This paper presents a deep learning-based approach to detect early blight and late blight in potato leaves using the EfficientNetB0 model. The study aims to address the challenge of accurate disease detection in potato crops, which is crucial for maintaining agricultural productivity and food security in India. The dataset consists of 2152 potato leaf images, including 2000 diseased images (1000 each for early blight and late blight) and 152 healthy images. The EfficientNetB0 model, known for its efficient use of computational resources, was chosen for its ability to scale width, depth, and resolution uniformly. The model achieved a testing accuracy of 99.05%, surpassing several widely used techniques. The study also includes a literature review, methodological details, and experimental results, demonstrating the effectiveness of the proposed model in disease detection and classification. The authors conclude by highlighting the potential of deep learning in agricultural practices and future research directions, including the development of a user-friendly web application for disease recognition and management.This paper presents a deep learning-based approach to detect early blight and late blight in potato leaves using the EfficientNetB0 model. The study aims to address the challenge of accurate disease detection in potato crops, which is crucial for maintaining agricultural productivity and food security in India. The dataset consists of 2152 potato leaf images, including 2000 diseased images (1000 each for early blight and late blight) and 152 healthy images. The EfficientNetB0 model, known for its efficient use of computational resources, was chosen for its ability to scale width, depth, and resolution uniformly. The model achieved a testing accuracy of 99.05%, surpassing several widely used techniques. The study also includes a literature review, methodological details, and experimental results, demonstrating the effectiveness of the proposed model in disease detection and classification. The authors conclude by highlighting the potential of deep learning in agricultural practices and future research directions, including the development of a user-friendly web application for disease recognition and management.