CN108133204A - A kind of hand body recognition methods, device, equipment and computer readable storage medium - Google Patents
A kind of hand body recognition methods, device, equipment and computer readable storage medium Download PDFInfo
- Publication number
- CN108133204A CN108133204A CN201810064021.8A CN201810064021A CN108133204A CN 108133204 A CN108133204 A CN 108133204A CN 201810064021 A CN201810064021 A CN 201810064021A CN 108133204 A CN108133204 A CN 108133204A
- Authority
- CN
- China
- Prior art keywords
- values
- image
- skin color
- color model
- outdoor scene
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/107—Static hand or arm
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30088—Skin; Dermal
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses hand body recognition methods, including:Obtain background RGB image and outdoor scene RGB image;According to outdoor scene RGB image, skin color model image is calculated using skin color model algorithm;Skin color model image is handled using OTSU algorithms, obtains the corresponding skin color model image in different CrCb regions;Background RGB image and outdoor scene RGB image are made the difference, and extracts and makes the difference the correspondence Cr values of coordinates of targets point and Cb values in result;Calculate the characteristic value of Cr values and the characteristic value of Cb values;The corresponding skin color model image in CrCb regions is as hand body recognition result where choosing the characteristic value of Cr values and the characteristic value of Cb values;This method can utilize OTSU algorithms and Cr values, the characteristic value of Cb values accurately distinguish hand body and with the colour of skin similar in noise object, improve hand body recognition accuracy;The invention also discloses hand body identification device, equipment and computer readable storage mediums, have above-mentioned advantageous effect.
Description
Technical field
The present invention relates to technical field of image processing, more particularly to a kind of hand body recognition methods, device, equipment and computer
Readable storage medium storing program for executing.
Background technology
The accuracy of hand body identification is for needing to identify user's operation using hand body gesture or hand body movement locus
It is extremely important for scene.But existing hand body recognition methods there are one it is very big the defects of, i.e., when there is the skin with people in background
Similar in color during noise object, the recognition methods of hand body can not detach hand body and this noise object, eventually lead to recognition result
Mistake.
Therefore, how to improve hand body identification accuracy can distinguish hand body and with the colour of skin similar in noise object, be
Those skilled in the art's technical issues that need to address.
Invention content
The object of the present invention is to provide a kind of hand body recognition methods, device, equipment and computer readable storage mediums, utilize
The characteristic value of the characteristic value and Cb values of OTSU algorithms and Cr values accurately distinguishes in skin color model image hand body and close with the colour of skin
Noise object, improve hand body recognition accuracy.
In order to solve the above technical problems, the present invention provides a kind of hand body recognition methods, the method includes:
Obtain background RGB image and outdoor scene RGB image;
According to the outdoor scene RGB image, skin color model image is calculated using skin color model algorithm;
The skin color model image is handled using OTSU algorithms, obtains the corresponding skin color model in different CrCb regions
Image;
The background RGB image and the outdoor scene RGB image are made the difference, and extracts and makes the difference coordinates of targets point in result
Correspondence Cr values and Cb values;
Calculate the characteristic value of the Cr values and the characteristic value of the Cb values;
The corresponding skin color model image in CrCb regions where choosing the characteristic value of the Cr values and the characteristic value of the Cb values
As hand body recognition result.
Optionally, according to the outdoor scene RGB image, skin color model image is calculated using skin color model algorithm, including:
The outdoor scene RGB image is converted to outdoor scene YCrCb images;
The outdoor scene YCrCb images are handled using skin color model algorithm, obtain skin color model image.
Optionally, the background RGB image and the outdoor scene RGB image are made the difference, and extracts and make the difference mesh in result
The correspondence Cr values of coordinate points and Cb values are marked, including:
The background RGB image is carried out fuzzy difference made from the outdoor scene RGB image to handle, obtains obscuring and makees difference image;
Binary conversion treatment is carried out to the fuzzy difference image of making, obtains binary image, and according to the binary image
Obtain coordinates of targets point;
Obtain the RGB numerical value that the coordinates of targets point corresponds to the outdoor scene RGB image;
By the RGB numerical value conversions to YCrCb color spaces, obtain the coordinates of targets point and correspond to Cr values and Cb values.
Optionally, the characteristic value of the Cr values and the characteristic value of the Cb values are calculated, including:
Calculate the average value of the Cr values and the average value of the Cb values.
The present invention also provides a kind of hand body identification device, described device includes:
Image collection module, for obtaining background RGB image and outdoor scene RGB image;
Skin color model module, for according to the outdoor scene RGB image, skin color model to be calculated using skin color model algorithm
Image;
OTSU computing modules for being handled using OTSU algorithms the skin color model image, obtain different CrCb
The corresponding skin color model image in region;
Target point feature color extraction module, for making the difference place to the background RGB image and the outdoor scene RGB image
Reason, and extract and make the difference the correspondence Cr values of coordinates of targets point and Cb values in result;Calculate the characteristic value of the Cr values and the Cb values
Characteristic value;
Hand body identification module, for choosing the characteristic value of the characteristic value of the Cr values and Cb values place CrCb regions pair
The skin color model image answered is as hand body recognition result.
Optionally, the skin color model module, including:
Converting unit, for the outdoor scene RGB image to be converted to outdoor scene YCrCb images;
Skin color model unit for being handled using skin color model algorithm the outdoor scene YCrCb images, obtains the colour of skin
Identify image.
Optionally, the target point feature color extraction module, including:
Unit is made the difference, makes the difference, obtains for the outdoor scene RGB image obscure the background RGB image
It is fuzzy to do difference image;
Coordinate of ground point acquiring unit for carrying out binary conversion treatment to the fuzzy difference image of making, obtains binary picture
Picture, and coordinates of targets point is obtained according to the binary image;
Target point feature color extraction unit, for obtaining the RGB that the coordinates of targets point corresponds to the outdoor scene RGB image
Numerical value;By the RGB numerical value conversions to YCrCb color spaces, obtain the coordinates of targets point and correspond to Cr values and Cb values;Calculate institute
State the characteristic value of Cr values and the characteristic value of the Cb values.
Optionally, the target point feature color extraction unit, including:
Target point feature Color Picking subelement, for calculating the average value of the average value of the Cr values and the Cb values.
The present invention also provides a kind of hand body identification equipment, including:
Memory, for storing computer program;
The step of processor, for performing computer program when, realize hand body recognition methods described above.
The present invention also provides a kind of computer readable storage medium, calculating is stored on the computer readable storage medium
The step of machine program, the computer program realizes hand body recognition methods described above when being executed by processor.
A kind of hand body recognition methods provided by the present invention, including:Obtain background RGB image and outdoor scene RGB image;According to
Skin color model image is calculated using skin color model algorithm in outdoor scene RGB image;Using OTSU algorithms to skin color model image into
Row processing obtains the corresponding skin color model image in different CrCb regions;Background RGB image and outdoor scene RGB image are made the difference,
And it extracts and makes the difference the correspondence Cr values of coordinates of targets point and Cb values in result;Calculate the characteristic value of Cr values and the characteristic value of Cb values;Choosing
The corresponding skin color model image in CrCb regions is as hand body recognition result where taking the characteristic value of Cr values and the characteristic value of Cb values.
As it can be seen that this method to outdoor scene RGB image carry out skin color model after, using OTSU algorithms to skin color model image into
Row processing obtains the corresponding skin color model image in different CrCb regions;And the calculated correspondence outdoor scene RGB image
The characteristic value of colour of skin Cr values and the characteristic value of Cb values are as screening foundation, from the corresponding skin color model image choosing in different CrCb regions
The corresponding skin color model image in CrCb regions is as hand body recognition result where taking characteristic value;Therefore this method can accurately distinguish
Hand body and with the colour of skin similar in noise object, improve hand body recognition accuracy;The invention also discloses a kind of hand body identification device,
Equipment and computer readable storage medium have above-mentioned advantageous effect.
Description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, to embodiment or will show below
There is attached drawing needed in technology description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
The embodiment of invention, for those of ordinary skill in the art, without creative efforts, can also basis
The attached drawing of offer obtains other attached drawings.
The flow chart of hand body recognition methods that Fig. 1 is provided by the embodiment of the present invention;
Fig. 2 is converted to the corresponding figure in Y domains in outdoor scene YCrCb images by the outdoor scene RGB image that the embodiment of the present invention provides
Picture;
Fig. 3 is converted to the corresponding figure in Cb domains in outdoor scene YCrCb images by the outdoor scene RGB image that the embodiment of the present invention provides
Picture;
Fig. 4 is converted to the corresponding figure in Cr domains in outdoor scene YCrCb images by the outdoor scene RGB image that the embodiment of the present invention provides
Picture;
Skin color model image is calculated using skin color model algorithm by what the embodiment of the present invention provided in Fig. 5;
Fig. 6 by CbCr threshold values of skin color model that the embodiment of the present invention provides it is corresponding with obtaining CbCr values in OSTU as
Limit divides schematic diagram;
The corresponding skin color model image in 1 region in Fig. 6 that Fig. 7 is provided by the embodiment of the present invention;
The corresponding skin color model image in 2 regions in Fig. 6 that Fig. 8 is provided by the embodiment of the present invention;
The corresponding skin color model image in 3 regions in Fig. 6 that Fig. 9 is provided by the embodiment of the present invention;
The corresponding skin color model image in 4 regions in Fig. 6 that Figure 10 is provided by the embodiment of the present invention;
The fuzzy schematic diagram for making the difference process that Figure 11 is provided by the embodiment of the present invention;
The schematic diagram of binary image after the noise reduction that Figure 12 is provided by the embodiment of the present invention;
The structure diagram of hand body identification device that Figure 13 is provided by the embodiment of the present invention;
The structure diagram of hand body identification equipment that Figure 14 is provided by the embodiment of the present invention.
Specific embodiment
The core of the present invention is to provide a kind of hand body recognition methods, utilizes OTSU algorithms and the characteristic value of Cr values and Cb values
Characteristic value, accurately distinguish in skin color model image hand body and with the colour of skin similar in noise object;The present invention another core be
A kind of hand body identification device, equipment and computer readable storage medium are provided.
Purpose, technical scheme and advantage to make the embodiment of the present invention are clearer, below in conjunction with the embodiment of the present invention
In attached drawing, the technical solution in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is
Part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
All other embodiments obtained without making creative work shall fall within the protection scope of the present invention.
In the prior art when having in outdoor scene with noise object similar in the colour of skin of people, the colour of skin by hand body and this can not make an uproar
Sound object detaches, and eventually leads to hand body recognition result mistake.The present embodiment is extracted using background RGB image and outdoor scene RGB image
Can distinguish hand body and with the colour of skin similar in noise object features color, for follow-up hand body identification basis of characterization is provided;So as to real
Now accurately distinguish in skin color model image hand body and with the colour of skin similar in noise object, improve portable identification certainty and accurate
Property.It specifically please refers to Fig.1, the flow chart of hand body recognition methods that Fig. 1 is provided by the embodiment of the present invention;This method can wrap
It includes:
S100, background RGB image and outdoor scene RGB image are obtained;
Specifically, skin color model is the main method of hand body identification, it is the colour switching based on rgb color domain.Therefore, originally
Embodiment obtains outdoor scene RGB image first when carrying out the identification of hand body.At present, it is only needed when carrying out the identification of hand body in the prior art
Obtain outdoor scene RGB image.Why the present embodiment also needs to obtain background RGB image, primarily to utilizing background
RGB image come distinguish hand body and with the colour of skin similar in noise object features color, for follow-up hand body identification basis of characterization is provided.
The present embodiment does not limit the mode for obtaining background RGB image and outdoor scene RGB image, such as can pass through camera
It directly acquires;The background RGB image transmitted after other equipment shooting and outdoor scene RGB image (such as Background can also be obtained
As RGB domains and real scene image RGB domains).Further, the present embodiment does not limit the ruler of background RGB image and outdoor scene RGB image yet
Very little size, user can be selected according to actual demand.
S110, according to outdoor scene RGB image, skin color model image is calculated using skin color model algorithm;
Specifically, the step obtains skin color model image, i.e., primarily to outdoor scene RGB image progress skin color model
Hand body recognition result in outdoor scene RGB image.The present embodiment does not limit specific skin color model algorithm.Such as utilize existing skill
Existing skin color model algorithm can be realized in art.It is i.e. optional, according to outdoor scene RGB image, calculated using skin color model algorithm
Obtaining skin color model image can include:
Outdoor scene RGB image is converted to outdoor scene YCrCb images;
Outdoor scene YCrCb images are handled using skin color model algorithm, obtain skin color model image.
Wherein, YCrCb is one kind of color space, it will usually be taken the photograph for the image continuous processing in film or number
In shadow system.Y is brightness (luma) ingredient of color and Cb and Cr are then blue and red concentration excursion amount composition.Specifically
Conversion formula can be as follows:
Wherein, result outdoor scene RGB image being converted to after outdoor scene YCrCb images can be with reference chart 2-4.Y is corresponded to respectively
Domain, Cb domains, Cr domains.
General skin color model discrimination standard, the range of colour of skin point may be used in skin color model algorithm in the present embodiment
For:Cb=[77 127], Cr=[133 173].Outdoor scene YCrCb images are handled using skin color model algorithm, obtain skin
Color identification image can be with reference chart 5.Due in outdoor scene RGB image pinkie there are one packaging bag be with similar in the hand body colour of skin,
This packaging bag cannot be removed as a noise object by the skin color model in step S110, follow-up in the present embodiment
The purpose of step is exactly to remove this noise object close with the hand body colour of skin.
S120, skin color model image is handled using OTSU algorithms, obtains the corresponding colour of skin in different CrCb regions and know
Other image;
Specifically, the step mainly carries out the differentiation of noise background and entity using OTSU algorithms (i.e. Otsu algorithm).Its
In, the basic principle of OSTU is the value for giving all the points in figure, chooses a particular value, this value must be in figure intermediate value
In the range of, this value makes all values in figure be divided into two parts, and the variance of this two parts to the point is maximum.What the present embodiment used
OTSU algorithms are existing algorithms, are not being repeated specifically again.
The threshold value of Cb_OSTU and the threshold value of Cr_OSTU can be obtained by OTSU algorithms, according to the threshold value of Cb_OSTU and
In the threshold value and above-mentioned steps of Cr_OSTU the critical field of colour of skin point can obtain the CbCr threshold values of skin color model algorithm with
The relationship between CbCr values is obtained in OSTU;Specific can see forms a division for being similar to four quadrants;It can obtain
To the skin color model image of each quadrant of correspondence.Such as four quadrants, each quadrant area (i.e. a CrCb region) corresponding one
A skin color model image.Illustrate the process by taking above-mentioned Fig. 2 as an example below:
It is Cb_OSTU to pass through the threshold value that OSTU algorithms obtain for real scene image in Fig. 2:98, Cr_OSTU:146.It is logical
Skin color model discrimination standard, colour of skin point is ranging from:Cb=[77 127], Cr=[133 173].Therefore Fig. 6 is please referred to,
Be skin color model CbCr threshold values and OSTU in obtain the relationship between CbCr values, it can be seen that form one be similar to four as
The division of limit;It is respectively 1,2,3,4 that plan, which is divided into 4 quadrants,.Please refer to Fig. 7-10, respectively 1, the corresponding skin in 2,3,4 regions
Color identifies image.
Then pictorial information when following step S130 to step S140 is by the use of no hand body will as background image information
Currently available real-time outdoor scene image information is made the difference with background image information, and being extracted in result is made the difference can be with area
Departure body and with the colour of skin similar in noise object feature color, can be corresponding from different CrCb regions by this feature color
Skin color model image chooses the skin color model image to noise object similar in the removal colour of skin.
S130, background RGB image and outdoor scene RGB image are made the difference, and extracts and make the difference coordinates of targets point in result
Corresponding Cr values and Cb values;
Specifically, the main purpose of the step be in order to obtain in outdoor scene RGB image hand body image corresponding region it is each
The corresponding Cr values of point and Cb values.Difference lies in background RGB figures compared with outdoor scene RGB image for background RGB image in the present embodiment
There is no hand body as in.
Further, the present embodiment, which does not limit, specifically makes the difference mode.Such as can be fuzzy make the difference.This
Embodiment does not also limit extraction and makes the difference the concrete mode of the correspondence Cr values of coordinates of targets point and Cb values in result.Such as it can pass through
The extraction of two-value method makes the difference coordinates of targets point in result (i.e. the corresponding coordinate points of hand body region), and it is corresponding to obtain coordinates of targets point
Cr values and Cb values (such as can correspond to RGB numerical value in outdoor scene RGB image, by RGB numerical value by first obtaining coordinates of targets point
It is converted to Cr values and Cb values).User can carry out the selection of response algorithm according to actual conditions.It and can be according in calculating process
Actual conditions addition noise reduction etc. improve the process of accuracy.
Optionally, background RGB image and outdoor scene RGB image are made the difference, and extracts and make the difference coordinates of targets point in result
Correspondence Cr values and Cb values can include:
Background RGB image with outdoor scene RGB image obscure and is made the difference, obtains obscuring and makees difference image;
Specifically, because of camera, there may be shakes, will make outdoor scene RGB image and background RGB image matching degree not right and wrong
Chang Gao.Such as in view of image array, be exactly background RGB image this point be ceiling white, outdoor scene RGB image this point be
The black of school bag, and the white of ceiling may become point of the corresponding points behind neighbouring in outdoor scene RGB image.Specifically
It is fuzzy make the difference process can be with reference chart 11.
Binary conversion treatment is carried out to fuzzy difference image of making, obtains binary image, and target is obtained according to binary image
Coordinate points;
Specifically, make to need given threshold when difference image carries out binary conversion treatment to fuzzy, as C (i0, j0)<During threshold value,
C (i0, j0) is set as 0;>During threshold value, C (i0, j0) is set as 1.The present embodiment does not limit specific threshold value, can
To be configured according to actual conditions.Such as threshold value is 20.
Further, can also include dropping binary image before coordinates of targets point is obtained according to binary image
It makes an uproar processing.To improve the reliability of binary image.The present embodiment does not limit the mode of specific noise reduction process.It such as can be with
By binary image using morphologic filtering, i.e. dilation erosion operates, this is a kind of common denoising method, no longer superfluous here
It states.Binary image can be with reference chart 12 after noise reduction.
Wherein, coordinates of targets point is obtained according to binary image and obtains the corresponding coordinate points in 1 region in binary image
As coordinates of targets point.Such as value is the coordinates of targets point of 1 point after black, that is, binaryzation in Figure 12.
Obtain the RGB numerical value that coordinates of targets point corresponds to outdoor scene RGB image;
By RGB numerical value conversions to YCrCb color spaces, obtain coordinates of targets point and correspond to Cr values and Cb values.
Specifically, by the corresponding coordinates of targets point in 1 region in binary image in R, G, B number for corresponding to outdoor scene RGB image
Value is taken out.Then rgb color space is converted to by YCrCb color spaces according to above-mentioned conversion formula, obtains coordinates of targets point pair
Answer Cr values and Cb values.Wherein, Y represents brightness in YCbCr color spaces, and Cb and Cr are known as coloration, this space can be used for skin
The space of color detection.
The characteristic value of S140, the characteristic value for calculating Cr values and Cb values;
The purpose of the present embodiment be in order to will distinguish in skin color model image hand body and with the colour of skin similar in noise object, this
Place needs to correspond to Cr values according to coordinates of targets point and Cb values determine in the outdoor scene RGB image the corresponding Cr values of hand body and Cb values come from
The corresponding skin color model image in immediate Cr Cb regions is obtained in step S120.Since coordinates of targets point corresponds to Cr values and Cb
Can also exist in value and answer Cr values and Cb values comprising noise object is corresponding, but noise object is a small number of after all, therefore can be led to
The characteristic value of characteristic value and Cb values for calculating Cr values is crossed, determining that the CrCb regions belonging to the hand body colour of skin are more accurate.
The present embodiment is not defined characteristic value restriction, such as can be median, average value, coordinates of targets point pair
Answer Cr values mode and Cb value modes of Cr values and Cb values etc..In the case where ensureing to calculate accuracy and computational efficiency, preferably
, calculate the average value of Cr values and the average value of Cb values.Such as figure is directed to Figure 12, the average value of Cr values calculated
149.9306 the average value 102.4683 of Cb values.
The corresponding skin color model image conduct in CrCb regions where the characteristic value of S150, the characteristic value for choosing Cr values and Cb values
Hand body recognition result.
Specifically, this step is that affiliated CrCb regions are determined by the characteristic value of Cr values and the characteristic value of Cb values, into
And the corresponding skin color model image in the CrCb regions is chosen as hand body recognition result.Such as the average value 149.9306 of Cr values,
The average value 102.4683 of Cb values, therefore corresponding diagram 6 understands that its affiliated area is 1 region.The corresponding skin color model for choosing 1 region
Image (Fig. 7) is as hand body recognition result.Hand body point mean value is Cb:102.4683;Cr:149.9306, it is clear that in figure 61
Range, so final hand body testing result corresponding diagram 7, in this way, by Fig. 7 it can also be seen that it eliminates original skin color model
In can not remove with the hand body colour of skin similar in noise object.
Step S120 is performed after step silo in the present embodiment, and step S140 and step S150 is after step s 130
It performs.But the present embodiment do not limit step S110 and step S130 perform sequence.Certainly the two can also be held simultaneously
Row.Wherein, step S130 and step S140 using real scene image and background image make the difference body Cr, Cb in one's hands characteristic value mistake
Journey;It is identified, and Cr, Cb characteristic value for identifying hand body are made with the hand body that step is the real scene image based on skin color model threshold value
Hand body and noise the object phase separation that skin color model cannot be distinguished.
Based on above-mentioned technical proposal, hand body recognition methods that the embodiment of the present invention carries, this method to outdoor scene RGB image into
After row skin color model, skin color model image is handled using OTSU algorithms, the corresponding colour of skin in different CrCb regions is obtained and knows
Other image;And the characteristic value conduct of the characteristic value and Cb values of the colour of skin Cr values of the calculated correspondence outdoor scene RGB image
Foundation is screened, from the corresponding skin color model in CrCb regions where the corresponding skin color model image selected characteristic value in different CrCb regions
Image is as hand body recognition result;Therefore this method can accurately distinguish hand body and with the colour of skin similar in noise object, improve hand
Body recognition accuracy.
It is situated between below to hand body identification device provided in an embodiment of the present invention, equipment and computer readable storage medium
It continues, hand body identification device described below, equipment and computer readable storage medium can with above-described hand body recognition methods
Correspond reference.
Please refer to Fig.1 the structure diagram of hand body identification device that 3, Figure 13 is provided by the embodiment of the present invention;The device can
To include:
Image collection module 100, for obtaining background RGB image and outdoor scene RGB image;
Skin color model module 200, for according to outdoor scene RGB image, skin color model to be calculated using skin color model algorithm
Image;
OTSU computing modules 300 for being handled using OTSU algorithms skin color model image, obtain different CrCb areas
The corresponding skin color model image in domain;
Target point feature color extraction module 400, for being made the difference to background RGB image and outdoor scene RGB image, and carries
It takes and makes the difference the correspondence Cr values of coordinates of targets point and Cb values in result;Calculate the characteristic value of Cr values and the characteristic value of Cb values;
Hand body identification module 500, it is corresponding for choosing the characteristic value of the characteristic value of Cr values and Cb values place CrCb regions
Skin color model image is as hand body recognition result.
Based on above-described embodiment, skin color model module 200 can include:
Converting unit, for outdoor scene RGB image to be converted to outdoor scene YCrCb images;
Skin color model unit for being handled using skin color model algorithm outdoor scene YCrCb images, obtains skin color model
Image.
Based on above-mentioned any embodiment, target point feature color extraction module 400 can include:
Unit is made the difference, is made the difference for outdoor scene RGB image obscure background RGB image, obtains fuzzy make the difference
Image;
Coordinate of ground point acquiring unit for carrying out binary conversion treatment to fuzzy difference image of making, obtains binary image, and
Coordinates of targets point is obtained according to binary image;
Target point feature color extraction unit, for obtaining the RGB numerical value that coordinates of targets point corresponds to outdoor scene RGB image;It will
RGB numerical value conversions obtain coordinates of targets point and correspond to Cr values and Cb values to YCrCb color spaces;Calculate the characteristic value and Cb of Cr values
The characteristic value of value.
Based on above-described embodiment, target point feature color extraction unit can include:
Target point feature Color Picking subelement, for calculating the average value of the average value of Cr values and Cb values.
It should be noted that based on above-mentioned any embodiment, described device can be realized based on programmable logic device
, programmable logic device includes FPGA, CPLD, microcontroller etc..
Please refer to Fig.1 the structure diagram of hand body identification equipment that 4, Figure 14 is provided by the embodiment of the present invention;The hand body is known
Other equipment can include:
Memory 600, for storing computer program;
Processor 700 is realized during for performing computer program and obtains background RGB image and outdoor scene RGB image;According to reality
Skin color model image is calculated using skin color model algorithm in scape RGB image;Skin color model image is carried out using OTSU algorithms
Processing obtains the corresponding skin color model image in different CrCb regions;Background RGB image and outdoor scene RGB image are made the difference, and
Extraction makes the difference the correspondence Cr values of coordinates of targets point and Cb values in result;Calculate the characteristic value of Cr values and the characteristic value of Cb values;It chooses
The corresponding skin color model image in CrCb regions is as hand body recognition result where the characteristic value of Cr values and the characteristic value of Cb values.
Further, hand body identification equipment can be mobile terminal, such as mobile phone etc..
The present invention also provides a kind of computer readable storage medium, computer journey is stored on computer readable storage medium
Sequence is realized when computer program is executed by processor and obtains background RGB image and outdoor scene RGB image;According to outdoor scene RGB image,
Skin color model image is calculated using skin color model algorithm;Skin color model image is handled using OTSU algorithms, is obtained
The corresponding skin color model image in different CrCb regions;Background RGB image and outdoor scene RGB image are made the difference, and extracts and makes the difference
As a result the correspondence Cr values and Cb values of middle coordinates of targets point;Calculate the characteristic value of Cr values and the characteristic value of Cb values;Choose the spy of Cr values
The corresponding skin color model image in CrCb regions is as hand body recognition result where the characteristic value of value indicative and Cb values.
The computer readable storage medium can include:USB flash disk, mobile hard disk, read-only memory (Read-Only
Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disc or CD etc. are various to deposit
Store up the medium of program code.
Each embodiment is described by the way of progressive in specification, the highlights of each of the examples are with other realities
Apply the difference of example, just to refer each other for identical similar portion between each embodiment.For device disclosed in embodiment
Speech, since it is corresponded to the methods disclosed in the examples, so description is fairly simple, related part is referring to method part illustration
.
Professional further appreciates that, with reference to each exemplary unit of the embodiments described herein description
And algorithm steps, can be realized with the combination of electronic hardware, computer software or the two, in order to clearly demonstrate hardware and
The interchangeability of software generally describes each exemplary composition and step according to function in the above description.These
Function is performed actually with hardware or software mode, specific application and design constraint depending on technical solution.Profession
Technical staff can realize described function to each specific application using distinct methods, but this realization should not
Think beyond the scope of this invention.
It can directly be held with reference to the step of method or algorithm that the embodiments described herein describes with hardware, processor
The combination of capable software module or the two is implemented.Software module can be placed in random access memory (RAM), memory, read-only deposit
Reservoir (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technology
In any other form of storage medium well known in field.
Above to a kind of hand body recognition methods provided by the present invention, device, equipment and computer readable storage medium into
It has gone and has been discussed in detail.Specific case used herein is expounded the principle of the present invention and embodiment, implements above
The explanation of example is merely used to help understand the method and its core concept of the present invention.It should be pointed out that for the general of the art
For logical technical staff, without departing from the principle of the present invention, can also to the present invention, some improvement and modification can also be carried out, this
A little improvement and modification are also fallen within the protection scope of the claims of the present invention.
Claims (10)
1. a kind of hand body recognition methods, which is characterized in that the method includes:
Obtain background RGB image and outdoor scene RGB image;
According to the outdoor scene RGB image, skin color model image is calculated using skin color model algorithm;
The skin color model image is handled using OTSU algorithms, obtains the corresponding skin color model figure in different CrCb regions
Picture;
The background RGB image and the outdoor scene RGB image are made the difference, and extract pair for making the difference coordinates of targets point in result
Answer Cr values and Cb values;
Calculate the characteristic value of the Cr values and the characteristic value of the Cb values;
The corresponding skin color model image conduct in CrCb regions where choosing the characteristic value of the Cr values and the characteristic value of the Cb values
Hand body recognition result.
2. according to the method described in claim 1, it is characterized in that, according to the outdoor scene RGB image, skin color model algorithm is utilized
Skin color model image is calculated, including:
The outdoor scene RGB image is converted to outdoor scene YCrCb images;
The outdoor scene YCrCb images are handled using skin color model algorithm, obtain skin color model image.
3. method according to claim 1 or 2, which is characterized in that scheme to the background RGB image and the outdoor scene RGB
As making the difference, and extract and make the difference the correspondence Cr values of coordinates of targets point and Cb values in result, including:
The background RGB image is carried out fuzzy difference made from the outdoor scene RGB image to handle, obtains obscuring and makees difference image;
Binary conversion treatment is carried out to the fuzzy difference image of making, obtains binary image, and obtain according to the binary image
Coordinates of targets point;
Obtain the RGB numerical value that the coordinates of targets point corresponds to the outdoor scene RGB image;
By the RGB numerical value conversions to YCrCb color spaces, obtain the coordinates of targets point and correspond to Cr values and Cb values.
4. according to the method described in claim 3, it is characterized in that, calculate the characteristic value of the Cr values and the feature of the Cb values
Value, including:
Calculate the average value of the Cr values and the average value of the Cb values.
5. a kind of hand body identification device, which is characterized in that described device includes:
Image collection module, for obtaining background RGB image and outdoor scene RGB image;
Skin color model module, for according to the outdoor scene RGB image, skin color model figure to be calculated using skin color model algorithm
Picture;
OTSU computing modules for being handled using OTSU algorithms the skin color model image, obtain different CrCb regions
Corresponding skin color model image;
Target point feature color extraction module, for being made the difference to the background RGB image and the outdoor scene RGB image, and
Extraction makes the difference the correspondence Cr values of coordinates of targets point and Cb values in result;Calculate the characteristic value of the Cr values and the feature of the Cb values
Value;
Hand body identification module, it is corresponding for choosing the characteristic value of the characteristic value of the Cr values and Cb values place CrCb regions
Skin color model image is as hand body recognition result.
6. device according to claim 5, which is characterized in that the skin color model module, including:
Converting unit, for the outdoor scene RGB image to be converted to outdoor scene YCrCb images;
Skin color model unit for being handled using skin color model algorithm the outdoor scene YCrCb images, obtains skin color model
Image.
7. device according to claim 5 or 6, which is characterized in that the target point feature color extraction module, including:
Unit is made the difference, makes the difference, is obscured for the outdoor scene RGB image obscure the background RGB image
Do difference image;
Coordinate of ground point acquiring unit for carrying out binary conversion treatment to the fuzzy difference image of making, obtains binary image, and
Coordinates of targets point is obtained according to the binary image;
Target point feature color extraction unit, for obtaining the RGB numbers that the coordinates of targets point corresponds to the outdoor scene RGB image
Value;By the RGB numerical value conversions to YCrCb color spaces, obtain the coordinates of targets point and correspond to Cr values and Cb values;Described in calculating
The characteristic value of the characteristic value of Cr values and the Cb values.
8. device according to claim 7, which is characterized in that the target point feature color extraction unit, including:
Target point feature Color Picking subelement, for calculating the average value of the average value of the Cr values and the Cb values.
9. a kind of hand body identification equipment, which is characterized in that including:
Memory, for storing computer program;
Processor, realizing the hand body recognition methods as described in any one of Claims 1-4 during for performing the computer program
Step.
10. a kind of computer readable storage medium, which is characterized in that be stored with computer on the computer readable storage medium
Program realizes the step of the hand body recognition methods as described in any one of Claims 1-4 when the computer program is executed by processor
Suddenly.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810064021.8A CN108133204B (en) | 2018-01-23 | 2018-01-23 | Hand body identification method, device, equipment and computer readable storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810064021.8A CN108133204B (en) | 2018-01-23 | 2018-01-23 | Hand body identification method, device, equipment and computer readable storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108133204A true CN108133204A (en) | 2018-06-08 |
CN108133204B CN108133204B (en) | 2021-02-02 |
Family
ID=62400103
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810064021.8A Active CN108133204B (en) | 2018-01-23 | 2018-01-23 | Hand body identification method, device, equipment and computer readable storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108133204B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111815662A (en) * | 2019-04-11 | 2020-10-23 | 上海集森电器有限公司 | Behavior recognition implementation method based on face detection |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104331690A (en) * | 2014-11-17 | 2015-02-04 | 成都品果科技有限公司 | Skin color face detection method and system based on single picture |
US9196219B1 (en) * | 2012-07-18 | 2015-11-24 | Amazon Technologies, Inc. | Custom color spectrum for skin detection |
CN106529432A (en) * | 2016-11-01 | 2017-03-22 | 山东大学 | Hand area segmentation method deeply integrating significance detection and prior knowledge |
CN106778704A (en) * | 2017-01-23 | 2017-05-31 | 安徽理工大学 | A kind of recognition of face matching process and semi-automatic face matching system |
CN106845388A (en) * | 2017-01-18 | 2017-06-13 | 北京交通大学 | The extracting method of the mobile terminal palmmprint area-of-interest based on complex scene |
-
2018
- 2018-01-23 CN CN201810064021.8A patent/CN108133204B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9196219B1 (en) * | 2012-07-18 | 2015-11-24 | Amazon Technologies, Inc. | Custom color spectrum for skin detection |
CN104331690A (en) * | 2014-11-17 | 2015-02-04 | 成都品果科技有限公司 | Skin color face detection method and system based on single picture |
CN106529432A (en) * | 2016-11-01 | 2017-03-22 | 山东大学 | Hand area segmentation method deeply integrating significance detection and prior knowledge |
CN106845388A (en) * | 2017-01-18 | 2017-06-13 | 北京交通大学 | The extracting method of the mobile terminal palmmprint area-of-interest based on complex scene |
CN106778704A (en) * | 2017-01-23 | 2017-05-31 | 安徽理工大学 | A kind of recognition of face matching process and semi-automatic face matching system |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111815662A (en) * | 2019-04-11 | 2020-10-23 | 上海集森电器有限公司 | Behavior recognition implementation method based on face detection |
Also Published As
Publication number | Publication date |
---|---|
CN108133204B (en) | 2021-02-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107909138B (en) | Android platform-based circle-like particle counting method | |
CN106611429B (en) | Detect the method for skin area and the device of detection skin area | |
US6263113B1 (en) | Method for detecting a face in a digital image | |
CN108765278A (en) | A kind of image processing method, mobile terminal and computer readable storage medium | |
Ajmal et al. | A comparison of RGB and HSV colour spaces for visual attention models | |
CN105243371A (en) | Human face beauty degree detection method and system and shooting terminal | |
CN108510491B (en) | Method for filtering human skeleton key point detection result under virtual background | |
CN105184216A (en) | Cardiac second region palm print digital extraction method | |
CN108389215B (en) | Edge detection method and device, computer storage medium and terminal | |
JP5969460B2 (en) | Nail region detection method, program, storage medium, and nail region detection device | |
CN112287867B (en) | Multi-camera human body action recognition method and device | |
JP6932402B2 (en) | Multi-gesture fine division method for smart home scenes | |
CN108694709A (en) | A kind of image interfusion method and device | |
CN112801049B (en) | Image classification method, device and equipment | |
CN104574358A (en) | Method and apparatus for scene segmentation from focal stack images | |
CN113129390B (en) | Color blindness image re-coloring method and system based on joint significance | |
CN111161281A (en) | Face region identification method and device and storage medium | |
CN108805838A (en) | A kind of image processing method, mobile terminal and computer readable storage medium | |
CN109064490B (en) | Moving target tracking method based on MeanShift | |
CN108133204A (en) | A kind of hand body recognition methods, device, equipment and computer readable storage medium | |
CN113283439A (en) | Intelligent counting method, device and system based on image recognition | |
CN105893943B (en) | A kind of oil level detection method and system | |
CN109934152B (en) | Improved small-bent-arm image segmentation method for sign language image | |
CN108205641A (en) | Images of gestures processing method and processing device | |
CN110879983A (en) | Face feature key point extraction method and face image synthesis method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right |
Effective date of registration: 20210722 Address after: 264300 No. 699, Jiangjun South Road, Rongcheng City, Weihai City, Shandong Province Patentee after: Rongcheng goer Technology Co.,Ltd. Address before: 266100 Room 308, Beizhai Street Investment Service Center, Laoshan District, Qingdao City, Shandong Province Patentee before: GOERTEK TECHNOLOGY Co.,Ltd. |
|
TR01 | Transfer of patent right |