The leaf of Mango diseases needs to be identified precisely and quickly for the purpose to increase crop yield and ensure sustainable farming methods. In this research, we used Principal Component Analysis (PCA) and Convolutional Neural Networks (CNNs) to create a quantum-inspired deep learning framework for facilitating the identification of diseases in Harumanis mango leaves. PCA, a quantum-inspired dimensionality reduction technique, preserves the most significant features when compressing high-dimensional image data. This substantial decrease boosts computational efficiency and reduces overfitting in the CNN model. Our research includes preprocessing a carefully selected dataset of mango leaf images using image augmentation and PCA transformation, followed by training a CNN model on the condensed feature space. The suggested structure offers advantageous classification accuracy for a range of disease classifications. In addition to accuracy, we evaluate the model using precision, recall, F1-score, and a confusion matrix to guarantee that it is robust and dependable. We also show that the model is suitable for edge-based or real-time agricultural applications from thoroughly evaluating time complexity, which quantifies the time required for each stage of data loading, PCA transformation, training, and evaluation. Experimental results indicate that the proposed framework offers high accuracy and efficient implementation, making it a promising tool for smart agriculture. This study contributes to the growing field of quantum-inspired AI in plant disease diagnosis while also offering agricultural stakeholders a scalable and interpretable solution.