Tomato is a significant commercial crop in India, with production projected to reach approximately 208 lakh tonnes by 2024–25. Early identification of plant diseases plays a vital role in improving both the quality and productivity of tomato crops. This research aims to classify tomato leaf diseases by integrating image processing with machine learning algorithms. The pre-processing steps include image resizing, denoising, and enhancement, followed by LAB-based segmentation to isolate disease-affected areas. The segmented images are decomposed into eight sub-bands using two-level 2D discrete wavelet transforms (DWT). Various features, including energy, entropy, mean, and others, are extracted from the decomposed images, along with colour histogram features from the segmented colour images. These features are subsequently fed into support vector machine (SVM) and random forest (RF) classifiers for disease classification. The proposed approach effectively classifies nine different types of tomato leaf diseases, attaining an accuracy of 93.58% using the SVM classifier and 95.63% with the RF classifier. This wavelet transform-based method for diagnosing leaf diseases improves early disease detection accuracy, enabling targeted interventions and resource-efficient treatments, which can reduce costs, minimize pesticide use, and enhance overall crop health and productivity.