Image recognition method, device and system

Abstract

The present application proposes an image recognition method, device and system, the method comprising: obtaining an image to be recognized; obtaining an image recognition model, wherein the image recognition model is generated after machine learning is performed on training samples, and the training samples include : the training image, the clustering label of the training image and the binary classification label of the training image, the clustering label of the training image is obtained after clustering the training image; using the image recognition model, the image to be recognized is Perform recognition, and determine the binary classification score of the image to be recognized. This method can improve the recognition effect of binary classification images.

Description

technical field

[0001] The present application relates to the technical field of image processing, in particular to an image recognition method, device and system. Background technique

[0002] Due to the advantages of images such as being beautiful and intuitive, a lot of information can be displayed in the form of images, for example, product information is displayed in the form of images. However, many product images will contain non-subject information such as promotional text, advertising text, watermark, logo, etc. outside the main body of the product. This part of information can be commonly referred to as psoriasis on the image. for psoriasis images. Since the psoriasis on the image will seriously affect the user experience, it is necessary to identify the psoriasis image for subsequent processing such as image filtering.

[0003] In related technologies, a text line detection method is usually used to identify psoriasis images. However, since there are more and m...

Images