CN107480617B - Skin color detection self-adaptive unit analysis method and system - Google Patents
Skin color detection self-adaptive unit analysis method and system Download PDFInfo
- Publication number
- CN107480617B CN107480617B CN201710650617.1A CN201710650617A CN107480617B CN 107480617 B CN107480617 B CN 107480617B CN 201710650617 A CN201710650617 A CN 201710650617A CN 107480617 B CN107480617 B CN 107480617B
- Authority
- CN
- China
- Prior art keywords
- gmb
- current image
- block
- size
- chroma
- 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.)
- Active
Links
Images
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
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
- G06V40/162—Detection; Localisation; Normalisation using pixel segmentation or colour matching
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
Abstract
The invention provides a skin color detection self-adaptive unit analysis method and system. The method designs a skin color detection unit self-adaptive determination method based on chrominance analysis according to the characteristics of skin color detection, completes the skin color detection of the images, and then completes the skin color detection of subsequent related images through the time correlation between the images. Therefore, the appropriate block size is set for the video without the compressed information, and the higher judgment accuracy is ensured while the algorithm execution speed is increased.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a skin color detection self-adaptive unit analysis method and system.
Background
With the rapid development of multimedia technology and computer network technology, video is becoming one of the mainstream carriers for information dissemination. The accurate and rapid skin color detection technology can enhance the effect of double results with little effort no matter face video retrieval or online video beautifying.
If a unified pixel-based judgment method is adopted, although the judgment is accurate, the operation speed of the judgment statement in the algorithm execution is far higher than the conventional addition, subtraction, multiplication and division speed, and the large-scale adoption of the judgment statement can greatly reduce the execution speed of the algorithm, so that the timeliness of the algorithm is influenced, and the negative effect is particularly prominent in the application of high-definition, ultrahigh-definition and high-resolution video images.
If a uniform block-based decision method is employed, the operating speed of the algorithm can be increased. Note that in practical application, scenes are often complex, and situations such as multiple persons, single person, different resolutions, and the like exist. The cured block division cannot meet the complex situation of practical application.
Disclosure of Invention
The embodiment of the invention aims to provide a skin color detection unit analysis method based on image analysis, and aims to solve the problems of low efficiency or low accuracy of the skin color detection technology in the prior art.
The embodiment of the invention is realized in such a way that a skin color detection self-adaptive unit analysis method comprises the following steps:
step 0: let t be 1; t represents the video frame number, also becomes the current frame number;
step 1: determining the size of the generalized block according to the resolution of the current image;
step2 calculating a first chroma intensity variable Inuk，InukA u chroma intensity variable representing the current image;
step 3: if Inu is presentkNot less than Thres, first set the sizef (gmb)k) Setting k to k +1, and then proceeding to Step 6; otherwise, go to Step 4;
wherein, sizef (gmb)k) Denotes gmbkOne-dimensional size of skin color detection unit; gmbkRepresents the kth square block of the current image, Thres represents the threshold value, Thres is less than or equal to size (gmb)k)，size(gmbk) Represents a one-dimensional size of the generalized block;
step 4: calculating a second chroma-strength variable Invk，InvkA v chroma intensity variable representing a current image;
step 5: if Invk<Thres, then set the sizef (gmb) firstk)＝size(gmbk) Then setting k to k + 1; otherwise, set sizef (gmb) firstk) 1, then k is set to k + 1;
step 6: if k is less than or equal to num, then the Step2 is entered again; otherwise, go to Step 7;
num represents the number of generalized blocks of the image divided according to the size of the generalized blocks;
step 7: if the next frame image of the current image does not exist, ending; otherwise, let t be t +1, and set the next frame of the current image as the current image, and then go to Step 8;
step 8: first press size (gmb)k) The previous frame of the current image and the current image are divided into blocks according to the size of the current image, and then block statistical variables of each block are calculated;
step 9: for current picture pictAll blocks of the time similarity identification; pictRepresents the tth frame of the video, also called the current image;
step 10: if the time similarity identifiers of all the blocks of the current image are 0, then the Step2 is re-entered; otherwise, go to Step 11;
step 11: the one-dimensional size of the skin tone detection unit of all blocks of the current image is first calculated based on the temporal similarity identifier of the block, and then Step7 is re-entered.
Another objective of the present invention is to provide an adaptive unit analysis system for skin color detection, which includes:
and the frame sequence number setting module is used for Step 0: let t equal to 1, pictRepresents the tth frame of the video, also called the current image;
the generalized block size confirming module is used for confirming the size of the generalized block according to the resolution of the current image;
wherein, QVGA, VGA, 720P are the standard sizes of the images disclosed in the industry; gmbkRepresenting the kth square block of the current image, which is called generalized block for short, wherein the initial value of k is 1; size (gmb)k) Represents a one-dimensional size of the generalized block;
a first chroma intensity variable calculation module for calculating a first chroma intensity variable, Inuk＝std(u(i,j)|u(i,j)∈gmbk)；
Wherein u (i, j) represents u chroma value of current image in ith line and jth column; std represents the mean square error; inu (Inu)kA u chroma intensity variable representing the current image;
a first chroma strength variable threshold judgment processing module for judging if InukNot less than Thres, first set the sizef (gmb)k) Setting k to be k +1, and entering a judgment processing module; otherwise, entering a second chroma intensity variable calculation module;
wherein, sizef (gmb)k) Denotes gmbkOne-dimensional size of skin color detection unit; thres represents the threshold, Thres ≦ size (gmb)k)；
A second chroma intensity variable calculation module for calculating a second chroma intensity variableInvk＝std(v(i,j)|v(i,j)∈gmbk)；
Wherein v (i, j) respectively represents the v chroma value of the current image in the ith line and the jth column; invkA v chroma intensity variable representing a current image;
a second chroma strength variable threshold judgment processing module for judging if Invk<Thres, then set the sizef (gmb) firstk)＝size(gmbk) Then setting k to k + 1; otherwise, set sizef (gmb) firstk) 1, then k is set to k + 1;
the judgment processing module is used for judging whether k is less than or equal to num, and entering the first chroma intensity variable calculation module; otherwise, entering a next frame image judgment processing module of the current image;
num represents the number of generalized blocks of the image divided according to the size of the generalized blocks;
the next frame image judgment processing module of the current image is used for judging whether the next frame image of the current image does not exist or not, and then ending; otherwise, let t be t +1, and set the next frame of the current image as the current image, then enter the block statistical variable calculation module;
a block statistical variable calculation module for first size (gmb)k) The previous frame of the current image and the current image are divided into blocks, and then block statistical variables of each block are calculated
tik＝std(y(i,j)-py(i,j)|y(i,j)∈gmbkAnd py (i, j) is within pg mbk)；
Wherein y (i, j) is pictThe luminance value of the ith row and the jth column; py (i, j) is pictThe ith row and the jth column of the previous frame; pg mbkRepresents pictThe k-th square block of the previous frame; tikDenotes gmbkThe block statistical variable of (1);
the time similarity identification module is used for identifying the time similarity of all the blocks of the current image;
the specific method comprises the following steps:
wherein, Thres1Denotes a first decision threshold, Thres18 ═ 8 × (1+24/fps), fps denotes the video capture frame rate; note (r) notekDenotes gmbkA temporal similarity identifier of;
the time similarity identifier judging module is used for judging whether the time similarity identifiers of all the blocks of the current image are 0 or not, and reentering the first chroma intensity variable calculating module; otherwise, entering a one-dimensional size calculation module of a block skin color detection unit;
and the block skin color detection unit one-dimensional size calculation module is used for calculating the one-dimensional sizes of skin color detection units of all blocks of the current image according to the time similarity identifiers of the blocks, and then reentering the next frame image judgment processing module of the current image.
The invention has the advantages of
The invention provides a skin color detection self-adaptive unit analysis method. The method designs a skin color detection unit self-adaptive determination method based on chrominance analysis according to the characteristics of skin color detection, completes the skin color detection of the images, and then completes the skin color detection of subsequent related images through the time correlation between the images. Therefore, the appropriate block size is set for the video without the compressed information, and the higher judgment accuracy is ensured while the algorithm execution speed is increased.
Drawings
FIG. 1 is a flow chart of a skin tone detection adaptive unit analysis method in accordance with a preferred embodiment of the present invention;
fig. 2 is a structural diagram of a skin color detection adaptive unit analysis system according to a preferred embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to the accompanying drawings and examples, and for convenience of description, only parts related to the examples of the present invention are shown. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The invention provides a skin color detection self-adaptive unit analysis method and system. The method designs a skin color detection unit self-adaptive determination method based on chrominance analysis according to the characteristics of skin color detection, completes the skin color detection of the images, and then completes the skin color detection of subsequent related images through the time correlation between the images. Therefore, the appropriate block size is set for the video without the compressed information, and the higher judgment accuracy is ensured while the algorithm execution speed is increased.
Example one
FIG. 1 is a flow chart of a skin tone detection adaptive unit analysis method in accordance with a preferred embodiment of the present invention; the method comprises the following steps:
step 0: let t equal to 1, pictRepresenting the tth frame of the video, also referred to as the current image.
Step 1: the size of the generalized block is determined according to the current image resolution.
Wherein, QVGA, VGA, 720P are the standard sizes of the images disclosed in the industry; gmbkRepresenting the kth square block of the current image, which is called generalized block for short, wherein the initial value of k is 1; size (gmb)k) Representing the one-dimensional size of the generalized block.
Step2 calculating a first chroma intensity variable, Inuk＝std(u(i,j)|u(i,j)∈gmbk)。
Wherein u (i, j) represents u chroma value of current image in ith line and jth column; std represents the mean square error; inu (Inu)kRepresenting the u chroma intensity variation of the current image.
Step 3: if Inu is presentkNot less than Thres, first set the sizef (gmb)k) Setting k to k +1, and then proceeding to Step 6; otherwise, Step4 is entered.
Wherein, sizef (gmb)k) Denotes gmbkOne-dimensional size of skin color detection unit; thres represents the threshold, and is generally taken to be Thres ≦ size (gmb)k)。
Step 4: calculating a second chroma-strength variable Invk＝std(v(i,j)|v(i,j)∈gmbk)。
Wherein v (i, j) respectively represents the v chroma value of the current image in the ith line and the jth column; invkRepresenting the v chroma intensity variation of the current image.
Step 5: if Invk<Thres, then set the sizef (gmb) firstk)＝size(gmbk) Then setting k to k + 1; otherwise, set sizef (gmb) firstk) Then k is set to k + 1.
The first chromaticity and the second chromaticity are randomly selected, that is, the first chromaticity may be u chromaticity, the second chromaticity may be v chromaticity, or the first chromaticity may be v chromaticity, and the second chromaticity may be u chromaticity.
Step 6: if k is less than or equal to num, then the Step2 is entered again; otherwise, go to Step 7.
Where num denotes the number of generalized blocks into which the picture is divided by the size of the generalized block.
Step 7: if the next frame image of the current image does not exist, ending; otherwise, let t be t +1, and set the next frame of the current image as the current image, and then proceed to Step 8.
Step 8: first press size (gmb)k) The previous frame of the current image and the current image are divided into blocks, and then block statistical variables of each block are calculated
tik＝std(y(i,j)-py(i,j)|y(i,j)∈gmbkAnd py (i, j) is within pg mbk)。
Wherein y (i, j) is pictThe luminance value of the ith row and the jth column; py (i, j) is pictThe ith row and the jth column of the previous frame; pg mbkRepresents pictThe k-th square block of the previous frame; tikDenotes gmbkThe block statistical variable of (2).
Step 9: time similarity identification is carried out on all blocks of the current image, and the specific method is as follows:
wherein, Thres1Indicates a first decision threshold, which may be Thres18 ═ 8 × (1+24/fps), fps denotes the video capture frame rate; note (r) notekDenotes gmbkThe time similarity identifier of (a).
Step 10: if the time similarity identifiers of all the blocks of the current image are 0, then the Step2 is re-entered; otherwise, Step11 is entered.
Step 11: the one-dimensional size of the skin tone detection unit of all blocks of the current image is first calculated based on the temporal similarity identifier of the block, and then Step7 is re-entered.
The first mode is as follows: is suitable for notek1-block
If notekSet the sizef (gmb) as 1k)＝sizef(pgmbk)
Wherein, sizef (pgmb)k) Expression of pgmbkOne-dimensional size of skin color detection unit.
Step A: calculating a first chroma intensity variable, Inuk＝std(u(i,j)|u(i,j)∈gmbk)。
And B: if Inu is presentkMore than or equal to Thres, set the sizef (gmb)k) 1 is ═ 1; otherwise, go to step C.
And C: calculating a second chroma-strength variable Invk＝std(v(i,j)|v(i,j)∈gmbk)。
Step D: if Invk<Thres, then set sizef (gmb)k)＝size(gmbk) (ii) a Otherwise, set sizef (gmb)k)＝1。
Example two
Fig. 2 is a block diagram of an adaptive unit analysis system for skin color detection according to a preferred embodiment of the present invention, the system including:
and the frame sequence number setting module is used for Step 0: let t equal to 1, pictRepresents the tth frame of the video, also called the current image;
the generalized block size confirming module is used for confirming the size of the generalized block according to the resolution of the current image;
wherein, QVGA, VGA, 720P are the standard sizes of the images disclosed in the industry; gmbkRepresenting the kth square block of the current image, which is called generalized block for short, wherein the initial value of k is 1; size (gmb)k) Represents a one-dimensional size of the generalized block;
a first chroma intensity variable calculation module for calculating a first chroma intensity variable, Inuk＝std(u(i,j)|u(i,j)∈gmbk)；
Wherein u (i, j) represents u chroma value of current image in ith line and jth column; std represents the mean square error; inu (Inu)kA u chroma intensity variable representing the current image;
a first chroma strength variable threshold judgment processing module for judging if InukNot less than Thres, first set the sizef (gmb)k) Setting k to be k +1, and entering a judgment processing module; otherwise, entering a second chroma intensity variable calculation module;
wherein, sizef (gmb)k) Denotes gmbkOne-dimensional size of skin color detection unit; thres represents the threshold, and is generally taken to be Thres ≦ size (gmb)k)；
A second chroma strength variable calculation module for the second chroma strength variable Invk＝std(v(i,j)|v(i,j)∈gmbk)；
Wherein v (i, j) respectively represents the v chroma value of the current image in the ith line and the jth column; invkA v chroma intensity variable representing a current image;
a second chroma strength variable threshold judgment processing module for judging if Invk<Thres, then set the sizef (gmb) firstk)＝size(gmbk) Then setting k to k + 1; otherwise, set sizef (gmb) firstk) 1, then k is set to k + 1;
the first chromaticity and the second chromaticity are randomly selected, namely the first chromaticity can be u chromaticity, the second chromaticity can be v chromaticity, the first chromaticity can be v chromaticity, and the second chromaticity can be u chromaticity;
the judgment processing module is used for judging whether k is less than or equal to num, and entering the first chroma intensity variable calculation module; otherwise, entering a next frame image judgment processing module of the current image;
num represents the number of generalized blocks of the image divided according to the size of the generalized blocks;
the next frame image judgment processing module of the current image is used for judging whether the next frame image of the current image does not exist or not, and then ending; otherwise, let t be t +1, and set the next frame of the current image as the current image, then enter the block statistical variable calculation module;
a block statistical variable calculation module for first size (gmb)k) The previous frame of the current image and the current image are divided into blocks, and then block statistical variables of each block are calculated
tik＝std(y(i,j)-py(i,j)|y(i,j)∈gmbkAnd py (i, j) is within pg mbk)；
Wherein y (i, j) is pictThe luminance value of the ith row and the jth column; py (i, j) is pictThe ith row and the jth column of the previous frame; pg mbkRepresents pictThe k-th square block of the previous frame; tikDenotes gmbkThe block statistical variable of (1);
the time similarity identification module is used for identifying the time similarity of all the blocks of the current image, and the specific method comprises the following steps:
wherein, Thres1Indicates a first decision threshold, which may be Thres18 ═ 8 × (1+24/fps), fps denotes the video capture frame rate; note (r) notekDenotes gmbkA temporal similarity identifier of;
the time similarity identifier judging module is used for judging whether the time similarity identifiers of all the blocks of the current image are 0 or not, and reentering the first chroma intensity variable calculating module; otherwise, entering a one-dimensional size calculation module of the block skin color detection unit.
And the block skin color detection unit one-dimensional size calculation module is used for calculating the one-dimensional sizes of skin color detection units of all blocks of the current image according to the time similarity identifiers of the blocks, and then reentering the next frame image judgment processing module of the current image.
The first mode is as follows: is suitable for notek1-block
If notekSet the sizef (gmb) as 1k)＝sizef(pgmbk)
Wherein, sizef (pgmb)k) Expression of pgmbkOne-dimensional size of skin color detection unit.
And a second mode: is suitable for notek0 block
Step A: calculating a first chroma intensity variable, Inuk＝std(u(i,j)|u(i,j)∈gmbk)。
And B: if Inu is presentkMore than or equal to Thres, set the sizef (gmb)k) 1 is ═ 1; otherwise, go to step C.
And C: calculating a second chroma-strength variable Invk＝std(v(i,j)|v(i,j)∈gmbk)。
Step D: if Invk<Thres, then set sizef (gmb)k)＝size(gmbk) (ii) a Otherwise, set sizef (gmb)k)＝1。
It will be understood by those skilled in the art that all or part of the steps in the method according to the above embodiments may be implemented by hardware related to program instructions, and the program may be stored in a computer readable storage medium, such as ROM, RAM, magnetic disk, optical disk, etc.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.
Claims (8)
1. A skin color detection self-adaptive unit analysis method is characterized by comprising the following steps:
step 0: let t be 1; t represents a video frame sequence number, also called a current frame sequence number;
step 1: determining the size of the generalized block according to the resolution of the current image;
step2 calculating a first chroma intensity variable Inuk，InukA u chroma intensity variable representing the current image;
step 3: if Inu is presentkNot less than Thres, first set the sizef (gmb)k) Setting k to k +1, and then proceeding to Step 6; otherwise, go to Step 4;
wherein, sizef (gmb)k) Denotes gmbkOne-dimensional size of skin color detection unit; gmbkRepresents the kth square block of the current image, Thres represents the threshold value, Thres is less than or equal to size (gmb)k)，size(gmbk) Represents a one-dimensional size of the generalized block;
step 4: calculating a second chroma-strength variable Invk，InvkA v chroma intensity variable representing a current image;
step 5: if Invk<Thres, then set the sizef (gmb) firstk)＝size(gmbk) Then setting k to k + 1; otherwise, set sizef (gmb) firstk) 1, then k is set to k + 1;
step 6: if k is less than or equal to num, then the Step2 is entered again; otherwise, go to Step 7;
num represents the number of generalized blocks of the image divided according to the size of the generalized blocks;
step 7: if the next frame image of the current image does not exist, ending; otherwise, let t be t +1, and set the next frame of the current image as the current image, and then go to Step 8;
step 8: first press size (gmb)k) The previous frame of the current image and the current image are divided into blocks according to the size of the current image, and then block statistical variables of each block are calculated;
step 9: for current picture pictAll blocks of the time similarity identification; pictRepresents the tth frame of the video, also called the current image;
step 10: if the time similarity identifiers of all the blocks of the current image are 0, then the Step2 is re-entered; otherwise, go to Step 11;
step 11: the one-dimensional size of the skin tone detection unit of all blocks of the current image is first calculated based on the temporal similarity identifier of the block, and then Step7 is re-entered.
2. The skin color detection adaptive unit analysis method of claim 1,
the determining the size of the generalized block according to the resolution of the current image specifically includes:
wherein, QVGA, VGA, 720P are the standard sizes of the images disclosed in the industry; gmbkRepresenting the kth square block of the current image, which is called generalized block for short, wherein the initial value of k is 1; size (gmb)k) Represents a one-dimensional size of the generalized block;
calculating a first chroma intensity variable InukThe method specifically comprises the following steps:
Inuk＝std(u(i,j)|u(i,j)∈gmbk)；
wherein u (i, j) represents u chroma value of current image in ith line and jth column; std represents the mean square error; inu (Inu)kA u chroma intensity variable representing the current image;
the calculated second chroma intensity variable InvkThe method specifically comprises the following steps:
Invk＝std(v(i,j)|v(i,j)∈gmbk)；
wherein v (i, j) respectively represents the v chroma value of the current image in the ith line and the jth column; invkA v chroma intensity variable representing a current image;
said first according to size (gmb)k) The previous frame of the current image and the current image are divided into blocks, and then the block statistical variable of each block is calculated as follows:
tik＝std(y(i,j)-py(i,j)|y(i,j)∈gmbkand py (i, j) is within pg mbk)；
Wherein y (i, j) is pictThe luminance value of the ith row and the jth column; py (i, j) is pictThe ith row and the jth column of the previous frame; pg mbkRepresents pictThe k-th square block of the previous frame; tikDenotes gmbkThe block statistical variable of (2).
3. The skin color detection adaptive unit analysis method of claim 2,
the specific method for identifying the time similarity of all the blocks of the current image is as follows:
notek＝sign(tik,Thres1)，
wherein, Thres1Denotes a first decision threshold, Thres18 ═ 8 × (1+24/fps), fps denotes the video capture frame rate; note (r) notekDenotes gmbkThe time similarity identifier of (a).
4. The skin color detection adaptive unit analysis method of claim 3,
firstly, according to the time similarity identifier of the block, calculating the one-dimensional sizes of the skin color detection units of all the blocks of the current image specifically as follows:
the first mode is as follows: is suitable for notekA block of 1;
if notekSet the sizef (gmb) as 1k)＝sizef(pgmbk)；
Wherein, sizef (pgmb)k) Expression of pgmbkOne-dimensional size of skin color detection unit;
and a second mode: is suitable for notekA block of 0;
step A: calculating a first chroma intensity variable, Inuk＝std(u(i,j)|u(i,j)∈gmbk)；
And B: if Inu is presentkMore than or equal to Thres, set the sizef (gmb)k) 1 is ═ 1; otherwise, entering the step C;
and C: calculating a second chroma-strength variable Invk＝std(v(i,j)|v(i,j)∈gmbk)；
Step D: if Invk<Thres, then set sizef (gmb)k)＝size(gmbk) (ii) a Otherwise, set sizef (gmb)k)＝1。
5. The skin color detection adaptive unit analysis method of claim 4, wherein the first chromaticity and the second chromaticity are randomly selected;
if the first chromaticity is set as v chromaticity and the second chromaticity is set as u chromaticity, u and v in Step1-Step11 are interchanged.
6. An adaptive unit analysis system for skin color detection, the system comprising:
and the frame sequence number setting module is used for Step 0: let t equal to 1, pictRepresents the tth frame of the video, also called the current image;
the generalized block size confirming module is used for confirming the size of the generalized block according to the resolution of the current image;
wherein, QVGA, VGA, 720P are the standard sizes of the images disclosed in the industry; gmbkRepresenting the kth square block of the current image, which is called generalized block for short, wherein the initial value of k is 1; size (gmb)k) Represents a one-dimensional size of the generalized block;
a first chroma intensity variable calculation module for calculating a first chroma intensity variable, Inuk＝std(u(i,j)|u(i,j)∈gmbk)；
Wherein u (i, j) represents u chroma value of current image in ith line and jth column; std represents the mean square error; inu (Inu)kA u chroma intensity variable representing the current image;
a first chroma strength variable threshold judgment processing module for judging if InukNot less than Thres, first set the sizef (gmb)k) Setting k to be k +1, and entering a judgment processing module; otherwise, entering a second chroma intensity variable calculation module;
wherein, sizef (gmb)k) Denotes gmbkOne-dimensional size of skin color detection unit; thres represents the threshold, Thres ≦ size (gmb)k)；
A second chroma strength variable calculation module for the second chroma strength variable Invk＝std(v(i,j)|v(i,j)∈gmbk)；
Wherein v (i, j) respectively represents the v chroma value of the current image in the ith line and the jth column; invkA v chroma intensity variable representing a current image;
a second chroma strength variable threshold judgment processing module for judging if Invk<Thres, then set the sizef (gmb) firstk)＝size(gmbk) Then setting k to k + 1; otherwise, set sizef (gmb) firstk) 1, then k is set to k + 1;
the judgment processing module is used for judging whether k is less than or equal to num, and entering the first chroma intensity variable calculation module; otherwise, entering a next frame image judgment processing module of the current image;
num represents the number of generalized blocks of the image divided according to the size of the generalized blocks;
the next frame image judgment processing module of the current image is used for judging whether the next frame image of the current image does not exist or not, and then ending; otherwise, let t be t +1, and set the next frame of the current image as the current image, then enter the block statistical variable calculation module;
a block statistical variable calculation module for first size (gmb)k) The previous frame of the current image and the current image are divided into blocks, and then a block statistical variable ti of each block is calculatedk＝std(y(i,j)-py(i,j)|y(i,j)∈gmbkAnd py (i, j) is within pg mbk)；
Wherein y (i, j) is pictThe luminance value of the ith row and the jth column; py (i, j) is pictThe ith row and the jth column of the previous frame; pg mbkRepresents pictPrevious frame ofThe kth square block; tikDenotes gmbkThe block statistical variable of (1);
the time similarity identification module is used for identifying the time similarity of all the blocks of the current image;
the specific method comprises the following steps:
notek＝sign(tik,Thres1)，
wherein, Thres1Denotes a first decision threshold, Thres18 ═ 8 × (1+24/fps), fps denotes the video capture frame rate; note (r) notekDenotes gmbkA temporal similarity identifier of;
the time similarity identifier judging module is used for judging whether the time similarity identifiers of all the blocks of the current image are 0 or not, and reentering the first chroma intensity variable calculating module; otherwise, entering a one-dimensional size calculation module of a block skin color detection unit;
and the block skin color detection unit one-dimensional size calculation module is used for calculating the one-dimensional sizes of skin color detection units of all blocks of the current image according to the time similarity identifiers of the blocks, and then reentering the next frame image judgment processing module of the current image.
7. The skin tone detection adaptive unit analysis system of claim 6,
the block skin color detection unit one-dimensional size calculation module is configured to calculate, according to the time similarity identifier of the block, one-dimensional sizes of skin color detection units of all blocks of the current image, specifically:
the first mode is as follows: is suitable for notekA block of 1;
if notekSet the sizef (gmb) as 1k)＝sizef(pgmbk)；
Wherein, sizef (pgmb)k) Expression of pgmbkOne-dimensional size of skin color detection unit;
and a second mode: is suitable for notekA block of 0;
step A: calculating a first chroma intensity variable, Inuk＝std(u(i,j)|u(i,j)∈gmbk)；
And B: if Inu is presentkMore than or equal to Thres, set the sizef (gmb)k) 1 is ═ 1; otherwise, entering the step C;
and C: calculating a second chroma-strength variable Invk＝std(v(i,j)|v(i,j)∈gmbk)；
Step D: if Invk<Thres, then set sizef (gmb)k)＝size(gmbk) (ii) a Otherwise, set sizef (gmb)k)＝1。
8. The skin tone detection adaptive unit analysis system of claim 7,
if the first chromaticity is set as v chromaticity and the second chromaticity is set as u chromaticity, u and v in the above claim 7 are interchanged.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710650617.1A CN107480617B (en) | 2017-08-02 | 2017-08-02 | Skin color detection self-adaptive unit analysis method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710650617.1A CN107480617B (en) | 2017-08-02 | 2017-08-02 | Skin color detection self-adaptive unit analysis method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107480617A CN107480617A (en) | 2017-12-15 |
| CN107480617B true CN107480617B (en) | 2020-03-17 |
Family
ID=60597502
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710650617.1A Active CN107480617B (en) | 2017-08-02 | 2017-08-02 | Skin color detection self-adaptive unit analysis method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107480617B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1751522A (en) * | 2002-12-18 | 2006-03-22 | 罗伯特·戈登大学 | Video encoding with skipping motion estimation for selected macroblocks |
| CN101650830A (en) * | 2009-08-06 | 2010-02-17 | 中国科学院声学研究所 | Compressed domain video lens mutation and gradient union automatic segmentation method and system |
-
2017
- 2017-08-02 CN CN201710650617.1A patent/CN107480617B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1751522A (en) * | 2002-12-18 | 2006-03-22 | 罗伯特·戈登大学 | Video encoding with skipping motion estimation for selected macroblocks |
| CN101650830A (en) * | 2009-08-06 | 2010-02-17 | 中国科学院声学研究所 | Compressed domain video lens mutation and gradient union automatic segmentation method and system |
Non-Patent Citations (2)
| Title |
|---|
| 基于肤色的人脸检测及其在红眼消除中的应用研究;纪晓宁;《中国优秀硕士学位论文全文数据库信息科技辑》;20120315(第3期);第I138-2041页 * |
| 自适应方向插值及相似性约束人脸区域的H.264视频流错误隐藏研究;张小红 等;《计算机应用研究》;20150831(第08期);第1-3页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107480617A (en) | 2017-12-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110378264B (en) | Target tracking method and device | |
| CN108921782B (en) | Image processing method, device and storage medium | |
| CN112308095A (en) | Picture preprocessing and model training method and device, server and storage medium | |
| CN111639564B (en) | Video pedestrian re-identification method based on multi-attention heterogeneous network | |
| WO2023082784A1 (en) | Person re-identification method and apparatus based on local feature attention | |
| CN110458115B (en) | Multi-frame integrated target detection algorithm based on time sequence | |
| CN110738160A (en) | human face quality evaluation method combining with human face detection | |
| CN111008608B (en) | Night vehicle detection method based on deep learning | |
| CN112381104A (en) | Image identification method and device, computer equipment and storage medium | |
| WO2014036813A1 (en) | Method and device for extracting image features | |
| WO2022213540A1 (en) | Object detecting, attribute identifying and tracking method and system | |
| CN111382647B (en) | Picture processing method, device, equipment and storage medium | |
| CN111414938B (en) | Target detection method for bubbles in plate heat exchanger | |
| WO2023169318A1 (en) | Image quality determination method, apparatus, device, and storage medium | |
| CN107480617B (en) | Skin color detection self-adaptive unit analysis method and system | |
| CN112565674A (en) | Exhibition hall central control system capable of realizing remote video monitoring and control | |
| CN107480616B (en) | Skin color detection unit analysis method and system based on image analysis | |
| CN112487927B (en) | Method and system for realizing indoor scene recognition based on object associated attention | |
| CN111708907B (en) | Target person query method, device, equipment and storage medium | |
| CN115239551A (en) | Video enhancement method and device | |
| CN110443244B (en) | Graphics processing method and related device | |
| CN113269205A (en) | Video key frame extraction method and device, electronic equipment and storage medium | |
| CN112579824A (en) | Video data classification method and device, electronic equipment and storage medium | |
| CN109493349B (en) | Image feature processing module, augmented reality equipment and corner detection method | |
| CN108764106B (en) | Multi-scale color image face comparison method based on cascade structure |
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 | ||
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 518057 Guangdong city of Shenzhen province Nanshan District Guangdong streets high in the four Longtaili Technology Building Room 325 No. 30 Patentee after: Shenzhen mengwang video Co., Ltd Address before: 518057 Guangdong city of Shenzhen province Nanshan District Guangdong streets high in the four Longtaili Technology Building Room 325 No. 30 Patentee before: SHENZHEN MONTNETS ENCYCLOPEDIA INFORMATION TECHNOLOGY Co.,Ltd. |