Unlock instant, AI-driven research and patent intelligence for your innovation.

Video Face Recognition Method Based on Multiple Instance Learning

A multi-instance learning and face recognition technology, which is applied in the fields of digital image processing and computer vision, can solve the problems of low recognition rate and difficult to accurately locate key frames, etc., achieve strong anti-interference ability, solve difficult selection, and good robustness Effect

Active Publication Date: 2017-12-01
JILIN UNIV
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The technical problem to be solved by the present invention is to propose a video face recognition method based on multi-example learning in order to solve the problems of difficult accurate positioning of key frames and low recognition rate caused by high signal-to-noise ratio in the video face recognition problem

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Video Face Recognition Method Based on Multiple Instance Learning
  • Video Face Recognition Method Based on Multiple Instance Learning
  • Video Face Recognition Method Based on Multiple Instance Learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0015] Below in conjunction with accompanying drawing, a kind of color image segmentation method based on histogram that the present invention proposes is described in detail:

[0016] like figure 1 Shown, the video face recognition method of the present invention, its steps are as follows:

[0017] Combine below figure 1 The multi-instance learning-based video face recognition algorithm of the present invention is described in detail.

[0018] First, the preprocessing of the face image is carried out. Gabor wavelet transform can extract the multi-scale and multi-directional local frequency information of the image, can enhance some key features, and has good characteristics in extracting the local spatial frequency domain information of the target. In the field of face recognition, Gabor transform has been widely used.

[0019] The two-dimensional Gabor wavelet function is defined as:

[0020]

[0021] Among them, μ and ν represent the direction and scale of the Gabor...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention proposes a video face recognition algorithm based on multi-instance learning. The algorithm regards each face video as a package, takes the normalized face frame image in the video as an example in the package, and adopts weighted-based The cascaded histogram of the local binary pattern of the block is used as an example feature. In the multi-instance feature space of the training set, a classifier is obtained by using a multi-instance learning algorithm, and then the classification and prediction of the test sample are realized. Through related experiments in the face video database, the algorithm has achieved relatively high recognition accuracy. At the same time, the method has good robustness to illumination changes, expression changes, etc., which verifies the effectiveness of the algorithm.

Description

technical field [0001] The invention relates to the fields of digital image processing and computer vision, in particular to a video face recognition method. Background technique [0002] Video face recognition has become a research hotspot and difficult problem in the field of computer vision in recent years. With the development of the Internet of Things and network security, it has broad application prospects. Compared with static images, the feature information that can be selected in dynamic videos is more abundant and diverse. For example, the temporal dynamic information of videos helps to improve the recognition rate; images with relatively high resolution can be selected from video sequences to improve recognition performance. ; The 3D model of the target can also be reconstructed through video learning, and the target recognition can be realized efficiently by using these models. In summary, temporal and motion information play a crucial role in video-based object...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/00
Inventor 陈海鹏申铉京王玉吕颖达王子瑜徐浩然
Owner JILIN UNIV