Plant diseases pose a significant danger to world agricultural production, which requires scalable and effective solutions for early prediction and management. The current work explores deep technology that uses learning to identify plant illnesses, leveraging Convolutional Neural Networks (CNN) for the detection and diagnosis of plant disorders in multiple crop species: corn, potato, rice, tea, and tomato. The system adopts a two-model strategy, involving the fine-tuning of a pre-trained ResNet50 and the development of a custom CNN architecture. Our dual-model framework combines a fine-tuned ResNet50 for high-accuracy diagnosis and a lightweight custom CNN optimized for edge deployment, achieving up to 99.33% validation accuracy on rice diseases. Both models are optimized for robust feature extraction and high-accuracy classification, utilizing techniques such as data augmentation, pre-processing, and hyperparameter tuning. Five data sets comprised of over 27,000 labeled images are processed by resizing, normalizing, and enhancing to improve model generalization and reduce overfitting. The standard pre-trained ResNet50 model, fine-tuned with a modified Softmax layer, achieves superior classification accuracy due to its advanced residual learning framework. Meanwhile, the custom CNN architecture balances computational efficiency and performance, making it appropriate for implementation on limited-resource systems. Measures such as precision and recall demonstrate exceptional performance, particularly for crops such as corn and rice, where classification precision exceeds 95%. This work underscores the revolutionary impact of deep learning on crop disease detection, offering accurate and real-time plant disease identification to improve crop management and reduce losses.