Rapid accurate human face detection method

A face detection, accurate technology, applied in the direction of instruments, character and pattern recognition, computer components, etc., can solve the problems of decreased accuracy, prone to false detection, low detection speed, etc., to reduce computational complexity and false detection The effect of improving efficiency and improving accuracy

Inactive Publication Date: 2017-04-26
NANJING UNIV OF POSTS & TELECOMM
View PDF1 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Since the human face is a kind of natural structural object with quite complex detail changes, the facial features "eyes, ears, nose, mouth, and eyebrows" of the human face affect the facial features, and the shape of the facial features will vary greatly depending on the emotion , Different faces have different contours and other factors also bring great challenges to face detection
The AdaBoost algorithm proposed by Paul Viola and Michael Jones is very representative in the field of face detection. However, due to its complex algorithm calculatio

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
  • Rapid accurate human face detection method
  • Rapid accurate human face detection method
  • Rapid accurate human face detection method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0046] Below in conjunction with accompanying drawing, the present invention is described in further detail.

[0047] The face detection flow chart of the present invention is as figure 1 As shown, the face detection method of the present invention will be described in detail below.

[0048] 1. Threshold and operate skin color segmentation

[0049] First, for the input sample image, the RGB color space is converted to CMYK, HSV and YCbCr color space respectively.

[0050] The threshold segmentation formulas corresponding to the three color spaces of CMYK, HSV and YCbCr are:

[0051]

[0052]Combine the skin color segmentation thresholds of the CMYK, HSV and YCbCr color spaces, and segment the image to obtain face candidate areas according to the combined thresholds. After skin color segmentation, the effect is as follows figure 2 shown.

[0053] 2. Construct MWR-AdaBoost face detector

[0054] A. Construction of Weak Classifier

[0055] (1) Construct a haar-like rec...

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 discloses a rapid accurate human face detection method. The rapid accurate human face detection method comprises steps that step 1: skin color segmentation thresholds are respectively set in CMYK color space, HSV color space, and YCbCr color space, and three thresholds are acquired for combination operation, and a threshold after the combination operation is used as the threshold for the skin color detection; step 2: the threshold after the combination operation is used for skin color segmentation pretreatment of an input image, and a human face detection candidate area is acquired; step 3: an MWR-AdaBoost algorithm is used for the human face detection of the candidate area, and the human face area is marked in the image. The combination operation of the skin color segmentation thresholds of the CMYK color space, the HSV color space, and the YCbCr color space is used for the pretreatment of the image to acquire the candidate area, and the problem of the skin color segmentation of easy missing of the human face candidate area in a complicated scene caused by using the single color space is solved; the MWR-AdaBoost algorithm is used for the human face detection of the candidate area, and the degradation phenomenon of the image after the skin color treatment caused by using the AdaBoost algorithm is inhibited. By using the method provided by the invention, the human face detection is completed rapidly and accurately.

Description

Technical field: [0001] The invention relates to a fast and accurate human face detection method, which belongs to the technical field of image pattern recognition. Background technique: [0002] Face is the most distinctive feature of a person. Even if two people are twins, their faces will still have subtle differences. Therefore, in today's era of intelligent information, faces provide a lot of valuable information. In recent years, more and more attention has been paid to the research on face recognition. Due to the natural advantages of face recognition and the characteristics of not being detected by the tested individual, it has been widely used in access control systems, camera surveillance systems, etc. [0003] Face detection is the key link of face recognition. The premise of face recognition is to accurately locate the area where the face is located in the background image. Since the human face is a kind of natural structural object with quite complex detail cha...

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
IPC IPC(8): G06K9/00
CPCG06V40/161G06V40/172
Inventor 方承志宦太杰袁海峰
Owner NANJING UNIV OF POSTS & TELECOMM
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products