Rice plays a crucial role in the economy of South-East Asia, but the region is grappling with the decline in rice crop production and quality due to diseases and pests. The manual monitoring and identification of these issues pose considerable difficulties for farmers. To tackle this problem, a real-time automated system is proposed, which identifies common rice crop diseases based on leaf symptoms. The system incorporates a user-friendly web interface that enables farmers to upload leaf images for analysis. To improve accuracy, the system employs a segmentation approach during the preprocessing stage, effectively isolating diseased portions in the leaves. Disease classification is then performed using a lightweight Convolutional Neural Network (CNN) model, resulting in an impressive overall precision of 97.07% with k-fold cross-validation. The proposed model surpasses the performance of pre-trained CNN models in disease classification. Furthermore, to showcase its effectiveness, the model is validated on a maize dataset, achieving an accuracy of 94.00%. By offering prompt and accurate disease identification, this automated system provides invaluable support to farmers, empowering them to make informed decisions and mitigate the adverse effects of diseases on rice crops.