Sugarcane Classification of Image Processing and Convolutional Neural Networks

Abstract

Sugarcane is a significant crop with several uses in the food, bio-energy, and bio-based product sectors. Many elements, including climate, soil fertility, and plant diseases, can have an impact on the quality of sugarcane. In this study, it suggest a sugarcane classification system based on convolutional neural networks (CNNs) and image processing to automatically evaluate sugarcane crops’ quality automatically. Using image processing techniques, the system uses digital photographs of sugarcane to extract attributes including color, texture, and form. These attributes are then sent into a CNN to be categorized into various grades or quality levels. By testing the proposed system using a sizable collection of photographs of sugarcane, comparing the outcomes to those attained using conventional visual inspection techniques. The findings indicate that the suggested system can classify sugarcane crops with a high degree of accuracy and dependability. The sugarcane industry may employ the suggested approach to increase the efficacy and precision of sugarcane quality evaluation. This research highlights the possibility of fusing deep learning models with image processing methods for agricultural applications.