CN106156700B - The extracting method and device of characteristics of human body in a kind of image procossing - Google Patents

The extracting method and device of characteristics of human body in a kind of image procossing Download PDF

Info

Publication number
CN106156700B
CN106156700B CN201510149262.9A CN201510149262A CN106156700B CN 106156700 B CN106156700 B CN 106156700B CN 201510149262 A CN201510149262 A CN 201510149262A CN 106156700 B CN106156700 B CN 106156700B
Authority
CN
China
Prior art keywords
point
probability
gravity
center
human
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
Application number
CN201510149262.9A
Other languages
Chinese (zh)
Other versions
CN106156700A (en
Inventor
余大勇
马昆
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
BEIJING SUMAVISION TECHNOLOGIES Co Ltd
Original Assignee
BEIJING SUMAVISION TECHNOLOGIES Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by BEIJING SUMAVISION TECHNOLOGIES Co Ltd filed Critical BEIJING SUMAVISION TECHNOLOGIES Co Ltd
Priority to CN201510149262.9A priority Critical patent/CN106156700B/en
Publication of CN106156700A publication Critical patent/CN106156700A/en
Application granted granted Critical
Publication of CN106156700B publication Critical patent/CN106156700B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Abstract

The present invention relates to the extracting methods and device of characteristics of human body in a kind of image procossing, this method comprises:, for the human body of central point to be extracted, concentrating from the basic point of the people's body region in body image, choose growth point set;Any point concentrated for growing point, using the point as seed point, determines the seed region of the seed point when the point meets preset condition;For each seed region, the center of gravity of the seed region is calculated;For each center of gravity, the prediction probability for the central point that the center of gravity is human body is calculated;According to the prediction probability for calculating each center of gravity obtained, the central point of human body is determined.The method provided through the invention can be improved the accuracy for extracting central point, and meet the speed requirement that real time processed images extract center.

Description

The extracting method and device of characteristics of human body in a kind of image procossing
Technical field
The present invention relates to the extracting methods and dress of characteristics of human body in field of image processing more particularly to a kind of image procossing It sets.
Background technique
Body-sensing technology is not only applicable to field of play, it may also be used for the fields such as security, health medical treatment, amusement shopping. In short, the development of body-sensing technology, brings revolutionary change to man-machine interaction mode.
Body-sensing technology manipulates somatosensory device by identifying the motion characteristic of human body, and in the process of identification maneuver feature In, need to extract the central point at each position of human body.For example, in optical sensing, it is necessary first to identify human body, it is subdivided go out Then each position (such as head, trunk, ancon, hand) of human body extracts central point as knowledge from each position The parameter of other motion feature.It can be seen that the extraction of human body central point is a key technology in body-sensing technology, central point is mentioned The accuracy taken directly affects the accuracy to motion characteristic identification.And body-sensing technology is that somatosensory device operator passes through limb Body acts controlling equipment, then the extraction rate for position central point just requires.
In the mainly type heart method that in body-sensing technology, can satisfy real-time processing speed demand at present.Type this method Process object is body image, and this method will belong to each pixel at same position coordinate in body image is added, then It averages, then using the average value acquired as the central point at the position.
However, type heart method extracts the accuracy of central point, it is severely limited by the accuracy of human body classification.This is because In currently used human body sorting technique based on probability, the classification at position is inaccurate.In general, a position is by belonging to The probability value at the position is not 0 pixel composition, and in practice, a pixel tends to belong to several positions, therefore, There is no specific boundaries between position and position.And type heart method does not account for this inaccurate factor of position classification, and simple Only with the coordinate of pixel be single parameter cause the accuracy of the extraction result of central point to need to extract central point It improves.
Summary of the invention
The object of the present invention is to provide the extracting methods and device of characteristics of human body in a kind of image procossing, to overcome related skill Human body central point extracts the low problem of accuracy in art.
On the one hand, the present invention provides the extracting method of characteristics of human body in image procossing a kind of, comprising:
For the body image obtained in advance, each pixel calculated in the body image belongs to each human body Probability;
For the human body of any central point to be extracted, the probability by belonging to the human body is greater than the first default threshold The point of value forms the basic point set of the human body;
It is concentrated from the basic point, acquisition probability is greater than the point of the second preset threshold, growth point set is formed, wherein described the Two preset thresholds are greater than first preset threshold;
For any point that the growing point is concentrated, when the point meets preset condition, executes following operation: being with the point Seed point obtains seed region corresponding with the seed point by region-growing method;The preset condition includes: not to be selected as For seed point and not in the seed region of any sub- point;
For each seed region of the human body, the probability of each of seed region point is considered as this The density of point and the density put according to each in the seed region and each position of point in the body image Coordinate calculates the center of gravity of the seed region;
For the center of gravity of each seed region, is concentrated according to the basic point and be less than or equal to the pixel distance of the center of gravity The probability of the point of first pre-determined distance calculates the prediction probability for the central point that the center of gravity is the human body;
The point for concentrating the center of gravity of range prediction maximum probability nearest basic point, is determined as the center of the human body Point.
On the other hand, the present invention also provides the extraction element of characteristics of human body in image procossing a kind of, described device includes:
Probability evaluation entity, for calculating each pixel in the body image for the body image obtained in advance Point belongs to the probability of each human body;
Basic point set obtains module, for being directed to the human body of any central point to be extracted, by belonging to the human body portion The point that the probability of position is greater than the first preset threshold forms the basic point set of the human body;
Point set determining module is grown, for concentrating from the basic point, acquisition probability is greater than the point of the second preset threshold, shape At growth point set, wherein second preset threshold is greater than first preset threshold;
Seed region forms module, any point for concentrating for the growing point, when the point meets preset condition, It executes following operation: using the point as seed point, seed region corresponding with the seed point being obtained by region-growing method;It is described pre- If condition includes: not to be selected as seed point and not in the seed region of any sub- point;
Center of gravity calculation module will be every in the seed region for being directed to each seed region of the human body The probability of one point is considered as the density of the point and the density put according to each in the seed region and each point are in institute The position coordinates in body image are stated, the center of gravity of the seed region is calculated;
Prediction probability computing module is concentrated and is somebody's turn to do according to the basic point for being directed to the center of gravity of each seed region The pixel distance of center of gravity is less than or equal to the probability of the point of the first pre-determined distance, calculates the central point that the center of gravity is the human body Prediction probability;
Central point determining module, the point for concentrating the center of gravity of range prediction maximum probability nearest basic point, is determined as The central point of the human body.
The present invention at least has the advantages that the embodiment of the present invention, introduces the human body portion for belonging to central point to be extracted The point for belonging to human body is divided to different seed zones as the parameter for extracting central point, and by seed point by the probability of position Domain obtains the center of gravity of each seed region then according to seed region, according further to the probability put around center of gravity, determines Out most possibly close to the center of gravity of central point, so that the central point finally obtained is closely connected with probability, to reduce based on general Influence of the human body inaccuracy that the position sorting technique of rate is got to central point is extracted, improves and extracts the accurate of central point Property, and the method for extraction central point of the invention is easy, can satisfy the processing speed demand handled in real time.
It should be understood that above general description and following detailed description be only it is exemplary and explanatory, not It can the limitation present invention.
Detailed description of the invention
Fig. 1 is one of the exemplary process diagram of extracting method of characteristics of human body in image procossing in the embodiment of the present invention;
Fig. 2 is to divide preset range in the embodiment of the present invention to show schematic diagram;
Fig. 3 is two of the exemplary process diagram of the extracting method of characteristics of human body in image procossing in the embodiment of the present invention;
Fig. 4 is one of the schematic diagram of extraction element of characteristics of human body in image procossing in the embodiment of the present invention;
Fig. 5 is two of the schematic diagram of the extraction element of characteristics of human body in image procossing in the embodiment of the present invention.
Specific embodiment
The embodiment of the present invention is illustrated below in conjunction with attached drawing, it should be understood that embodiment described herein is only used In the description and interpretation present invention, it is not intended to limit the present invention.
Example embodiments are described in detail here, and the example is illustrated in the accompanying drawings.Following description is related to When attached drawing, unless otherwise indicated, the same numbers in different drawings indicate the same or similar elements.Following exemplary embodiment Described in embodiment do not represent all embodiments consistented with the present invention.On the contrary, they be only with it is such as appended The example of device and method being described in detail in claims, some aspects of the invention are consistent.
Although fast for the medium-sized heart method processing speed of the prior art, the demand of real-time processing speed, the party can satisfy The central point accuracy problem to be improved for the human body that method is extracted, the embodiment of the present invention provide people in a kind of image procossing The extracting method of body characteristics, for extracting the central point of human body, this method can not only improve in extraction with respect to type heart method The accuracy of heart point, additionally it is possible to meet the needs of real time handling requirement processing speed is fast.
Firstly, in image procossing provided in an embodiment of the present invention characteristics of human body extracting method, by the point in body image The probability for belonging to the human body of central point to be extracted takes into account, for determining the position of human body central point, so as to Enough reduce position classification inaccuracy to extract central point influence, so as to improve extract human body central point it is accurate Property.
In addition, in the embodiment of the present invention, before the central point for extracting human body according to probability, first to belonging to the human body The point at position is pre-processed, such as filters out the small point of probability and isolated point, by reducing the point for participating in extracting central point Quantity reduces calculation amount, can be improved the speed for extracting central point, by filter out belong to the small point of probability of human body can also It is enough to improve the accuracy for extracting central point.
Below by simple embodiment, the extracting method of characteristics of human body in image procossing in the embodiment of the present invention is carried out It is described in detail.
Embodiment one
As shown in Figure 1, for the exemplary stream of the extracting method of characteristics of human body in image procossing provided in an embodiment of the present invention Cheng Tu, method includes the following steps:
Step 101: for the body image obtained in advance, calculate each pixel in the body image belong to it is each The probability of human body.
Step 102: for the human body of any central point to be extracted, the probability by belonging to the human body is greater than the The point of one preset threshold forms the basic point set of the human body.
Step 103: it is concentrated from the basic point, acquisition probability is greater than the point of the second preset threshold, growth point set is formed, Described in the second preset threshold be greater than first preset threshold.
Step 104: any point concentrated for the growing point executes following operation when the point meets preset condition: Using the point as seed point, seed region corresponding with the seed point is obtained by region-growing method;The preset condition includes: not Once it was selected as seed point and not in the seed region of any sub- point.
Step 105: for each seed region of the human body, by the general of each of seed region point Rate is considered as the density of the point and the density put according to each in the seed region and each point are in the body image In position coordinates, calculate the center of gravity of the seed region.
Wherein, in one embodiment can according to the following formula (1) calculate seed region center of gravity:
Wherein, G=Σ Gi (1)
Wherein, in formula (1), XiIndicate i-th point of seed region of abscissa;YiIndicate i-th point of seed region Ordinate;GiIndicate i-th point of seed region of probability;Xc indicates the abscissa of the center of gravity of seed region;Yc indicates seed zone The ordinate of the center of gravity in domain;The sum of the probability of all the points of G expression seed region;* it indicates to seek product.
Step 106: for the center of gravity of each seed region, the pixel distance with the center of gravity being concentrated according to the basic point Less than or equal to the probability of the point of the first pre-determined distance, the prediction probability for the central point that the center of gravity is the human body is calculated.
Step 107: the point for concentrating the center of gravity of range prediction maximum probability nearest basic point is determined as the human body Central point.
In the following, above steps is described in detail:
1) seed region corresponding with the seed point, wherein, is obtained by region-growing method in step 104, tool can be specific It executes are as follows: concentrated from the basic point, acquisition probability is greater than third predetermined threshold value and is less than with the pixel distance of the seed point The point of second pre-determined distance forms seed region;Wherein, the third predetermined threshold value is less than second preset threshold.
Wherein, the forming process of seed region for ease of understanding carries out the process below by step A1 to step A2 For example:
Step A1: it after obtaining growth point set, is concentrated optionally a bit from growing point, as seed point, is subordinated to the human body The basic point at position is concentrated, acquisition probability be greater than third predetermined threshold value and by the basis of the position in the body image, with The pixel distance of seed point forms the seed region of seed point less than the point of the second pre-determined distance.
Step A2: and then for any point that growing point is concentrated, when the point be not selected as seed point and was not selecting When doing in the seed region of seed point, using the point as seed point, the basic point for being subordinated to the human body is concentrated, and is obtained general Rate be greater than third predetermined threshold value and by the basis of the position in the body image, with the pixel distance of seed point less than the The point of two pre-determined distances, forms the seed region of seed point.
Wherein, when seed point is concentrated, there is no be not elected to be for seed point and not in the seed region of any sub- point When point, the division of seed region is just completed.
Wherein, by the basis of the position in the body image, with the pixel distance of seed point it is default less than second away from From point be, for example: according to position of the seed point in body image, concentrated from basic point, it is finding with seed point pixel Distance is the point of presetted pixel point.
2), wherein step 106 can be executed specifically as following steps:
Step B1: for the center of gravity of each seed region, concentrated from the basic point obtain with the pixel of the center of gravity away from Probability from the point for being less than or equal to the first pre-determined distance.
Step B2: calculate the probability for being less than or equal to the point of the first pre-determined distance with the pixel distance of the center of gravity and, will count Prediction probability of the sum calculated as the center of gravity.
Wherein, using probability and value be used as the prediction probability of center of gravity, the center of gravity can be calculated simply and easily as human body portion The probability size of the central point of position improves calculating speed.
Wherein, naturally it is also possible to the prediction probability of each center of gravity is calculated using other methods, for example, when preset range is with weight Centered on the heart, using the distance of n (n is positive integer) a pixel as the border circular areas of radius when, by the border circular areas according to distance The distance of center of gravity is divided into the circular annular region of preset quantity.For example, as shown in Fig. 2, first pre-determined distance is pressed when n is 3 It is pre- by first respectively using the distance of 1 pixel, 2 pixels and 3 pixels as the circle of radius according to centered on center of gravity O If the range of distance marks off 3 circular annular regions, these three circular annular regions are respectively region 1, (the black mark in Fig. 2 of region 2 The circular annular region of knowledge) and region 3.Then weight is set for each region, wherein region 1 indicates, the area Ze Gai nearest apart from center of gravity The corresponding maximum weight in domain, region 3 is farthest apart from center of gravity, then the weight in region 3 is minimum.It, will to any point in preset range The point belongs to new probability of the probability of human body multiplied by the weight of the region as the point, then will be in preset range Each point new probability prediction probability of the sum as center of gravity.Due to bigger apart from the nearlyr weight of center of gravity, can increase in this way The effect of the probability of point around the heart.It is of course also possible to calculate the pre- of the central point that center of gravity is human body using other methods Probability is surveyed, which is not limited by the present invention.
3), wherein, in step 107, when putting the point of concentration based on the maximum center of gravity of prediction probability, basic point concentrate away from The point nearest from the center of gravity is center of gravity itself.It include using the maximum center of gravity of prediction probability as basic point set in step 107 Center of gravity the case where, as a result, by step 107, the central point finally obtained can be on human body.In this way, even if base In probability position sorting technique divide position inaccuracy (such as even if due to environment or clothing influence, a position quilt At least two parts are divided into, or due to being influenced by acting, when occurring cavity among a position), it is determined according to step 107 Central point can fall on human body, to improve the accuracy of determining central point.
4) it, is calculated to simplify, and optimizes the basic point set for extracting central point, in the embodiment of the present invention, in step 103 form before growth point set, can also filter out the basic point and concentrate, non-conterminous with other points of the human body Point.In this way, can not only reduce the data volume for extracting central point, additionally it is possible to optimize the basic number for extracting central point According to improve the accuracy for extracting central point.
To sum up, the embodiment of the present invention introduces the probability for belonging to the human body of central point to be extracted as extraction central point Parameter, and the point for belonging to human body is divided to by different seed regions by seed point, then according to seed region, obtained The center of gravity of each seed region determines the weight most possibly close to central point according further to the probability put around center of gravity The heart, so that the central point finally obtained is closely connected with probability, to reduce the people that position sorting technique based on probability is got Influence of the body region inaccuracy to central point is extracted improves the accuracy for extracting central point, and extraction central point of the invention Method it is easy, can satisfy the processing speed demand handled in real time.
Embodiment two
As shown in figure 3, by taking the central point for extracting head as an example, to human body in the image procossing provided in the embodiment of the present invention The extracting method of feature is illustrated, method includes the following steps:
Step 301: in the body image obtained in advance, obtain belong to head probability be not 0 point, form head Initial point set.
Step 302: filtering out initial point concentration, probability is equal to the point of the first preset threshold less than the, and filters out and head Other non-conterminous points of point, basis of formation point set.
Step 303: being concentrated from the basic point, acquisition probability is greater than the point of the second preset threshold, forms growth point set.
Step 304: any point concentrated for the growing point executes following operation when the point meets preset condition: Using the point as seed point, seed region corresponding with the seed point is obtained by region-growing method;The preset condition includes: not Once it was selected as seed point and not in the seed region of any sub- point.
Step 305: for each seed region on the head, the probability of each of seed region point being regarded For the point density and according in the seed region each put density and each point in the body image Position coordinates calculate the center of gravity of the seed region.
Step 306: for the center of gravity of each seed region, concentrated from the basic point obtain with the pixel of the center of gravity away from Probability from the point for being less than or equal to the first pre-determined distance, and the sum of the probability of the point of acquisition is calculated, the sum of calculating is heavy as this The heart is the prediction probability of the central point on head.
Step 307: the point for concentrating the center of gravity of range prediction maximum probability nearest basic point is determined as the center on head Point.
The embodiment of the present invention by according to belonging to the probability on head, and belongs to the position coordinates extraction of the point on head and lifts one's head The central point in portion improves so as to reduce influence of the position classification inaccuracy to central point is extracted and extracts the accurate of central point Property.
Embodiment three
Based on identical inventive concept, the embodiment of the present invention also provides the extraction dress of characteristics of human body in image procossing a kind of It sets, as shown in figure 4, described device includes:
Probability evaluation entity 401, for calculating each picture in the body image for the body image obtained in advance Vegetarian refreshments belongs to the probability of each human body;
Basic point set obtains module 402, for being directed to the human body of any central point to be extracted, by belonging to the human body The point that the probability at position is greater than the first preset threshold forms the basic point set of the human body;
Point set determining module 403 is grown, for concentrating from the basic point, acquisition probability is greater than the second preset threshold Point forms growth point set, wherein second preset threshold is greater than first preset threshold;
Seed region forms module 404, any point for concentrating for the growing point, when the point meets preset condition When, it executes following operation: using the point as seed point, seed region corresponding with the seed point being obtained by region-growing method;Institute Stating preset condition includes: not to be selected as seed point and not in the seed region of any sub- point;
Center of gravity calculation module 405 will be in the seed region for being directed to each seed region of the human body The probability of each point is considered as the density of the point and the density put according to each in the seed region and each point exist Position coordinates in the body image calculate the center of gravity of the seed region;
Prediction probability computing module 406, for be directed to each seed region center of gravity, according to the basic point concentrate with The pixel distance of the center of gravity is less than or equal to the probability of the point of the first pre-determined distance, calculates the center that the center of gravity is the human body The prediction probability of point;
Central point determining module 407, the point for concentrating the center of gravity of range prediction maximum probability nearest basic point, determines For the central point of the human body.
Wherein, in one embodiment, the seed region forms module, is specifically used for concentrating from the basic point, obtain It takes probability to be greater than third predetermined threshold value and the point with the pixel distance of the seed point less than the second pre-determined distance, forms seed Region;Wherein, the third predetermined threshold value is less than second preset threshold.
Wherein, in one embodiment, as shown in figure 5, described device further include:
Module 408 is filtered out, is concentrated for the growth point set determining module from the basic point, acquisition probability is greater than second The point of preset threshold is formed before growth point set, is filtered out the basic point and is concentrated, non-conterminous with other points of the human body Point.
Wherein, in one embodiment, as shown in figure 5, the prediction probability computing module 406, specifically includes:
Probability acquiring unit 409 is concentrated from the basic point for being directed to the center of gravity of each seed region and obtains and be somebody's turn to do The pixel distance of center of gravity is less than or equal to the probability of the point of the first pre-determined distance;
Prediction probability computing unit 410 is less than or equal to the first pre-determined distance with the pixel distance of the center of gravity for calculating The sum of the probability of point, using the sum of calculating as the prediction probability of the center of gravity.
About the device in above-described embodiment, wherein modules execute the concrete mode of operation in related this method Embodiment in be described in detail, no detailed explanation will be given here.
It should be understood by those skilled in the art that, the embodiment of the present invention can provide as method, system or computer program Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the present invention Apply the form of example.Moreover, it wherein includes the computer of computer usable program code that the present invention, which can be used in one or more, The computer program implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) produces The form of product.
The present invention be referring to according to the method for the embodiment of the present invention, the process of equipment (system) and computer program product Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates, Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one The step of function of being specified in a box or multiple boxes.
Although preferred embodiments of the present invention have been described, it is created once a person skilled in the art knows basic Property concept, then additional changes and modifications may be made to these embodiments.So it includes excellent that the following claims are intended to be interpreted as It selects embodiment and falls into all change and modification of the scope of the invention.
Obviously, various changes and modifications can be made to the invention without departing from essence of the invention by those skilled in the art Mind and range.In this way, if these modifications and changes of the present invention belongs to the range of the claims in the present invention and its equivalent technologies Within, then the present invention is also intended to include these modifications and variations.

Claims (8)

1. the extracting method of characteristics of human body in a kind of image procossing, which is characterized in that the described method includes:
For the body image obtained in advance, each pixel calculated in the body image belongs to the general of each human body Rate;
For the human body of any central point to be extracted, the probability by belonging to the human body is greater than the first preset threshold Point forms the basic point set of the human body;
It is concentrated from the basic point, acquisition probability is greater than the point of the second preset threshold, growth point set is formed, wherein described second is pre- If threshold value is greater than first preset threshold;
For any point that the growing point is concentrated, when the point meets preset condition, following operation is executed: using the point as seed Point obtains seed region corresponding with the seed point by region-growing method;The preset condition includes: not to be selected as planting Sub- point and not in the seed region of any sub- point;
For each seed region of the human body, the probability of each of seed region point is considered as the point Density and the density put according to each in the seed region and each position coordinates of point in the body image, Calculate the center of gravity of the seed region;
For the center of gravity of each seed region, is concentrated according to the basic point and be less than or equal to first with the pixel distance of the center of gravity The probability of the point of pre-determined distance calculates the prediction probability for the central point that the center of gravity is the human body;
The point for concentrating the center of gravity of range prediction maximum probability nearest basic point, is determined as the central point of the human body.
2. the method according to claim 1, wherein described corresponding with the seed point by region-growing method acquisition Seed region, specifically include:
It is concentrated from the basic point, acquisition probability is greater than third predetermined threshold value and with the pixel distance of the seed point less than the The point of two pre-determined distances forms seed region;Wherein, the third predetermined threshold value is less than second preset threshold.
3. according to the method described in claim 2, acquisition probability is greater than the it is characterized in that, described concentrate from the basic point The point of two preset thresholds is formed before growth point set, the method also includes:
It filters out the basic point to concentrate, the non-conterminous point of other points with the human body.
4. the method according to claim 1, wherein the center of gravity for each seed region, according to institute It states basic point and concentrates the probability for being less than or equal to the point of the first pre-determined distance with the pixel distance of the center of gravity, it is described for calculating the center of gravity The prediction probability of the central point of human body, specifically includes:
For the center of gravity of each seed region, concentrates to obtain from the basic point and be less than or equal to the with the pixel distance of the center of gravity The probability of the point of one pre-determined distance;And
Calculate the probability for being less than or equal to the point of the first pre-determined distance with the pixel distance of the center of gravity and, using the sum of calculating as this The prediction probability of center of gravity.
5. the extraction element of characteristics of human body in a kind of image procossing, which is characterized in that described device includes:
Probability evaluation entity, for calculating each pixel category in the body image for the body image obtained in advance In the probability of each human body;
Basic point set obtains module, for being directed to the human body of any central point to be extracted, by belonging to the human body The point that probability is greater than the first preset threshold forms the basic point set of the human body;
Point set determining module is grown, for concentrating from the basic point, acquisition probability is greater than the point of the second preset threshold, forms life Long point set, wherein second preset threshold is greater than first preset threshold;
Seed region forms module, any point for concentrating for the growing point, when the point meets preset condition, executes It operates below: using the point as seed point, seed region corresponding with the seed point being obtained by region-growing method;The default item Part includes: not to be selected as seed point and not in the seed region of any sub- point;
Center of gravity calculation module, for being directed to each seed region of the human body, by each of the seed region The probability of point is considered as the density of the point and the density put according to each in the seed region and each point are in the people Position coordinates in body image calculate the center of gravity of the seed region;
Prediction probability computing module is concentrated and the center of gravity for being directed to the center of gravity of each seed region according to the basic point Pixel distance be less than or equal to the first pre-determined distance point probability, calculate the center of gravity be the human body central point it is pre- Survey probability;
Central point determining module, the point for concentrating the center of gravity of range prediction maximum probability nearest basic point are determined as described The central point of human body.
6. device according to claim 5, which is characterized in that the seed region forms module, is specifically used for from described Basic point is concentrated, and acquisition probability is greater than third predetermined threshold value and with the pixel distance of the seed point less than the second pre-determined distance Point, formed seed region;Wherein, the third predetermined threshold value is less than second preset threshold.
7. device according to claim 6, which is characterized in that described device further include:
Module is filtered out, is concentrated for the growth point set determining module from the basic point, acquisition probability is greater than the second default threshold The point of value is formed before growth point set, is filtered out the basic point and is concentrated, the non-conterminous point of other points with the human body.
8. device according to claim 5, which is characterized in that the prediction probability computing module specifically includes:
Probability acquiring unit is concentrated from the basic point and is obtained and the center of gravity for being directed to the center of gravity of each seed region Pixel distance is less than or equal to the probability of the point of the first pre-determined distance;
Prediction probability computing unit, for calculating the probability for being less than or equal to the point of the first pre-determined distance with the pixel distance of the center of gravity Sum, using the sum of calculating as the prediction probability of the center of gravity.
CN201510149262.9A 2015-03-31 2015-03-31 The extracting method and device of characteristics of human body in a kind of image procossing Active CN106156700B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510149262.9A CN106156700B (en) 2015-03-31 2015-03-31 The extracting method and device of characteristics of human body in a kind of image procossing

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510149262.9A CN106156700B (en) 2015-03-31 2015-03-31 The extracting method and device of characteristics of human body in a kind of image procossing

Publications (2)

Publication Number Publication Date
CN106156700A CN106156700A (en) 2016-11-23
CN106156700B true CN106156700B (en) 2019-06-18

Family

ID=57338089

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510149262.9A Active CN106156700B (en) 2015-03-31 2015-03-31 The extracting method and device of characteristics of human body in a kind of image procossing

Country Status (1)

Country Link
CN (1) CN106156700B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10672174B2 (en) 2018-06-28 2020-06-02 Adobe Inc. Determining image handle locations
US10621764B2 (en) 2018-07-05 2020-04-14 Adobe Inc. Colorizing vector graphic objects

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101251898A (en) * 2008-03-25 2008-08-27 腾讯科技(深圳)有限公司 Skin color detection method and apparatus
CN102005034A (en) * 2010-12-01 2011-04-06 南京大学 Remote sensing image segmentation method based on region clustering
JP2012003358A (en) * 2010-06-15 2012-01-05 Yahoo Japan Corp Background determination device, method, and program
CN103093215A (en) * 2013-02-01 2013-05-08 北京天诚盛业科技有限公司 Eye location method and device

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101251898A (en) * 2008-03-25 2008-08-27 腾讯科技(深圳)有限公司 Skin color detection method and apparatus
JP2012003358A (en) * 2010-06-15 2012-01-05 Yahoo Japan Corp Background determination device, method, and program
CN102005034A (en) * 2010-12-01 2011-04-06 南京大学 Remote sensing image segmentation method based on region clustering
CN103093215A (en) * 2013-02-01 2013-05-08 北京天诚盛业科技有限公司 Eye location method and device

Also Published As

Publication number Publication date
CN106156700A (en) 2016-11-23

Similar Documents

Publication Publication Date Title
US20200192489A1 (en) Information processing device and method, program and recording medium for identifying a gesture of a person from captured image data
KR102148502B1 (en) Automated three dimensional model generation
Rodríguez et al. Agents and computer vision for processing stereoscopic images
US10055851B2 (en) Determining dimension of target object in an image using reference object
JP6295645B2 (en) Object detection method and object detection apparatus
CN105745687B (en) Context aware Moving target detection
CN103604371B (en) Mobile terminal and object measurement method thereof
JP4396430B2 (en) Gaze guidance information generation system, gaze guidance information generation program, and gaze guidance information generation method
CN105701820B (en) A kind of point cloud registration method based on matching area
CN107341473B (en) Palm characteristic recognition method, palm characteristic identificating equipment and storage medium
CN102542289B (en) Pedestrian volume statistical method based on plurality of Gaussian counting models
JP5848551B2 (en) Learning device, learning device control method, detection device, detection device control method, and program
CN106845421A (en) Face characteristic recognition methods and system based on multi-region feature and metric learning
US10417775B2 (en) Method for implementing human skeleton tracking system based on depth data
US10659670B2 (en) Monitoring system and control method thereof
CN105825192B (en) A kind of facial expression recognizing method and system
CN104143080B (en) Three-dimensional face identifying device and method based on three-dimensional point cloud
JP5954712B2 (en) Image processing apparatus, image processing method, and program thereof
CN104346619B (en) Living body feature vector extraction device and the vectorial extracting method of living body feature
CN109076198B (en) Video-based object tracking occlusion detection system, method and equipment
CN105550678B (en) Human action feature extracting method based on global prominent edge region
JP4670976B2 (en) Learning device and method, recognition device and method, program, and recording medium
CN105405154B (en) Target object tracking based on color-structure feature
CN103988233B (en) Posture estimation device and posture estimation method
CN103221977B (en) Posture state estimation unit and posture state method of estimation

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant