Agriculture supports a large portion of India’s population and drives its economy, with approximately 70% of rural households depending on agricultural activities. It ranks globally second in crop production and ninth in farm exports. So, the primary goal of this research is to create a system for detecting plant leaf diseases using deep learning models, specifically ImageNet, and a custom CNN model. A benchmark plant leaf disease dataset is accessed from the Kaggle open-source data repository. It has 43,444 training images and 10,861 testing images for model building. In this research work, we have designed a model by applying the ImageNet technique, which is well-known for its classification abilities. Then, leaf disease detection is adapted using the transfer learning technique, with a remarkable accuracy of 99.51%. In addition, the CNN model is used, resulting in a 92.57% accuracy. To make the system more accessible, the ImageNet model is embedded with an Android app using Kotlin and the TensorFlow Lite (TFLite) framework. The user-friendly interface allows users to upload or take new images for disease classification using their camera. The pre-trained model’s integration ensures efficient inference on mobile devices, giving users a simple and effective tool for detecting plant leaf disease.