The agricultural sector presents myriad challenges, with farmers grappling to accurately identify diseases in their crops, limited effective treatment methods, the impact of changing climatic conditions on yields, and the persistent issue of low crop prices. This research paper is centered on a novel approach to disease identification in bell pepper plants, leveraging deep learning architectures, specifically AlexNet, GoogleNet, ResNet (18, 50, 101), and Vgg (16, 19). The primary objective is a comprehensive analysis of pre-trained convolutional neural network (CNN) architectures to assess their performance and identify the most suitable model for disease classification in bell peppers. This study introduces a valuable resource for bell pepper farmers, offering an innovative solution to enhance disease identification accuracy beyond conventional methods. The proposed automation concept streamlines the crop disease identification process, significantly reducing the time and effort required by farmers, and ultimately facilitating their agricultural practices. Early disease identification, achieved with reduced effort, holds the potential to substantially boost crop yields. The research uniquely contributes to the field by providing insights into the performance of different pre-trained CNN architectures, considering both augmented and non-augmented image data. Through rigorous comparisons, the findings highlight the effectiveness of Vgg 19, as a highly suitable choice for bell pepper image classification. This research not only addresses critical challenges in agriculture but also introduces novel methodologies that promise tangible benefits for bell pepper farmers, marking a significant contribution to the field of crop disease management.