An image recognition method and device

Abstract

The invention discloses an image identifying method and device so as to accurately identify images with various qualities. The image identifying method comprises calculating image feature vectors of a target image; inquiring representative feature vectors matching with the image feature vectors of the target image; and returning inquiry results, wherein one representative feature vector is obtained by extracting features of image feature vectors of sample images which are clustered; the image feature vectors of the sample images which are clustered are obtained according to similarity clustering of the image feature vectors of multiple sample images; and the image feature vectors of the sample images are obtained by extracting features the same with the image feature vectors of the target image. Every representative feature vector is the feature of a sample image of one kind of image quality, features of sample images of a variety of image qualities can be embodied, more accurate feature descriptions are provided for images, so that target image identification is more accurate, and higher identification capacity can be achieved.

Description

technical field [0001] The invention relates to the field of image processing, in particular to an image recognition method and device. Background technique [0002] In the human image recognition system, image recognition is based on the main features of the image. Each image has its characteristics, such as the letter A has a point, P has a circle, and the center of Y has an acute angle, etc. It can be seen that in the process of machine image recognition, the image can be recognized by eliminating redundant information and extracting key features. [0003] Image recognition using extracted image features, the general method is: to determine whether the extracted image features meet a key feature, if so, the recognition is successful, for example, there is currently a coin image recognition method, the method is the first choice to extract the coin image The diameter of the outer contour circle of the coin. If the diameter of the outer contour circle of the extracted coi...

AI Technical Summary

Problems solved by technology

[0004] However, the current description of the key features of the image is too simple. When the image quality is not good, the extracted image features may not have the key features, resulting in recognition failure. When the time is right, the outer contour of the extracted coin is deformed and does not have the agreed allowable diameter range, resulting in the failure of recognition

Images