As food becomes more accessible, the prevalence of obesity is rising, which is a serious chronic disease. Maintaining good health requires precise monitoring of caloric intake. Traditional methods, nevertheless, could be tedious and wasteful. Calorie estimation from food photos using deep learning neural networks is the subject of this research. We provide a technique that analyses food images using Deep Neural Network (DNN) to calculate calorie content. Deep Neural Network (DNN) is with 22 layers to accurately identify the food in the system. A massive collection of tagged food pictures with calorie information is used to train the DNN. As part of the training process, the model extracts shape, colour, and texture information from the images, and then converts it to calorie content. One of the many advantages of this technology is that it provides calorie estimates more efficiently and without invasiveness than manual methods. To further assist with dietary tracking and weight management goals, it may also be integrated with mobile applications. Properly evaluating calorie content for complex recipes with several components is difficult, and obtaining high-quality and diverse training data to avoid bias is another difficulty. When compared to the state-of-the-art method, the suggested approach performs better.