Advances In CNN Classification of Food Products

Abstract

Deep neural networks have been developed in recent years to solve exceedingly complicated computer vision categorization challenges. Although the results obtained with these classifiers are frequently excellent, there are some industries that require better precision from these systems. Increasing Ensemble learning, which integrates several methods, can improve the accuracy of neural networks. Classifiers with the goal of picking a winner based on various characteristics about them These strategies have Despite the fact that they include distinct types of training models and can even produce over fitting with respect to the training data, hence datasets must be carefully selected the outcome. In this research, we are using the different transfer learning CNN algorithms applied on the grocery dataset. Once after considering the grocery dataset the preprocessing is performed and then the transfer learning algorithms are used for training the data. Where we have used multiple transfer learning algorithms, we can select different models during the output checking.