CNN-Based Intelligent Classification of Bell Pepper Plant Diseases: A Novel Approach for Precision Agriculture
Gunjan Sharma, Vatsala Anand, Sonal Malhotra, Sanjeev Kukreti, Sheifali Gupta
Abstract
Capsicum, also referred to as bell pepper, is a widely cultivated plant with substantial economic significance in various industries, including condiments, vegetables, and medicinal applications. Accurate diagnosis of leaf diseases poses a significant difficulty in cultivating capsicum. The occurrence of leaf diseases in capsicum plants can significantly impact the overall production quality, resulting in considerable financial losses for agricultural practitioners. The process of manually monitoring plants necessitates specialized knowledge, significant labor, and a substantial amount of time. The implementation of automated detection systems offers the potential to not only streamline processes and minimize labor requirements but also to facilitate the early identification of diseases as they manifest on plant leaves. In recent times, there has been a notable increase in the attention given by researchers to image processing in the field of agriculture. This research introduces convolutional neural network architecture consisting of four layers, designed to automatically detect plant diseases by analyzing leaf images. To enhance the training of the proposed model, a dataset including 20,000 augmented images is developed. The experimental findings provide evidence that the optimized-CNN model, as presented, exhibits a remarkable accuracy of 99.89% in predicting the health status of pepper bell plant leaves, distinguishing between healthy and bacterial-infected ones. The proposed CNN model exhibits strong and reliable outcomes, positioning it as a preliminary warning tool suitable for implementation as a disease diagnostic system in the agriculture domain.