Active Contour Integration Segmentation with ann Classification Based Plant Leaf Disease Detection

Resumen

Modern emerging nations are a land of constant innovation, but some farming methods are centuries old. The agricultural industry is expanding quickly, and new diseases are emerging every day, necessitating the development of suitable treatment plans. Finding a plant that has contracted a disease requires more time, money, and effort than it does with the present approaches. It takes a lot of image processing to analyze pictures with different standards. In addition, a substantial waste of labour, resources, time, yield, and product quality can arise from inaccurate plant disease identification. Therefore, for optimal cultivation, a precise assessment of the plant's state is required. Image processing is essential for analysing photographs of different standards. This work's objective is to create an efficient algorithm for segmentation and cataloguing of plant leaf diseases. This study suggests segmenting plant leaf images using proposed Active Contour Integration (ACI) with Artificial Neural Network (ANN) classification algorithm. Neural network-based automatic image recognition systems may improve on existing techniques for early plant disease identification, according to recent studies. The Active Contour Integration (ACI) segmented leaf image is given to Artificial Neural Network (ANN) classification technique was implemented for categorising affected and unaffected part of leaves. The suggested ACI with ANN results are compared with two well-liked segmentation techniques: Watershed, Thresholding, K-Means Clustering and Fuzzy C-Means algorithm. Suggested ACIANN segmentation algorithm is an effective method it can more precisely separate sick leaves from a live background, as shown by simulation results. The recommended Active Contour Integration with ANN method produced the best accuracy result 88.11%, dice result of 89.65%, and jaccard result is 86.28% when compared to the other methods.