Prediction Of Plant Leaf Health By Image Contour Segmentation And LSTM Model

Authors: Arpit Pethe Scholar, Arjun Rajput Assistant Prof, Dr. Sanjay Sharma, Dr. Sanjay Sharma

Abstract: Agriculture is the backbone of the Indian economy and contributes significantly to the GDP. Crop diseases, particularly those affecting leaves, lead to a decline in both the quality and quantity of agricultural produce. Traditional methods like expert diagnosis and pathogen analysis depend on skilled professionals and may be time-consuming. These approaches are also prone to human errors, affecting the accuracy of disease identification and management. This paper has proposed Plant Leaf Heath Prediction Model (PLHPM) leaf health prediction model that segment input image and extract features for learning. Image segmentation was done by active contour method, while histogram features was extracted from the image. Extracted histogram features were used for the LSTM model training. Experiment was done on real dataset images of potato. Result shows that proposed Plant Leaf Heath Prediction Model (PLHPM) has increases the detection precision and accuracy in less execution time.

DOI: http://doi.org/10.5281/zenodo.17440544