Fatigue expression recognition method based on hierarchical structural features
A technology for structural feature and expression recognition, applied in the field of image processing, can solve the problem that the dictionary is not discriminative, and achieve the effect of improving the time complexity and space complexity.
Inactive Publication Date: 2018-04-13
GUANGDONG UNIV OF TECH
View PDF4 Cites 3 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0006] In order to solve the problem that the dictionary constructed by the traditional sparse representation algorithm is not discriminative, the present invention proposes a new fatigue expression recognition method based on structural feature hierarchy in combination with the reality of fatigue expressions
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0022] 1. Extract Gabor features
[0023] Gabor wavelet is a windowed short-time Fourier transform, which is an important tool for signal processing in the time domain and space domain, and is widely used in pattern recognition. The characteristics of the quadrature phase of Gabor wavelet and the selectivity of capturing information in spatial frequency, position and direction, the approximate two-dimensional Gabor wavelet function schematic diagram and the human visual cortex cell perception are also very similar. The formula for defining the kernel function of the two-dimensional Gabor wavelet filter is:
[0024]
[0025] where ν∈{0,1,…,α−1} and μ∈{1,2,…,b−1}, correspondingly j∈{1,2,…,a×b}. ν and μ determine the scale and direction of the Gabor filter, and the amplitude of the expression data map in eight directions corresponds to image 3 shown. In the definition of the Gabor kernel function in formula (3), It is the Gauss envelope function, which limits the numeric...
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
Login to View More Abstract
The invention discloses a fatigue expression recognition method based on hierarchical structural features. The whole recognition process is mainly divided into five parts which are blocking, Gabor feature extraction, dictionary vector building, weight assignment and solving, recognition and classification. The fatigue expression recognition method specifically comprises the steps of (1) cutting atraining sample into three parts such as the eyebrows, the cheeks and the mouth; (2) acquiring a block dictionary vector of each divided part by using a K-SVD algorithm; (3) solving a weight value ofeach block dictionary vector by using a weight assignment method of an analytic hierarchy process, and forming various types of sub-dictionaries; and (4) combining all of the sub-dictionaries, and solving by using a structured dictionary learning algorithm. The advantages lie in that the built sparse dictionary has excellent judgment performance, texture features of a face fatigue expression are optimized according to the analytic hierarchy process, and both the time complexity and the space complexity are improved to a certain extent compared with the traditional sparse representation classification solving.
Description
technical field [0001] The invention belongs to the technical field of image processing, in particular to a fatigue recognition method, which can be used for fatigue driving. Background technique [0002] With the acceleration of social development, mental work occupies the vast majority of the proportion. Under today's high-pressure life, it is easy to cause people's mental fatigue. Human fatigue can lead to a decrease in people's ability to work, mainly manifested in inattention, sluggishness, forgetfulness, dizziness, constant yawning, frequent blinking, etc. These manifestations of fatigue can lead to many negative effects. How to detect a person's fatigue level has become a hot issue in the field of image processing. [0003] Sparse coding is derived from neuroscience and is an artificial neural network method that simulates the simple cell receptive field in V1 area of the main visual cortex of the mammalian visual system. This method has spatial locality, directi...
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
Login to View More IPC IPC(8): G06K9/00G06K9/34G06K9/46G06K9/62
CPCG06V40/175G06V10/267G06V10/507G06F18/214
Inventor 罗源张灵向文姬秋敏
Owner GUANGDONG UNIV OF TECH