Face image glasses removal method, device and storage medium

Abstract

The invention discloses a method for removing glasses from a face image. The method includes: acquiring a real-time image captured by a camera device, extracting a face image from the real-time image; performing normalization processing on the face image, Performing face posture correction using affine transformation to obtain a front face image; by performing binarization processing and edge detection on the front face image, judging whether the front face image includes a glasses area; and, Determining the glasses area in the front face image, searching for pixels around the glasses area in the front face image to fill the glasses area, and obtaining a face image without glasses. The present invention removes the glasses in the human face image to generate the human face image without glasses, and at the same time retains the original human face features in the human face image, thereby improving the recognition rate of human face recognition. The invention also discloses an electronic device and a computer-readable storage medium.

Description

technical field [0001] The present invention relates to the technical field of computer vision processing, in particular to a method and device for removing glasses from a human face image, and a computer-readable storage medium. Background technique [0002] In the field of face recognition, because many people wear glasses, especially deep-frame glasses, during face recognition, the face images with deep-frame glasses have a high similarity, and accurate face recognition cannot be performed. [0003] At present, the face image glasses removal scheme adopted in the industry is to use the two-dimensional generalized principal component analysis method. This method uses the face image without glasses to train the feature space, and reconstructs the face wearing glasses. Contrast, extract the occluded area of ​​the glasses, and then perform error compensation on the image through error iteration to synthesize the final face image without glasses. This method works better for ...

AI Technical Summary

Problems solved by technology

This method works better for input images that are similar to the training image, but it takes a certain amount of time and a certain number of pictures to train, and for input images that are quite different from the training image, although the glasses in the face image are eliminated, the The damage to the face features is serious, and it is impossible to perform accurate face recognition

Images