<p>The agriculture sector is a cornerstone of human sustenance and economic stability. Timely detection of plant diseases is essential for optimizing crop yields and mitigating potential losses. This paper presents a novel approach for plant identification and disease diagnosis based on leaf analysis. The proposed approach harnesses the capabilities of transfer learning through SqueezeNet, an eighteen-layer deep Convolutional Neural Network (CNN). A pre-trained SqueezeNet model is utilized for feature extraction from plant leaf images, followed by classification using a supervised machine learning technique, namely the Support Vector Machine (SVM) classifier. A stepwise disease detection framework was developed, integrating SqueezeNet and SVM classifier, and evaluated using image datasets comprising both diseased and healthy plant pairs. The experimental results reveal that the proposed approach achieves exceptional accuracy, ranging from 96.9% to 100%, in plant identification and disease detection through leaf analysis.</p>