Face image normalizing method

Abstract

The invention discloses a face image normalizing method. The method comprises the following steps of 1, normalizing a target face image to acquire an initial face gray image; and 2, adjusting the gray value of the initial face gray image, wherein the gray value of a relatively flat face area is relatively reduced, and the gray value of a relatively un-flat face area is relatively increased. The scaling factor of a Weber face is adaptively adjusted according to a space flatness mask, intrinsic information of different space positions of a face base image under normal lighting condition can be used, and when shadow is caused by illumination in a target image, the shadow part has relatively small scaling factor, so that large response is not caused, and the problem that a shadow edge cannot be effectively processed by the primary Weber face method is effectively solved. Complementary information under different scales can be effectively used by a multi-scale adaptive Weber fusion method provided by the invention, so that more useful information for face identification / authentication is kept.

Description

【Technical Field】 [0001] The invention relates to the technical field of computer vision and image processing, in particular to a method for normalizing a face image. 【Background technique】 [0002] In recent years, face recognition / authentication has attracted the attention of many scientific research institutes and enterprises due to its wide application in public security, identity verification, multimedia retrieval, and human-computer interaction, and a large number of related researches have also been carried out. However, for most of the existing face recognition / authentication systems, changes in the external environment light still severely restrict their performance. This is mainly because the difference in facial imaging of the same individual caused by changes in illumination may even be greater than the difference between different individuals. In the design of the actual application system, due to the difference in identification / authentication and registration time ...

AI Technical Summary

Problems solved by technology

One is to learn a face model under variable illumination based on a large number of face samples under different illumination conditions, such as the illumination normalization method based on the quadratic polynomial model proposed by the patent application number 200710027817.8. This method deals with The effect is good, but it is often computationally intensive and time-consuming, and the demand for a large number of training samples also limits its practical application

The second type of method uses traditional image processing methods to preprocess face images under changing lighting conditions, such as histogram equalization, logarithmic transformation, etc. This type of method does not take into account the light formation model, but only adjusts the target image Therefore, the ideal practical effect cannot be achieved; the third type of method extracts the light insensitive amount according to the Lambertian reflection model to normalize the light. This type of method is more effective and more complex than the first type of method. low, thus gaining widespread attention

However, the method based on the Weber face still has the following two shortcomings: (1) The obtained local second-order relative gradient has no light insensitivity at the edge of the shadow; information at different scales, and many recent researches have shown that information at different scales has complementary characteristics

Images