Green Chili Classification using Deep Learning Models: CNN and VGGNet

The classification of chilies is essential for the agriculture and food industries, as it determines the quality and classification of the three green chili varieties. This study proposes a deep learning-based classification model to categorize green chilies into three classes, namely green chilies, big chilies, and curly chilies. The primary objective of this work is to develop an accurate and automated classification system using the VGGNet(Visual Geometry Group) and CNN(Convolutional Neural Network) deep learning technique to identify comparatively which method has obtained higher accuracy. The proposed model extracts essential color and the shape features, where the color features include the average and standard deviation of the red, green, and blue channels. Comparative study shows that the image processing based green chili classification, a dataset of 170 chili images as used, with 136 images allocating for training across the three classes in both CNN and VGGNet models. Both models are accurate in this classification, with 90% and 100% accuracy respectively. The findings suggest that VGGNet, due to its deeper architecture, outperforms CNN in green chili classification, making it a promising approach for precision agriculture.