Plant health assessment and disease diagnosis are critical for optimizing agricultural productivity, yet traditional methods relying on visual inspection often lack precision and scalability. This project explores the integration of advanced image processing techniques and artificial intelligence (AI) to improve the detection and classification of powdery mildew disease, characterized by a white, powder-like substance on green leaves. The initial phase of the project involved developing a computer vision model to process images by isolating plants, filtering out the background, and categorizing pixel colors into specific ranges (green, yellow, white, and brown) using the HSV color space. Variations in green-to-white pixel ratios were analyzed to estimate the healthiness levels of plants. Subsequent iterations incorporated clustering methods like k-means and AI-based models to refine the detection of anomalies, such as powdery mildew and discolorations, on plant leaves. Principal Component Analysis (PCA) was employed to identify key factors contributing to plant health, including areas of infection and health-to-infection ratios. The AI model achieved a preliminary confidence level of 75% in marking infected areas, with the potential for further improvement through training. The program workflow integrates boundary detection, masking, infection marking, and healthiness scoring, producing outputs that quantify plant health based on the number and area of infections. Challenges such as variations in pixel intensity due to dirt, light reflection, and overlapping color ranges were addressed incrementally, though minor inconsistencies remain. This study demonstrates the potential of AI and image processing in achieving scalable, accurate, and efficient plant health diagnostics, advancing precision agriculture and disease management practices.