Ensuring the health of rice crops is vital for global food security, with disease detection in rice leaves playing a pivotal role in generating a robust yield. Despite rice being an extensively grown crop, illnesses often offer considerable economic hurdles. This research offers a unique technique to enhance the precision and robustness of rice leaf disease classification by seamlessly integrating morphological and spectral information. Our approach combines Convolutional Neural Networks (CNNs) with transfer learning on the rice leaf disease dataset. Morphological features, comprising observable and measurable physical aspects of rice leaves such as shape, size, color, and texture, are extracted using deep learning approaches like edge detection, thresholding, and shape analysis. These metrics, including leaf length, width, aspect ratio, and surface roughness, collectively contribute to a comprehensive morphological feature set, affording insights into structural changes associated with healthy and sick leaves. Simultaneously, spectral properties collected by spectral analysis provide vital information on the interaction of light with rice leaves. By analyzing light reflectance or transmittance at various wavelengths, specific pigments or molecules suggestive of illness presence are found. These spectral characteristics permit separation between healthy and sick leaves, synergizing with morphological information to considerably boost disease classification accuracy. Our recommended approach develops a thorough categorization model by incorporating spectral and morphological information. This shared feature set is used to train a variety of machine learning algorithms, such as neural networks, Support Vector Machines (SVMs), Decision Trees, Random Forests, and neural networks. Making the most of both data types advantages, our classifier effectively predicts the disease class of new, unseen rice leaves. The methodology comprises fine-tuning a pre-trained CNN model, VGG16, on the rice leaf dataset to classify six major rice leaf diseases. Performance evaluation, which covers measurements like recall, accuracy, precision, and F1-score, underscores the efficacy of our system, reaching an amazing 97.3% accuracy in disease classification, outperforming previous machine learning approaches. This integrated strategy not only enhances disease identification but also permits early action for effective crop management.