In this paper, an in-depth investigation is done on the most prevalent tomato diseases that affect the growth of the plant. The current paper has used the hybrid approach that combines Convolutional Neural Network (CNN) and Random Forest algorithms to determine the diseases by providing images of infected plants, from the area of Punjab and Haryana. The paper has achieved an impressive overall accuracy of 75.82%, apart from that the micro-average is 75.83% and the macro-average is 76.05% from the total of 7 classes. The results that are predicted by this paper are like a boon and significant breakthrough for the farmers, as it is very less expensive to use and reduces the efforts of farmers by simply providing images of the crops and checking the diseases. The accuracy of the paper can be improved in the future by using the other Hybrid model that too is also cost-effective and time-consuming for the farmers of the country. Overall this study provides meaningful and valuable insight into the detection and identification of tomato diseases, and its results have noteworthy implications for the agriculture sector in India.