Method and device for generating texture image and composite image

Abstract

The embodiment of the invention provides a method and device for generating a texture image and a composite image executed by a computer, and the method comprises the steps: applying a basic graphic function to a coordinate value of any first position in a to-be-generated image, and obtaining a basic value; for the first position, generating a disturbance noise value at least according to the noise disturbance function; applying an adjustment function to the sum of the basic value and the disturbance noise value to obtain a pixel value of the first position; and generating a texture image according to the pixel value of each position. By using the method, a large number of complex, non-repeated and low-cost texture images are generated, and the method can be used for training an image recognition model. A composite image can be generated by combining the texture image and the text image, and position information of each text in the composite image is recorded as text position label data of the composite image. And the synthesized image is used as a verification code to train a verification code recognition model, so the recognition capability of the recognition model is effectivelyimproved.

Description

technical field [0001] One or more embodiments of this specification relate to the fields of machine learning and data security, and in particular, to a method and device for generating texture images and composite images. Background technique [0002] In many image recognition projects, it is necessary to use a large number of complex and non-repetitive images to train the image recognition model to enhance its recognition ability. This requires an automatic image generation method to generate the required images in large quantities. The images generated by the existing image generation methods are relatively simple, and the style is relatively single. The recognition model trained with such images has poor recognition ability for complex images and is prone to overfitting problems, that is, the image style for effective recognition It is relatively single and does not have the ability to recognize images of various types and styles. Therefore, there is a need for a metho...

AI Technical Summary

Problems solved by technology

The verification codes generated by these solutions are relatively simple, and the trained recognition models have poor ability to recognize complex verification codes; the verification codes generated by each specific verification code generator have a single style, which is difficult to cover multiple types of verification codes, and lacks universality. sex

Images