEPI SOBEL DAN CANNY

Abstract

Abstract
Mango leaf disease is one of the significant agricultural problems that can affect the productivity and quality of
mango fruit. Therefore, to reduce economic losses and improve agricultural efficiency, it is very important to detect
mango leaf diseases quickly and accurately. In this study, a system was developed to identify mango leaf diseases
based on digital image analysis with the Sobel edge detection technique. This system consists of several stages,
namely: image acquisition, determining the Region of Interest (ROI), converting the image to grayscale, using the
Sobel operator to detect edges, and performing histogram analysis. Mango leaf images are taken from the prepared
dataset to obtain images. Furthermore, the ROI is automatically determined from the center area of the image to
focus the analysis on a more representative part of the leaf. Then the ROI image is automatically converted to
grayscale to facilitate further analysis. The Sobel method is used on grayscale images to find significant edges that
indicate disease symptoms. The results show that this system can quickly find and analyze mango leaf diseases
and provide clear and structured visual information. The results show that this system can be used as a tool to
maintain the health of mango plants effectively and efficiently.