Smart Food Analyser: Deterrence of selling Rotten food etymology by Deep Neural Network Applications

Food is an important source of energy for all living things. Sometimes we consume or eat food that is not edible, poisonous, or contaminated and goes unreported. People grow ill and suffer from serious health problems. It costs the government a lot of money to cure illnesses like food poisoning. There hasn't been a lot of study done in this field of food analysis to identify rotting meat or other contaminated food items. According to the news [1], the Muscat Primary Court ordered an eatery to close for a month for supplying sandwiches which is unsuitable for eating. The court also fined an Asian worker 1,000 riyals for making the sandwiches with bad chicken flesh. The Public Authority for Consumer Protection (PACP) works hard to keep Omani and other nationalities safe from dangerous or unhealthy food. Other investigations suggest that manual identification of these sensitive food products falls short of the criteria. This study develops a smart system that uses deep transfer learning to identify and discourage the sale of bad food using image processing and artificial intelligence. The technology employed in this study assists in identifying dangerous food and reporting it to the PACP authority, who then takes the required measures to prevent it from being sold in the market. Python, Keras, Tensor Flow, and Resnet50 were utilized in the system's development. A pre-trained version of the ResNet-50, a 50-layer convolutional neural network, can be used in conjunction with millions of images from the ImageNet database [2]. Fish, meat, chicken, and a range of other animal food items are among the 1000 categories of food that the pretrained network can recognize in food photos. The accuracy of 99.14% is achieved by Resnet50 classifiers. This research assists PACP and other food companies in detecting and avoiding the use of inedible foods. This artificial authority-based tool identifies food quality using a user-friendly interface and informs the appropriate authorities. Also, this research exhibits results using Gradio Interface and android mobile application. The importance of this study is to prevent many Omanis from becoming ill as a result of eating such improper sustenance in their everyday lives.