Detection of Plant Leaf-based Diseases Using Machine Learning Approach

Agriculture productivity is increasing day-by-day based on recent advances and research growth in technology. Detection of plant leaf-based diseases and for improving the quality of plant leaf-based is very essential in agriculture. Detecting various plant leaf-based diseases with human sight, many laboratory-based approaches like polymerase chain reaction, decrease in food production, pest management, hyper spectral techniques are identified for detection of diseases but they are very high time consuming and high cost to the farmers. Identification of recent advanced techniques and various systematic models using Machine Learning (ML) approaches may increase the agriculture productivity. Researchers worked on modern approaches in ML algorithms for detection of leaf diseases for increasing the accuracy results. Every approach has its importance and is focused towards the direction of ML applications and is also based on issues faced by the farmers. In this research paper, detection of leaf-based diseases is analyzed using Support Vector Machine (SVM), Random Forest algorithms. The performance metrics like Root Mean Square Error (RMSE), Peak Signal Noise Ratio (PSNR), Disease affected area of the leaf by using Euclidian Distance method and Accuracy results are compared to benefit the farmers with less time, low cost and increase our agriculture productivity.