Vegetable Leaf Identification System Using Image Processing

Abstract

Identification of vegetables Leaf is essential to various agricultural and food-related applications Manual plant recognition by human experts is achievable, but the method could be more efficient and efficient. This study presents a robotic leaf recognition technique for vegetables using MATLAB, a powerful platform for software, a popular way for processing images and pattern recognition. The proposed system employs a multi-stage approach to accurately identify and classify vegetable leaves. Initially, picture of a leaf preprocessed to enhance their quality the contrast enhancement removes any noise or unwanted artifacts. and are applied to improve efficiency of future analysis. Next, extraction of features is performed on the preprocessed images to convert them into meaningful representations. A group of relevant and discriminative features, such as shape, texture, and color, are removed utilizing methods such as grey level co-occurrence (GLCM), In the classification stage, a algorithm for machine learning is on a dataset of labeled vegetable leaves to learn the mapping between the classes that match to the retrieved features. SVMs, or vector support machines, are classifiers, k-Nearest Neighbors (k-NN), After training, the classifier can forecast the class labels of unseen vegetable leaves since extracted features. The proposed MATLAB-based system is evaluated using a diverse dataset of vegetable leaf images, encompassing various species and leaf types, the proposed system attends improved accuracy compare to traditional human being The proposed system SVM algorithm can give reliable outcome using the percentage of 96.77%.