Method for correcting face cutting frame based on face feature points

Abstract

The invention provides a method for correcting a face cutting frame based on face feature points. The method comprises the following steps: S1, preliminarily determining the face cutting frame; s2, expanding the face position according to the five-point feature distance; s3, correcting the face cutting frame; s3.1, expanding the cutting upper limit to be above the eyebrows; s3.2, expanding a cutting lower limit, and ensuring that the feature information of the mouth corner and the jaw is in a cutting frame; and S3.3, when the human face cutting proportion is not appropriate, properly expanding the left and right cutting boundaries. Due to reasons such as face shapes, if human face features such as eyes and eyebrows are cut off only according to the preliminarily determined human face cutting frame, human face recognition errors or non-recognition is caused; and according to the correction method provided by the invention, after the face cutting frame is corrected, the face cutting accuracy is increased and the recognition precision is effectively improved because the conditions are met. In this way, it is ensured that the human face is properly cut, and false detection is avoided.

Description

technical field [0001] The invention relates to the technical field of intelligent video processing, in particular to a method for correcting a human face clipping frame based on human face feature points. Background technique [0002] At present, face recognition technology is widely used in identity authentication, security check, access control and other systems. The face recognition system mainly includes four components: face image acquisition and detection, face image preprocessing, face image feature extraction, and matching and recognition. Among them, the cropped face in the face image acquisition and detection process will directly affect the result of face recognition, because if the cropped face is different from the face stored in the system, such as shaved beard, changed hairstyle, or added glasses , Changing the expression may cause the comparison to fail. Therefore, in order to increase the accuracy of face recognition, the recognized face will cut out irre...

AI Technical Summary

Problems solved by technology

This simple cropping strategy can easily crop the important features of the face (eyebrows, eyes, chin, etc.), resulting in an incomplete face, which affects the accuracy of face recognition

For example attached image 3 As shown, when the face shape is long, in order to obtain a face with equal width and height, when cutting half of the face, the eyes will be cut off

Images