Implementation of Item Management Process and Food Recognition in a Fridge using Computer Vision and IoT

Several studies have reported the use of Radio Frequency Identification(RFID) Sensors in smart fridges; however, smart fridges using computer vision remains to be accomplished. Here we build a Smart Fridge using Computer Vision, and Internet of things. This required us to train our model on Google’s Cloud Vision API (Application Programming Interface), a solution for large-scale detection. The model is able to accurately predict what comes in front of the camera. The next step was to improve the Human-Fridge interaction. This was done with the help of Mobile Application, considering the growing demand of IoT Applications. The main focus of the App was to improve User Interface and functionality, by using React Native at Frontend. The main features included are; Inventory management, Recipe suggestion using API, Nutritional Information using food API, and Spoilage detection to limit the food wastage. Considering how fast life is today, we are building an enhanced Recipe suggestion engine based on the contents of the fridge. Our project implies the importance of Human-centric applications, where ease and comfort of end-user is the main priority, by enhancing User Interface and Functionality, and keeping the process simple yet efficient.