Fecha de publicación:
Fuente: WIPO Food Portada
The current invention is a food adulteration detection system using machine learning (ML), Internet of Things (IoT) technology integration and GSM module. It is meant for automatic detection of food adulteration so as to help all stakeholders to have necessary benefits. The system captures images of food items in question live and detects any possible food adulteration with the help of ground truth samples or quality food items’ images. The system includes high imaging sensor for capturing food images, IoT module, GSM module and Wi-Fi module. The IoT integrated system enables the automated examination of food items and take necessary steps. The system is cloud-assisted to store sensed data and results of data analytics. Convolutional Neural Network (CNN) classifier is used for detection of food adulteration. Both input data and output data are stored in cloud database for scalability and availability. The given input data (test data/food item image) is subjected to noise removal, feature selection prior to giving it to a food adulteration detection model which is based on deep learning. The detection model uses ground truth samples stored in cloud for accurate adulteration detection. The current invention is useful to authorities of Food Safety and Standards Authority of India (FSSAI). The current invention is beneficial to many stakeholders such as food item owners, food sellers, food inspectors, food inspection authorities and general public besides researchers and academia.