CN103646255A - Face detection method based on Gabor characteristics and extreme learning machine - Google Patents
Face detection method based on Gabor characteristics and extreme learning machine Download PDFInfo
- Publication number
- CN103646255A CN103646255A CN201310563932.2A CN201310563932A CN103646255A CN 103646255 A CN103646255 A CN 103646255A CN 201310563932 A CN201310563932 A CN 201310563932A CN 103646255 A CN103646255 A CN 103646255A
- Authority
- CN
- China
- Prior art keywords
- learning machine
- extreme learning
- gabor
- image
- feature
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention relates to a face detection method based on Gabor characteristics and an extreme learning machine. The method includes the following steps: step1: using Gabor wavelet transformation to extract sample characteristics of an image; step2: after performing characteristic extraction processing on a training image, converting the training image into a one-dimension row vector according to a row superposition method and then inputting the extracted characteristics into an extreme learning machine network for training; step3: based on the network acquired from learning, detecting an image to be tested and obtaining a final detection result. Compared with the prior art, the face detection method has the following beneficial effects: when a face image is detected under a comparatively simple background condition, the detection rate can reach nearly 90% and at the same time error rate is comparatively low. The method has great robustness with respect to effects of the change of illumination angle and the change of face rotating angle.
Description
Technical field
People's face detects and derives from recognition of face.People's face detects and refers to for any given image, adopts certain method to process it, whether to comprise people's face in process decision chart picture.To belong to mode identification technology.
Background technology
It is the vital first step of recognition of face that people's face detects, and is also the focus in pattern-recognition and computer vision research field.In the present invention, introduce a kind of extreme learning machine method for detecting human face based on Gabor feature.In the method, we extract the feature of training sample with Gabor wavelet transformation, and the feature of extracting is carried out to dimension-reduction treatment.Sample image is being carried out after feature extraction processing, image is being become to one dimension row vector by the mode conversion of row stack, then these features are being input in extreme learning machine and are trained, finally using the method for template matches to detect and in image, whether comprise people's face.
Summary of the invention
Technical matters solved by the invention is to propose a kind of method for detecting human face based on Gabor feature and extreme learning machine, adopts Two-Dimensional Gabor Wavelets to come realization character to extract, and adopts extreme learning machine as sorter, thereby obtained, detects more accurately performance.According to the solution of the present invention, a kind of method for detecting human face based on Gabor feature and extreme learning machine has been proposed, comprise the following steps:
Step 1: in the method for the invention, we have used Gabor wavelet transformation, the sample characteristics of extraction image.
Step 2: training image is being carried out after feature extraction processing, training image is become to one dimension row vector according to the mode conversion of row stack, the feature then these being extracted is input in extreme learning machine network and trains.
Step 3: the network obtaining based on study, treat detecting of test pattern, obtain final testing result.
Compared to existing technology, the inventive method has following beneficial effect: during the facial image of method of the present invention under detecting comparatively simple background, when can accomplish that verification and measurement ratio reaches nearly 90%, error rate is lower.Method of the present invention has good robustness for the impact that lighting angle changes and people's face anglec of rotation changes.
Accompanying drawing explanation
fig. 1 is the process flow diagram that inventor's face detects.
fig. 2 is extreme learning machine network diagram.
the complex background plurality of human faces image detection result of Fig. 3 sample size 150.
fig. 4 rotates the testing result of people's face.
Embodiment
Below in conjunction with accompanying drawing, the present invention is illustrated.Described enforcement example is only for illustrative purposes, rather than limitation of the scope of the invention.
The present invention propose a kind of we with Gabor wavelet transformation, extract the feature of training sample, and the feature of extracting is carried out to dimension-reduction treatment.Sample image is being carried out after feature extraction processing, image is being become to one dimension row vector by the mode conversion of row stack, then these features are being input in extreme learning machine and are trained, finally using the method for template matches to detect and in image, whether comprise people's face.
Fig. 1 is process flow diagram of the present invention.With reference to Fig. 1, performing step of the present invention is as follows:
Adopt Two-Dimensional Gabor Wavelets in time domain, spatial domain and direction, to obtain optimum resolution, two-dimensional Gabor functional form can be expressed as simultaneously:
The value of M represents the direction of Gabor kernel function, and K represents general direction number (K=8, u=0,1,2,3 ... 7,8,8 directions), the wavelength of the value of V decision Gabor filtering (v=0,1,2,3,4, represent 5 frequencies),
determine the size of Gauss's window,
.After reading training sample image information, use matrix turning and Matrix Translation to realize reading images is done to 10 kinds of different mirror images, then these mirror images are done to convolution with Gabor kernel function respectively, and result is converted to vector form, form proper vector.
The foundation of step 2, extreme learning machine model
Use suitable sorter, set up sorter learning model.Conventional sorter has k nearest neighbor, SVM, Bayes and BP neural network etc., and the present invention adopts extreme learning machine (ELM) sorter.
As a class list hidden layer feedforward neural network, different from classic method, it can be random selection network in the connection weights of hidden neuron, input weights and hidden layer deviation can random assignments, output layer weights calculate by analytical algorithm, have the good characteristics such as pace of learning is fast, generalization ability is good.
Suppose to have N different training sample
,
for input sample,
for output sample, wherein
,
.Network has
individual hidden node, activation function is
the unified model of SLFN be
, wherein,
i hidden node and the weights that are connected of inputting node,
?
the weights that are connected of individual hidden node and output node,
?
the deviation of individual hidden node.Network structure as shown in Figure 3.
The matrix form of an above-mentioned N equation can be written as:
H is network hidden layer output matrix.If activation function g (x) infinitely can be micro-, according to the theorem of extreme learning machine,
.
So the training process of single hidden layer feedforward neural network, is equivalent to searching linear system
least square solution
,
.It is the Moore-Penrose descriptor matrix of matrix H.
So far, we have learnt extreme learning machine (ELM) sorter.
Step 3, the identification based on extreme learning machine sorter
Test sample book data are sent in corresponding extreme learning machine sorter model, and Output rusults is recognition result.
Claims (4)
1. the method for detecting human face based on Gabor feature and extreme learning machine, is characterized in that, comprises following step:
Step 1: use Gabor wavelet transformation, extract the sample characteristics of image;
Step 2: training image is being carried out after feature extraction processing, training image is become to one dimension row vector according to the mode conversion of row stack, the feature then these being extracted is input in extreme learning machine network and trains;
Step 3: the network obtaining based on study, treat detecting of test pattern, obtain final testing result.
2. a kind of method for detecting human face based on Gabor feature and extreme learning machine according to claim 1, it is characterized in that, the extraction of the image pattern feature in described step 1, is specially: adopt Two-Dimensional Gabor Wavelets in time domain, spatial domain and direction, to obtain optimum resolution simultaneously; After reading training sample image information, use matrix turning and Matrix Translation to realize reading images is done to 10 kinds of different mirror images, then these mirror images are done to convolution with Gabor kernel function respectively, and result is converted to vector form, form proper vector.
3. a kind of method for detecting human face based on Gabor feature and extreme learning machine according to claim 1, it is characterized in that, the study of extreme learning machine network in described step 2, be specially: the connection weights of selecting at random the hidden neuron in network, input weights and hidden layer deviation can random assignments, by training sample, can calculate output layer weights, sorter has been arrived in namely study.
4. a kind of method for detecting human face based on Gabor feature and extreme learning machine according to claim 1, is characterized in that, the network obtaining based on study in described step 3, treats detecting of test pattern, obtains final testing result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310563932.2A CN103646255A (en) | 2013-11-13 | 2013-11-13 | Face detection method based on Gabor characteristics and extreme learning machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310563932.2A CN103646255A (en) | 2013-11-13 | 2013-11-13 | Face detection method based on Gabor characteristics and extreme learning machine |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103646255A true CN103646255A (en) | 2014-03-19 |
Family
ID=50251466
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310563932.2A Pending CN103646255A (en) | 2013-11-13 | 2013-11-13 | Face detection method based on Gabor characteristics and extreme learning machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103646255A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104102913A (en) * | 2014-07-15 | 2014-10-15 | 无锡优辰电子信息科技有限公司 | Wrist vein certification system |
| CN104537391A (en) * | 2014-12-23 | 2015-04-22 | 天津大学 | Meta learning method of extreme learning machine |
| CN104598920A (en) * | 2014-12-30 | 2015-05-06 | 中国人民解放军国防科学技术大学 | Scene classification method based on Gist characteristics and extreme learning machine |
| CN104616030A (en) * | 2015-01-21 | 2015-05-13 | 北京工业大学 | Extreme learning machine algorithm-based recognition method |
| CN104657718A (en) * | 2015-02-13 | 2015-05-27 | 武汉工程大学 | Face recognition method based on face image feature extreme learning machine |
| CN104700078A (en) * | 2015-02-13 | 2015-06-10 | 武汉工程大学 | Scale-invariant feature extreme learning machine-based robot scene recognition method |
| CN104751141A (en) * | 2015-03-30 | 2015-07-01 | 东南大学 | ELM gesture recognition algorithm based on feature image full pixel gray values |
| CN104809450A (en) * | 2015-05-14 | 2015-07-29 | 郑州大学 | Wrist vein authentication system based on on-line extreme learning machine |
| CN106447691A (en) * | 2016-07-19 | 2017-02-22 | 西安电子科技大学 | Weighted extreme learning machine video target tracking method based on weighted multi-example learning |
| CN106682574A (en) * | 2016-11-18 | 2017-05-17 | 哈尔滨工程大学 | One-dimensional deep convolution network underwater multi-target recognition method |
| CN107067029A (en) * | 2017-03-20 | 2017-08-18 | 新智认知数据服务有限公司 | The image classification method that a kind of ELM and DE based on multi-channel feature are combined |
| CN108510534A (en) * | 2018-04-10 | 2018-09-07 | 辽宁石油化工大学 | A kind of aluminium foil sealing method for conducting leak test based on gal cypress transformation and extreme learning machine neural network |
| CN109948583A (en) * | 2019-03-28 | 2019-06-28 | 合肥京东方显示技术有限公司 | Extreme learning machine, face identification method, readable storage medium storing program for executing and computer equipment |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102622585A (en) * | 2012-03-06 | 2012-08-01 | 同济大学 | Back propagation (BP) neural network face recognition method based on local feature Gabor wavelets |
-
2013
- 2013-11-13 CN CN201310563932.2A patent/CN103646255A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102622585A (en) * | 2012-03-06 | 2012-08-01 | 同济大学 | Back propagation (BP) neural network face recognition method based on local feature Gabor wavelets |
Non-Patent Citations (4)
| Title |
|---|
| 李定珍 等: "B2DPCA和ELM人脸识别算法研究", 《液晶与显示》 * |
| 杨靖尧 等: "基于Gabor变换和极限学习机的贝类图像种类识别", 《大连工业大学学报》 * |
| 王丽娟: "基于Gabor变换和BP神经网络的人脸检测算法的Matlab实现", 《郑州轻工业学院学报（自然科学版）》 * |
| 许学斌 等: "基于多分类投影极速学习机的快速人脸识别方法", 《微电子学与计算机》 * |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104102913A (en) * | 2014-07-15 | 2014-10-15 | 无锡优辰电子信息科技有限公司 | Wrist vein certification system |
| CN104102913B (en) * | 2014-07-15 | 2018-10-16 | 无锡优辰电子信息科技有限公司 | Wrist vena identification system |
| CN104537391A (en) * | 2014-12-23 | 2015-04-22 | 天津大学 | Meta learning method of extreme learning machine |
| CN104598920B (en) * | 2014-12-30 | 2016-05-18 | 中国人民解放军国防科学技术大学 | Scene classification method based on Gist feature and extreme learning machine |
| CN104598920A (en) * | 2014-12-30 | 2015-05-06 | 中国人民解放军国防科学技术大学 | Scene classification method based on Gist characteristics and extreme learning machine |
| CN104616030A (en) * | 2015-01-21 | 2015-05-13 | 北京工业大学 | Extreme learning machine algorithm-based recognition method |
| CN104616030B (en) * | 2015-01-21 | 2019-03-29 | 北京工业大学 | A kind of recognition methods based on extreme learning machine algorithm |
| CN104657718A (en) * | 2015-02-13 | 2015-05-27 | 武汉工程大学 | Face recognition method based on face image feature extreme learning machine |
| CN104700078B (en) * | 2015-02-13 | 2018-06-08 | 武汉工程大学 | A kind of robot scene recognition methods based on scale invariant feature extreme learning machine |
| CN104657718B (en) * | 2015-02-13 | 2018-12-14 | 武汉工程大学 | A kind of face identification method based on facial image feature extreme learning machine |
| CN104700078A (en) * | 2015-02-13 | 2015-06-10 | 武汉工程大学 | Scale-invariant feature extreme learning machine-based robot scene recognition method |
| CN104751141A (en) * | 2015-03-30 | 2015-07-01 | 东南大学 | ELM gesture recognition algorithm based on feature image full pixel gray values |
| CN104809450A (en) * | 2015-05-14 | 2015-07-29 | 郑州大学 | Wrist vein authentication system based on on-line extreme learning machine |
| CN104809450B (en) * | 2015-05-14 | 2018-01-26 | 郑州大学 | Wrist vena identification system based on online extreme learning machine |
| CN106447691A (en) * | 2016-07-19 | 2017-02-22 | 西安电子科技大学 | Weighted extreme learning machine video target tracking method based on weighted multi-example learning |
| CN106682574A (en) * | 2016-11-18 | 2017-05-17 | 哈尔滨工程大学 | One-dimensional deep convolution network underwater multi-target recognition method |
| CN107067029A (en) * | 2017-03-20 | 2017-08-18 | 新智认知数据服务有限公司 | The image classification method that a kind of ELM and DE based on multi-channel feature are combined |
| CN108510534A (en) * | 2018-04-10 | 2018-09-07 | 辽宁石油化工大学 | A kind of aluminium foil sealing method for conducting leak test based on gal cypress transformation and extreme learning machine neural network |
| CN109948583A (en) * | 2019-03-28 | 2019-06-28 | 合肥京东方显示技术有限公司 | Extreme learning machine, face identification method, readable storage medium storing program for executing and computer equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103646255A (en) | Face detection method based on Gabor characteristics and extreme learning machine | |
| Alaslani | Convolutional neural network based feature extraction for iris recognition | |
| Rahman et al. | Face recognition using gabor filters | |
| CN111814574B (en) | Face living body detection system, terminal and storage medium applying double-branch three-dimensional convolution model | |
| CN107273845B (en) | Facial expression recognition method based on confidence region and multi-feature weighted fusion | |
| Baccouche et al. | Sequential deep learning for human action recognition | |
| Thai et al. | Image classification using support vector machine and artificial neural network | |
| CN105718889B (en) | Based on GB (2D)2The face personal identification method of PCANet depth convolution model | |
| Taheri et al. | Animal classification using facial images with score‐level fusion | |
| CN109902615B (en) | Multi-age-group image generation method based on countermeasure network | |
| CN105224942A (en) | A kind of RGB-D image classification method and system | |
| CN108564040B (en) | Fingerprint activity detection method based on deep convolution characteristics | |
| Dong et al. | Finger vein verification based on a personalized best patches map | |
| Chen et al. | Action recognition using ensemble weighted multi-instance learning | |
| Hassan et al. | Pose invariant palm vein identification system using convolutional neural network | |
| CN112949469A (en) | Image recognition method, system and equipment for face tampered image characteristic distribution | |
| Huang et al. | Chinese herbal medicine leaves classification based on improved AlexNet convolutional neural network | |
| Hussein | Robust iris recognition framework using computer vision algorithms | |
| Sharma et al. | Deep convolutional neural network with ResNet-50 learning algorithm for copy-move forgery detection | |
| Agab et al. | New combined DT-CWT and HOG descriptor for static and dynamic hand gesture recognition | |
| CN107133579A (en) | Based on CSGF (2D)2The face identification method of PCANet convolutional networks | |
| Ren et al. | Alignment free and distortion robust iris recognition | |
| Jain et al. | Online signature verification using energy, angle and directional gradient feature with neural network | |
| Angadi et al. | Offline signature recognition system using radon transform | |
| Arigbabu et al. | Gender recognition on real world faces based on shape representation and neural network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20140319 |
|
| RJ01 | Rejection of invention patent application after publication |