9th International Conference on ICT for Sustainable Development

Authors - Shruti Jha, Ishmeen Kaur Garewal, Lancy Aathisaya, Leona Alphonso, Bhakti Aher
Abstract - This project proposes a novel approach to streamline the extraction and organization of nutritional details from food packaging using OCR. The proposed system efficiently retrieves vital nutritional data from images or text of food products, providing users with accurate and easily accessible information. Additionally, it tracks users' dietary restrictions and fitness goals, enabling customized diet plans. Nutritional information is an essential consideration for making informed food choices. However, extracting this information from food packaging can be time-consuming and challenging, especially for individuals with dietary restrictions or fitness goals. Traditional methods of manual data entry are prone to errors, while existing automated solutions often lack the accuracy and flexibility required. The proposed system addresses these limitations by leveraging OCR technology to extract nutritional information from food packaging with high accuracy. The system first pre-processes the input image to improve readability and reduce noise. Next, it applies OCR algorithm to extract the text from the image. The extracted text from the food label is parsed to identify key nutritional information which includes calorie count, macronutrients, and vitamins. This is done using a data analysis and random forest classifier. Finally, the results include nutritional information in easy language, data visualization to give a visual appeal and information regarding allergies from the ingredients used in the product.