Application And Research Of Intelligent Shelf Image Recognition Technology

Due to the convenience of shopping,supermarkets have gradually become the primary choice for people's daily shopping.In order to further improve the management level of the supermarket and provide a good shopping experience for consumers,it is necessary to further improve the intelligent management le vel of the supermarket.However,the existing method for obtaining commodity information by the commodity information management software has some shortcomings,the labor cost is high,and the product information cannot be updated in time.Therefore,researching an image recognition method for automatic identification of shelf goods is of great significance for improving the efficiency of supermarket product management.The intelligent shelf image recognition method is mainly divided into three steps.The first step is to collect the image of the shelf product by using the Raspberry Party.The second step is to perform image preprocessing and feature extraction on the collected product image.During the experiment,the shape features and color features of the product images were extracted.At the same time,the extracted image features were reduced by principal component analysis(PCA)and linear discriminant analysis(LDA)to facilitate image recognition.The third step: the image recognition method is used to identify the sample image.Firstly,the average distance method and the error correction SVM method are used as a classification method to identify the product image.In order to improve the accuracy of the classification image recognition method and analog face recognition,a secondary classification image recognition method is designed.The neural network algorithm is combined with a classification image recognition method,and simulation experiments are carried out to verify the effectiveness of the s econdary classification image recognition method.It has been verified that the secondary classification image recognition method improves the accuracy of recognition of the product image.The secondary classification image recognition method is applied to the image recognition process of smart shelves,and the correct rate of image recognition reaches over 90%,which verifies the effectiveness of the secondary classification image recognition method.