In Malaysia, health monitoring has become a top priority, especially in regard to calorie intake, which is essential for keeping a fit body. The prevalence of overweight people and obesity-related disorders, including diabetes and cardiovascular problems, is alarmingly rising across the nation. The current approach of manually calculating calories based on the amount of food consumed can produce unreliable results. A project utilizing a Convolutional Neural Network (CNN) to detect food classes and a TensorFlow model to calculate calorie counts to address this issue. Combining different Kaggle datasets resulted in a collection of popular Malaysian cuisines, including three regional specialties and three fruits. The TensorFlow model was then trained and tested using the TensorFlow Keras framework. To evaluate the project's accuracy, the performance of the developed model was compared to that of commercial calories using data from the United States Department of Agriculture (USDA) and the Malaysia Ministry of Health (KKM) as a reference.