Commodity image classifying method based on complementary features and class description

Abstract

The invention discloses a commodity image classifying method based on complementary features and class description, comprising the following steps of: firstly, taking the classified image as a training sample; secondly, carrying out the resolution compression on all marking image class and test pictures by using a picture batch editing tool; thirdly, extracting the complementary features of towertype gradient direction histograms and tower type key word histograms of the pictures in various marking images; fourthly, extracting the features of the tower type gradient direction histograms and the tower type key word histograms of images of commodities to be classified; fifthly, constructing a class descriptor of each marking image class; and sixthly, classifying the obtained feature vectors by using a nearest neighbor classification algorithm, calculating the distance between the images of the commodities to be classified and various marking image class descriptors and using the image class with the shortest distance as the classification result. According to the invention, two complementary features can be fully utilized and the classification result is more accurate by using the nearest neighbor classification algorithm based on the image-class distance.

Description

technical field [0001] The invention relates to a method for automatic classification of commodity images, in particular to an automatic classification algorithm for commodity images based on complementary features and an improved image-class distance algorithm. Background technique [0002] With the popularization and development of the Internet, e-commerce has gradually entered a new era, the number of e-commerce websites has increased dramatically, and a number of well-known e-commerce websites at home and abroad have emerged, such as Amazon, ebay, Taobao, etc. E-commerce websites need to mark the products sold online to facilitate users' search. Under the current circumstances, these labels only describe the basic information (meta information) of the product, such as the name, place of origin, size, price, etc. of the product, and it is difficult to reflect the complete characteristics of the product. For example: whether women’s leather shoes are round or pointed, whe...

AI Technical Summary

Problems solved by technology

If the labeling of these potential interest information is done manually, it will undoubtedly be very time-consuming and labor-intensive for e-commerce websites with a large number of products and varieties.

Images