A Novel Approach for Identification of Healthy and Unhealthy Leaves Using Scale Invariant Feature Transform and Shading Histogram-PCA Techniques

Tomato is one of the significant cultivation crop devoured by dominant part of the populace in everywhere throughout the world because of medical advantages. As a result of huge number of tomato buyers and tomato being a high return and a multi-time crop, huge number of ranchers develops this cultivation crop for an enormous scope. In any case, the serious issue is that the varieties in climatic conditions make the agriculture crops and different pieces of the plants, for example, roots, stem, leaf, and seeds helpless to the infection assaults. Of every one of these parts, leaf is generally defenseless to the infection assaults. Moreover, the cultivation crop maladies spread at a quicker rate contrasted with the other classification of yields. Notwithstanding these issues, the manual recognizable proof of leaf illnesses is testing and dreary. To address these issues, we have robotized the procedure of recognizable proof of the wellbeing state of the tomato leaf, with the presumption that an appropriate picture obtaining process exists. We have proposed two calculations to distinguish whether the tomato leaf is sound or unfortunate. One of them depends on the scale invariant feature transform (SIFT), and the other depends on the shading histogram and the principal component analysis (PCA). To train the calculation, a sum of 80 tomato leaf pictures (solid and unfortunate) are caught utilizing the camera. Utilizing the proposed calculations, the highlights are extricated from these pictures and put away in the database. At the point, when a question tomato leaf picture is given as information, the calculation removes the highlights and contrasts these highlights and the put away highlights and gives the yield of the calculation as wellbeing or undesirable leaf. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.