Optimized Deep Neuro Fuzzy Network based Automatic Approach for Segmentation and Food Recognition

Abstract

In the era of Covid-19 pandemic, the people are very much cautious about their health and hence their food choice. It is very important for every individual to be healthy by taking nutritious food. In these days, due lockdowns in county, people could buy the food online only. While doing this, the food choices are made by using the images in online applications and websites. Also, for the restaurants and hotels, it is very much important to post the better quality and truly identifiable images on the portals for customer attractions. In this research work, the food recognition based on food images is targeted to automatically enable, the food items and type of food information for users. From the images, the system will recognize about the type of food as non-invasive or harmful food based on reading and extracting the image characteristics. Several attempts are made for this type of food identification, but these approaches are related to standard object identification, hence couldn't produce the significantly accurate results. Here the approach is developed by using the Poor and Rich Spider Monkey Optimization (PRSM) algorithm based Deep Neuro Fuzzy Network for effective food recognition. This algorithm helps us to reduce the False Positive Rate (FPR) and False Negative Rate (FNR), and increases the accuracy of the image recognition ensuring the maximally accurate classification of the food type.