CN108573216A - A kind of limbs posture judgment method and device - Google Patents
A kind of limbs posture judgment method and device Download PDFInfo
- Publication number
- CN108573216A CN108573216A CN201810229630.4A CN201810229630A CN108573216A CN 108573216 A CN108573216 A CN 108573216A CN 201810229630 A CN201810229630 A CN 201810229630A CN 108573216 A CN108573216 A CN 108573216A
- Authority
- CN
- China
- Prior art keywords
- node
- human body
- posture
- wrist
- hand
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
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
- G06V40/20—Movements or behaviour, e.g. gesture recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- Image Analysis (AREA)
Abstract
A kind of limbs posture judgment method of offer of the embodiment of the present invention and device, to realize that accurate and effective limbs posture judges.Wherein, limbs posture judgment method includes：Determine that multiple human body nodes of the first human body in human body image, multiple human body nodes include at least multiple nodes on wrist node, elbow node, shoulder node and head and neck；According to the wrist node relative position relation between elbow node, shoulder node and head zone respectively, the posture of the first human body is determined, wherein head zone is the region for being used to characterize the first human body head determined according to multiple nodes of head and neck.
Description
Technical field
The present invention relates to image processing field more particularly to a kind of limbs posture judgment method and devices.
Background technology
There are mainly two types of the methods for being presently used for the judgement of limbs posture, and one is optical flow analysis method, another kind is depth map
As analytic approach.Wherein, optical flow analysis method by capture picture in light stream variation, to determine the movable information of target, Jin Ergen
Judge limbs posture according to the motion conditions of target；Range image analysis method using the distance of each point in scene to image acquisition device as
The pixel value of each point in depth image, and then limbs posture is judged according to depth image.
However, when implementation optical flow analysis method progress limbs posture judges, human body raises one's hand and stands up, and both act generation
Light stream vector it is more similar, thus can not accurately distinguish human body based on optical flow analysis method and raise one's hand or stand up.Implementing deeply
When degree image analytical method progress limbs posture judges, limbs (such as arm) are usually relatively thin, and the limbs information in depth image is unknown
It is aobvious；Also, when arm is close to head, the depth difference unobvious on arm and head can not judge depth difference information, lead
Limbs posture is caused to judge failure.
As it can be seen that lacking a kind of technology that can accurately and effectively carry out limbs posture judgement in the prior art.
Invention content
A kind of limbs posture judgment method of offer of the embodiment of the present invention and device, to realize accurate and effective limbs appearance
State judges.
In a first aspect, a kind of limbs posture judgment method is provided, including：
Determine that multiple human body nodes of the first human body in human body image, the multiple human body node include at least wrist section
Multiple nodes on point, elbow node, shoulder node and head and neck；
According to the wrist node opposite position between the elbow node, the shoulder node and head zone respectively
Relationship is set, determines the posture of first human body, wherein the head zone is multiple nodes according to the head and neck
The determining region for characterizing first human body head.
In one possible implementation, according to the wrist node respectively with the elbow node, the shoulder section
Relative position relation between point and head zone, determines the posture of first human body, including：It is determined in condition according to following
One or more determine conditions, determine whether the posture of first human body is preset posture：
(1) determine whether the wrist node is higher than or is less than within the scope of preset height difference the shoulder node；
(2) determine whether the wrist node is higher than the elbow node；
(3) determine whether the wrist node is located at the head zone.
In one possible implementation, according to the wrist node respectively with the elbow node, the shoulder section
Relative position relation between point and head zone, determines the posture of first human body, including：
Step 1：Determine the first-hand carpopodium point in the wrist node whether higher than first-hand in the elbow node
Otherwise toggle point determines that the posture of first human body is different from the preset posture if so, thening follow the steps 2；Wherein, institute
It is any wrist node in the wrist node to state first-hand carpopodium point, and the first-hand toggle point is and first wrist
Node is located at the wrist node of the first human body homonymy；
Step 2：Determine whether the first-hand carpopodium point is higher than or is less than within the scope of preset height difference the shoulder section
Otherwise it is described default to determine that the posture of first human body is different from if so, thening follow the steps 3 for the first shoulder node in point
Posture；Wherein, the first shoulder node is the shoulder node for being located at the first human body homonymy with the first-hand carpopodium point；
Step 3：Determine whether the first-hand carpopodium point is located at the head zone, if so, determination is described the first
The posture of body is different from the preset posture, otherwise determines that the posture of first human body is the preset posture.
In one possible implementation, according to the wrist node respectively with the elbow node, the shoulder section
Relative position relation between point and head zone, determines the posture of first human body, including：
Step 1：Determine the first-hand carpopodium point in the wrist node whether be higher than or within the scope of preset height difference it is low
Otherwise the first shoulder node in the shoulder node determines the posture of first human body not if so, thening follow the steps 2
It is same as the preset posture；Wherein, the first-hand carpopodium point be the wrist node in any wrist node, described first
Shoulder node is the shoulder node for being located at the first human body homonymy with the first-hand carpopodium point；
Step 2：The first-hand carpopodium point is determined whether higher than the first-hand toggle point in the elbow node, if so,
3 are thened follow the steps, otherwise determines that the posture of first human body is different from the preset posture；Wherein, the first-hand toggle
Point is the wrist node for being located at the first human body homonymy with the first-hand carpopodium point；
Step 3：Determine whether the first-hand carpopodium point is located at the head zone, if so, determination is described the first
The posture of body is different from the preset posture, otherwise determines that the posture of first human body is the preset posture.
In one possible implementation, according to the wrist node respectively with the elbow node, the shoulder
Relative position relation between node and head zone before the posture for determining first human body, further includes：According to described more
Personal body node establishes the position coordinates system for determining the multiple human body node location, or, by the figure of the human body image
As pixel coordinate system is as the position coordinates system；
According to the wrist node opposite position between the elbow node, the shoulder node and head zone respectively
Relationship is set, determines the posture of first human body, including：According to the wrist node, the elbow node, the shoulder node
With coordinate position of the multiple nodes of the head and neck in the position coordinates system, the relative position relation is determined,
And the posture of first human body is determined according to the relative position relation.
Second aspect provides a kind of limbs posture judgment means, including：
Node determining module, multiple human body nodes for determining the first human body in human body image, the multiple human body
Node includes at least multiple nodes on wrist node, elbow node, shoulder node and head and neck；
Posture determining module, for according to the wrist node respectively with the elbow node, the shoulder node and head
Relative position relation between portion region determines the posture of first human body, wherein the head zone is according to the head
The region for characterizing first human body head that multiple nodes of portion and neck determine.
In one possible implementation, the posture determining module is used for：According to following one determined in condition
Or multiple determining conditions, determine whether the posture of first human body is preset posture：
(1) determine whether the wrist node is higher than or is less than within the scope of preset height difference the shoulder node；
(2) determine whether the wrist node is higher than the elbow node；
(3) determine whether the wrist node is located at the head zone.
In one possible implementation, the posture determining module is specifically used for executing following steps：
Step 1：Determine the first-hand carpopodium point in the wrist node whether higher than first-hand in the elbow node
Otherwise toggle point determines that the posture of first human body is different from the preset posture if so, thening follow the steps 2；Wherein, institute
It is any wrist node in the wrist node to state first-hand carpopodium point, and the first-hand toggle point is and first wrist
Node is located at the wrist node of the first human body homonymy；
Step 2：Determine whether the first-hand carpopodium point is higher than or is less than within the scope of preset height difference the shoulder section
Otherwise it is described default to determine that the posture of first human body is different from if so, thening follow the steps 3 for the first shoulder node in point
Posture；Wherein, the first shoulder node is the shoulder node for being located at the first human body homonymy with the first-hand carpopodium point；
Step 3：Determine whether the first-hand carpopodium point is located at the head zone, if so, determination is described the first
The posture of body is different from the preset posture, otherwise determines that the posture of first human body is the preset posture.
In one possible implementation, the posture determining module is specifically used for executing following steps：
Step 1：Determine the first-hand carpopodium point in the wrist node whether be higher than or within the scope of preset height difference it is low
Otherwise the first shoulder node in the shoulder node determines the posture of first human body not if so, thening follow the steps 2
It is same as the preset posture；Wherein, the first-hand carpopodium point be the wrist node in any wrist node, described first
Shoulder node is the shoulder node for being located at the first human body homonymy with the first-hand carpopodium point；
Step 2：The first-hand carpopodium point is determined whether higher than the first-hand toggle point in the elbow node, if so,
3 are thened follow the steps, otherwise determines that the posture of first human body is different from the preset posture；Wherein, the first-hand toggle
Point is the wrist node for being located at the first human body homonymy with the first-hand carpopodium point；
Step 3：Determine whether the first-hand carpopodium point is located at the head zone, if so, determination is described the first
The posture of body is different from the preset posture, otherwise determines that the posture of first human body is the preset posture.
In one possible implementation, described device further includes coordinate system determining module, is used for：According to the hand
The carpopodium point relative position relation between the elbow node, the shoulder node and head zone respectively, determines described
Before the posture of one human body, the position for determining the multiple human body node location is established according to the multiple human body node and is sat
Mark system, or, using the image pixel coordinates system of the human body image as the position coordinates system；
The posture determining module is specifically used for：According to the wrist node, the elbow node, the shoulder node and
Coordinate position of the multiple nodes of the head and neck in the position coordinates system, determines the relative position relation, and
The posture of first human body is determined according to the relative position relation.
The third aspect, provides a kind of computer equipment, and the equipment includes：
At least one processor, and
The memory being connect at least one processor；
Wherein, the memory is stored with the instruction that can be executed by least one processor, at least one place
The instruction that device is stored by executing the memory is managed, method as described in relation to the first aspect is executed.
Fourth aspect provides a kind of computer readable storage medium, including：
The computer-readable recording medium storage has computer instruction, when the computer instruction is run on computers
When so that computer executes the method described in first aspect.
In the embodiment of the present invention, multiple human body nodes of the first human body in human body image, multiple human body node are determined
Including at least multiple nodes on wrist node, elbow node, shoulder node and head and neck, in turn, according to wrist node point
Relative position relation not between elbow node, shoulder node and head zone, determines the posture of the first human body, wherein head
Portion region is the region for characterizing the first human body head determined according to multiple nodes of head and neck.
By determining multiple human body nodes of the first human body, and according between wrist node and other nodes, head zone
Relative position relation, the wrist node of human body can accurately be positioned, realize judgement to limbs posture.Further
Ground carries out posture judgement in the embodiment of the present invention after the nodal information for determining human body according to nodal information, is based on nodal information
Limbs and body trunk can be accurately distinguished, are avoided since limbs information unobvious cause to judge inaccurate problem.And
And limbs posture is judged according to the relative position between wrist node and other nodes, head zone, it avoids based on depth map
As the problem of in the case of judging limbs posture, limbs posture judges failure when arm is close to head.
Description of the drawings
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, 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
Inventive embodiments for those of ordinary skill in the art without creative efforts, can also be according to carrying
The attached drawing of confession obtains other attached drawings.
Fig. 1 is a kind of flow diagram of limbs posture judgment method in the embodiment of the present invention；
Fig. 2 is the schematic diagram of multiple human body nodes in the embodiment of the present invention；
Fig. 3 is a kind of schematic diagram of posture determination process in the embodiment of the present invention；
Fig. 4 is the schematic diagram of another posture determination process in the embodiment of the present invention；
Fig. 5 is a kind of schematic diagram of position coordinates system in the embodiment of the present invention；
Fig. 6 is the schematic diagram of another position coordinates system in the embodiment of the present invention；
Fig. 7 is a kind of schematic diagram of head zone in the embodiment of the present invention；
Fig. 8 is a kind of structure diagram of limbs posture judgment means in the embodiment of the present invention.
Specific implementation mode
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A 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
The every other embodiment obtained without creative efforts, shall fall within the protection scope of the present invention.
In addition, the terms "and/or", only a kind of incidence relation of description affiliated partner, indicates may exist
Three kinds of relationships, for example, A and/or B, can indicate：Individualism A exists simultaneously A and B, these three situations of individualism B.Separately
Outside, character "/" herein, in the case where not illustrating, it is a kind of relationship of "or" to typically represent forward-backward correlation object.
In addition, it is necessary to understand, in the description of the embodiment of the present invention, the vocabulary such as " first ", " second " are only used for distinguishing description
Purpose is not understood to indicate or imply relative importance, can not be interpreted as instruction or hint sequence.
Embodiment one
The embodiment of the present invention provides a kind of limbs posture judgment method, this method can be used for the limbs posture of human body into
Row judges, for example, can be used for judging whether human body raises one's hand.For concrete example, in classroom instruction scene, class can be shot
Hall picture, then through the embodiment of the present invention in limbs posture judgment method combine captured picture, judge to teach indoor
The state of raising one's hand of one or more students.
Method in embodiment in order to better understand the present invention, below in conjunction with Figure of description and specific implementation
The above described technique is demenstrated in detail for mode.
Fig. 1 is referred to, the flow of the limbs posture judgment method in the embodiment of the present invention is described as follows.
Step 11：Determine that multiple human body nodes of the first human body in human body image, multiple human body nodes include at least hand
Multiple nodes on carpopodium point, elbow node, shoulder node and head and neck.
In the embodiment of the present invention, by carrying out multiple people that the first human body in human body image is determined in analysis to human body image
Body node, analysis method of determination can be image characteristic analysis, deep learning Model Identification, can also be determining in the prior art
The mode of human body node, for example, existing OpenPose algorithms, clique percolation method (Clique may be used
Percolation Method, CPM) determine multiple human body nodes of the first human body in human body image.It is right in the embodiment of the present invention
The human body node for specifically which kind of mode being used to analyze determining human body image is not limited.
Wherein, human body node can be the critical positions of the human bodies such as the artis of human body, face.For ease of understanding, with such as
For human body shown in Fig. 2, multiple human body nodes in the first human body may include having left finesse node 201, right finesse node
202, left hand toggle point 203, right hand toggle point 204, left shoulder node 205, right shoulder node 206, neck node 207, nose
One or more of node 208, left eye node 209, right eye node 210, left ear node 211 and auris dextra node 212.These people
Body node is used to represent the human region of near nodal, such as left finesse node 201 is used to represent the entire left finesse area of human body
Domain, including palm, finger, rather than it is merely representative of 201 this point in Fig. 2.
In specific implementation process, in human body image may include there are one or multiple human bodies, in the embodiment of the present invention
First human body can be any human body therein.Also, when in human body image including multiple human bodies, for therein any
Human body, the limbs posture judgment method that may be used in the embodiment of the present invention carry out limbs posture judgement respectively, that is to say, that
The limbs posture judgement of each human body is mutually independent of each other.
Step 12：According to wrist node, the relative position between elbow node, shoulder node and head zone is closed respectively
System, determines the posture of the first human body, wherein head zone be according to multiple nodes of head and neck determines for characterization the
The region of one human body head.
It, can be according to wrist node and elbow after determining multiple human body nodes of the first human body in the embodiment of the present invention
The relative position relation of node, the relative position relation and wrist node of wrist node and shoulder node and head zone
Relative position relation determines the posture of the first human body.
Wherein, head zone is the region determined according to multiple nodes of head and neck, for example, can according to fig. 2 in
Neck node 207, nose node 208, left eye node 209, right eye node 210, left ear node 211 and auris dextra node 212 determine
The head zone of human body in Fig. 2.The shape for the head zone determined for example can be rectangle.
In specific implementation process, according to wrist node respectively between elbow node, shoulder node and head zone
Relative position relation, determine the process of the posture of the first human body, may include one or more in following three judgment step
It is a：
(1) determine whether wrist node is higher than or is less than within the scope of preset height difference shoulder node；
(2) determine whether wrist node is higher than elbow node；
(3) determine whether first-hand carpopodium point is located at head zone.
The execution sequence of these three above-mentioned judgment steps can be exchanged mutually.Also, determine the posture of the first human body
The deterministic process that process can be " from thick to thin ", that is to say, that these three judgment steps can be carried out one by one, such as with default
Sequence carries out these three judgment steps one by one, also, when only first judgment step meets preset condition just carry out it is posterior
Judgment step.
Refer to Fig. 3, Fig. 4, below to according to wrist node respectively between elbow node, shoulder node and head zone
Relative position relation, determine that the process of the posture of the first human body is illustrated：
Determination process shown in Fig. 3, includes the following steps：
Step 120：Determine whether the first-hand carpopodium point in wrist node is higher than or is less than within the scope of preset height difference
The first shoulder node in shoulder node, if so, 121 are thened follow the steps, the no posture for thening follow the steps 124 the first human bodies of determination
Different from preset posture.
Step 121：First-hand carpopodium point is determined whether higher than the first-hand toggle point in elbow node, if so, executing
Step 122, the no posture for thening follow the steps 124 the first human bodies of determination is different from preset posture.
Step 122：Determine whether first-hand carpopodium point is located at head zone, if so, it is the first to then follow the steps 124 determinations
The posture of body is different from preset posture, and the no posture for thening follow the steps 123 the first human bodies of determination is preset posture.
Wherein, first-hand carpopodium point is any wrist node in wrist node, and the first shoulder node is and the first wrist
Node is located at the shoulder node of the first human body homonymy, and first-hand toggle point is to be located at the first body homonymy with first-hand carpopodium point
Wrist node.For example, first-hand carpopodium point can be the left finesse node 201 in Fig. 2, corresponding, the first shoulder node is Fig. 2
In left shoulder node 205, first-hand toggle point be Fig. 2 in left hand toggle point 203.Also, in a kind of possible embodiment party
In formula, determine that the posture of the first human body can refer to determining the state of raising one's hand of the first human body, corresponding, posture is that preset posture can
It is to raise one's hand to refer to the state of raising one's hand, it is not raise one's hand that posture, which can refer to the state of raising one's hand different from preset posture,.
Determination process shown in Fig. 4, includes the following steps：
Step 125：Determine whether the first-hand carpopodium point in wrist node is higher than or is less than within the scope of preset height difference
The first shoulder node in shoulder node, if so, 126 are thened follow the steps, the no posture for thening follow the steps 129 the first human bodies of determination
Different from preset posture.
Step 126：First-hand carpopodium point is determined whether higher than the first-hand toggle point in elbow node, if so, executing
Step 127, the no posture for thening follow the steps 129 the first human bodies of determination is different from preset posture.
Step 127：Determine whether first-hand carpopodium point is located at head zone, if so, it is the first to then follow the steps 129 determinations
The posture of body is different from preset posture, and the no posture for thening follow the steps 128 the first human bodies of determination is preset posture.
Wherein, first-hand carpopodium point is any wrist node in wrist node, and the first shoulder node is and the first wrist
Node is located at the shoulder node of the first human body homonymy, and first-hand toggle point is to be located at the first body homonymy with first-hand carpopodium point
Wrist node.For example, first-hand carpopodium point can be the left finesse node 201 in Fig. 2, corresponding, the first shoulder node is Fig. 2
In left shoulder node 205, first-hand toggle point be Fig. 2 in left hand toggle point 203.Also, in a kind of possible embodiment party
In formula, determine that the posture of the first human body can refer to determining the state of raising one's hand of the first human body, corresponding, posture is that preset posture can
It is to raise one's hand to refer to the state of raising one's hand, it is not raise one's hand that posture, which can refer to the state of raising one's hand different from preset posture,.
In a kind of possible embodiment, according to wrist node respectively with elbow node, shoulder node and header area
Relative position relation between domain further includes the steps that determining position coordinates system, the position before the posture for determining the first human body
Coordinate system is for determining multiple human body node locations.In the position under coordinate system, it may be determined that the human body node of the first human body, area
Relative position relation between domain.
There are many modes for determining position coordinates system, is illustrated below with two of which：
The first method of determination：Using the image pixel coordinates system of human body image as position coordinates system
Wherein, image pixel coordinates system can refer in image processing field generally use, be original with the image upper left corner
Point O, image top be X-axis, image the left side be Y-axis establish coordinate system.As shown in figure 5, for people in the embodiment of the present invention
The schematic diagram of the image pixel coordinates system of body image.
Second of method of determination：Position coordinates system is established according to multiple human body nodes
The specific implementation mode that position coordinates system is established according to multiple human body nodes can determine the more of the first human body
On the basis of personal body node, according to the coordinate system of the position distribution situation of multiple human body nodes foundation.That is, according to more
The position coordinates system that personal body node is established is related to multiple positions of human body node, not by the other factors shadow of human body image
It rings.In this kind of method of determination, when, there are when multiple human bodies, the corresponding position coordinates system established of each human body can in human body image
To be different.As shown in fig. 6, when establishing position coordinates system according to multiple human body nodes, built corresponding to the human body 61 in Fig. 6
Vertical position coordinates system 62 is different from the corresponding position coordinates system 64 established of human body 63.
Based on the first method of determination, the entire operand judged during limbs posture can be reduced, limbs appearance is improved
The efficiency that state judges, also, each human body in human body image can share a position coordinates system.Based on second of determination side
Formula can reduce due to the influence of human body inclination or shooting angle inclination to judging result in human body image, improve limbs posture
The accuracy of judgement.
Further, in the case where position coordinates system is determined, according to wrist node respectively with elbow node, shoulder section
Relative position relation between point and head zone, determines the detailed process of the posture of the first human body, can be according to wrist section
Coordinate position of the multiple nodes of point, elbow node, shoulder node and head and neck in the coordinate system of position determines opposite position
Relationship is set, and determines the posture of the first human body according to relative position relation.I.e., it is possible to which the coordinate by comparing each one body node is joined
Several magnitude relationships determines relative position relation, and the posture of the first human body is determined according to relative position relation.
By taking position coordinates system shown in fig. 7 as an example, it is assumed that groups of people's body node of the human body in Fig. 7 is in position coordinate system
In coordinate position it is as follows：
Neck node 207 (NECK_X, NECK_Y), nose node 208 (NOSE_X, NOSE_Y), left eye node 209
(LEYE_X, LEYE_Y), right eye node 210 (REYE_X, REYE_Y), left ear node 211 (LEAR_X, LEAR_Y), auris dextra section
212 (REAR_X, REAR_Y) of point.Then, during head zone in determining Fig. 7 73：
A. coordinate LEFT_X of the upper left corner (putting 71) of head zone 73 in X-axis can be true with following formula (1)
Fixed, alpha therein is preset value, such as takes 0.1：
LEFT_X=REAR_X-alpha*ABS (LEYE_X-REYE_X) ... ... (1)
B. coordinate TOP_Y of the upper left corner (putting 71) of head zone 73 in Y-axis can be true with following formula (2)
It is fixed：
TOP_Y=NOSE_Y-(NECK_Y-NOSE_Y) ... ... (2)
C. coordinate RIGHT_X of the lower right corner (putting 72) of head zone 73 in X-axis can be true with following formula (3)
Fixed, alpha therein is preset value, such as takes 0.1：
RIGHT_X=LEAR_X+alpha*ABS (LEYE_X-REYE_X) ... ... (3)
D. coordinate BOTTOM_Y of the lower right corner (putting 72) of head zone 73 in Y-axis can use following formula (4)
It determines：
BOTTOM_Y=NOSE_Y+ (NECK_Y-NOSE_Y) ... ... (4)
Further, during determining whether first-hand carpopodium point is located at head zone 73, if first-hand carpopodium
The coordinate of point is (WRIST_X, WRIST_Y), then following formula (5) may be used and determine whether first-hand carpopodium point is located at head
Portion region 73.Specifically, if (WRIST_X, WRIST_Y) meets following formula (5), it is determined that first-hand carpopodium point not position
In head zone 73, otherwise determine that first-hand carpopodium point is located at head zone 73.
WRISTX>LEFTX&&WRISTX<RIGHTX&&WRISTY<BOTTOM_Y&&
WRISTY>TOP_Y…………(5)
Embodiment two
Based on same inventive concept, the embodiment of the present invention provides a kind of limbs posture judgment means.Fig. 8 is referred to, the limb
Body posture judgment means include at least node determining module 81 and posture determining module 82.Wherein：
Node determining module 81, multiple human body nodes for determining the first human body in human body image, multiple human body sections
Multiple nodes of the point including at least wrist node, elbow node, shoulder node and head and neck；
Posture determining module 82, for according to wrist node respectively between elbow node, shoulder node and head zone
Relative position relation, determine the posture of the first human body, wherein head zone be according to multiple nodes of head and neck determine
For characterizing the region of the first human body head.
In a kind of possible embodiment, posture determining module 82 is used for：According to it is following determine condition in one or
Multiple determining conditions determine whether the posture of the first human body is preset posture：
(1) determine whether wrist node is higher than or is less than within the scope of preset height difference shoulder node；
(2) determine whether wrist node is higher than elbow node；
(3) determine whether wrist node is located at head zone.
In a kind of possible embodiment, posture determining module 82 is specifically used for executing following steps：
Step 1：The first-hand carpopodium point in wrist node is determined whether higher than the first-hand toggle point in elbow node, if
It is to then follow the steps 2, otherwise determines that the posture of the first human body is different from preset posture；Wherein, first-hand carpopodium point is wrist section
Any wrist node in point, first-hand toggle point are to be located at the wrist node of the first body homonymy with first-hand carpopodium point；
Step 2：Determine whether first-hand carpopodium point is higher than or within the scope of preset height difference less than the in shoulder node
Otherwise one shoulder node determines that the posture of the first human body is different from preset posture if so, thening follow the steps 3；Wherein, the first shoulder
Portion's node is the shoulder node for being located at the first body homonymy with first-hand carpopodium point；
Step 3：Determine whether first-hand carpopodium point is located at head zone, if so, determining that the posture of the first human body is different
In preset posture, otherwise determine that the posture of the first human body is preset posture.
In a kind of possible embodiment, posture determining module 82 is specifically used for executing following steps：
Step 1：Determine whether the first-hand carpopodium point in wrist node is higher than or is less than within the scope of preset height difference shoulder
Otherwise the first shoulder node in portion's node determines that the posture of the first human body is different from preset posture if so, thening follow the steps 2；
Wherein, first-hand carpopodium point is any wrist node in wrist node, and the first shoulder node is to be located at first-hand carpopodium point
The shoulder node of first human body homonymy；
Step 2：First-hand carpopodium point is determined whether higher than the first-hand toggle point in elbow node, if so, executing step
Rapid 3, otherwise determine that the posture of the first human body is different from preset posture；Wherein, first-hand toggle point is and first-hand carpopodium point
In the wrist node of the first human body homonymy；
Step 3：Determine whether first-hand carpopodium point is located at head zone, if so, determining that the posture of the first human body is different
In preset posture, otherwise determine that the posture of the first human body is preset posture.
In a kind of possible embodiment, device further includes coordinate system determining module, is used for：According to wrist node point
Relative position relation not between elbow node, shoulder node and head zone, before the posture for determining the first human body, according to
Multiple human body nodes establish the position coordinates system for determining multiple human body node locations, or, by the image pixel of human body image
Coordinate system is as position coordinates system；
Posture determining module 82 is specifically used for：According to the more of wrist node, elbow node, shoulder node and head and neck
Coordinate position of a node in the coordinate system of position determines relative position relation, and is determined according to relative position relation the first
The posture of body.
Embodiment three
Based on same inventive concept, the embodiment of the present invention provides a kind of computer equipment, including：
At least one processor, and
The memory being connect at least one processor；
Wherein, memory is stored with the instruction that can be executed by least one processor, and at least one processor passes through execution
The instruction of memory storage, executes the method as described in embodiment one.
Example IV
Based on same inventive concept, the embodiment of the present invention provides a kind of computer readable storage medium, this is computer-readable
Storage medium is stored with computer instruction, when computer instruction is run on computers so that computer executes embodiment one
The method.
In specific implementation process, computer readable storage medium includes：General serial bus USB
(Universal Serial Bus flash drive, USB), mobile hard disk, read-only memory (Read-Only Memory,
ROM), random access memory (Random Access Memory, RAM), magnetic disc or CD etc. are various can store program
The storage medium of code.
Above-mentioned one or more technical solutions, at least have the advantages that：
In the embodiment of the present invention, multiple human body nodes of the first human body in human body image, multiple human body node are determined
Including at least multiple nodes on wrist node, elbow node, shoulder node and head and neck, in turn, according to wrist node point
Relative position relation not between elbow node, shoulder node and head zone, determines the posture of the first human body, wherein head
Portion region is the region for characterizing the first human body head determined according to multiple nodes of head and neck.
By determining multiple human body nodes of the first human body, and according between wrist node and other nodes, head zone
Relative position relation, the wrist node of human body can accurately be positioned, realize judgement to limbs posture.Further
Ground carries out posture judgement in the embodiment of the present invention after the nodal information for determining human body according to nodal information, is based on nodal information
Limbs and body trunk can be accurately distinguished, are avoided since limbs information unobvious cause to judge inaccurate problem.And
And limbs posture is judged according to the relative position between wrist node and other nodes, head zone, it avoids based on depth map
As the problem of in the case of judging limbs posture, limbs posture judges failure when arm is close to head.
Further, during determining the posture of the first human body, using the method for determination of " from thick to thin ", to wrist
Node is gradually judged with elbow node, shoulder node, the relative position relation of head zone respectively.Thus, on the one hand, it is logical
It crosses and repeatedly judges on the other hand the accuracy for improving limbs posture judging result judged due to being not necessarily to each limbs posture
Journey has been carried out all judgment steps, thus reduces the calculation amount of limbs posture deterministic process, improves execution efficiency.
Further, during limbs posture judges, it is set with the judgement of opponent's carpopodium point and head zone, is determining first
It when wrist node is located at head zone, determines that the posture of the first human body is different from preset posture, and is determining first-hand carpopodium
When point is not located at head zone, determine that the posture of the first human body is preset posture.Pass through the phase of opponent's carpopodium point and head node,
Judgement to position relationship can avoid scratching one's head, wipe the postures such as face and be mistaken for preset posture, will such as scratch one's head, and wipe face and be mistaken for
It raises one's hand.
The apparatus embodiments described above are merely exemplary, wherein the units/modules illustrated as separating component
It may or may not be physically separated, the component shown as units/modules may or may not be
Physical unit/module, you can be located at a place, or may be distributed in multiple network element/modules.It can basis
It is actual to need that some or all of module therein is selected to achieve the purpose of the solution of this embodiment.Ordinary skill people
Member is not in the case where paying performing creative labour, you can to understand and implement.
Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can
It is realized by the mode of software plus required general hardware platform, naturally it is also possible to pass through hardware.Based on this understanding, on
Stating technical solution, substantially the part that contributes to existing technology can be expressed in the form of software products in other words, should
Computer software product can store in a computer-readable storage medium, such as ROM/RAM, magnetic disc, CD, including several fingers
It enables and using so that a computer equipment (can be personal computer, server or the network equipment etc.) executes each implementation
Method described in certain parts of example or embodiment.
Finally it should be noted that：The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although
Present invention has been described in detail with reference to the aforementioned embodiments, it will be understood by those of ordinary skill in the art that：It still may be used
With technical scheme described in the above embodiments is modified or equivalent replacement of some of the technical features；
And these modifications or replacements, various embodiments of the present invention technical solution that it does not separate the essence of the corresponding technical solution spirit and
Range.
Claims (12)
1. a kind of limbs posture judgment method, which is characterized in that including：
Determine multiple human body nodes of the first human body in human body image, the multiple human body node include at least wrist node,
Multiple nodes on elbow node, shoulder node and head and neck；
According to the wrist node, the relative position between the elbow node, the shoulder node and head zone is closed respectively
System, determines the posture of first human body, wherein the head zone is to be determined according to multiple nodes of the head and neck
For characterizing the region of first human body head.
2. the method as described in claim 1, which is characterized in that according to the wrist node respectively with the elbow node, institute
The relative position relation between shoulder node and head zone is stated, determines the posture of first human body, including：According to following true
One or more of fixed condition determines condition, determines whether the posture of first human body is preset posture：
(1) determine whether the wrist node is higher than or is less than within the scope of preset height difference the shoulder node；
(2) determine whether the wrist node is higher than the elbow node；
(3) determine whether the wrist node is located at the head zone.
3. method as claimed in claim 2, which is characterized in that according to the wrist node respectively with the elbow node, institute
The relative position relation between shoulder node and head zone is stated, determines the posture of first human body, including：
Step 1：Determine the first-hand carpopodium point in the wrist node whether higher than the first-hand toggle in the elbow node
Otherwise point determines that the posture of first human body is different from the preset posture if so, thening follow the steps 2；Wherein, described
Proficiency carpopodium point is any wrist node in the wrist node, and the first-hand toggle point is and the first-hand carpopodium point
Positioned at the wrist node of the first human body homonymy；
Step 2：Determine whether the first-hand carpopodium point is higher than or within the scope of preset height difference less than in the shoulder node
The first shoulder node otherwise determine that the posture of first human body is different from the preset posture if so, then follow the steps 3；
Wherein, the first shoulder node is the shoulder node for being located at the first human body homonymy with the first-hand carpopodium point；
Step 3：Determine whether the first-hand carpopodium point is located at the head zone, if so, determining first human body
Posture is different from the preset posture, otherwise determines that the posture of first human body is the preset posture.
4. method as claimed in claim 2, which is characterized in that according to the wrist node respectively with the elbow node, institute
The relative position relation between shoulder node and head zone is stated, determines the posture of first human body, including：
Step 1：Determine whether the first-hand carpopodium point in the wrist node is higher than or is less than within the scope of preset height difference institute
The first shoulder node in shoulder node is stated, if so, thening follow the steps 2, otherwise determines that the posture of first human body is different from
The preset posture；Wherein, the first-hand carpopodium point is any wrist node in the wrist node, first shoulder
Node is the shoulder node for being located at the first human body homonymy with the first-hand carpopodium point；
Step 2：The first-hand carpopodium point is determined whether higher than the first-hand toggle point in the elbow node, if so, holding
Otherwise row step 3 determines that the posture of first human body is different from the preset posture；Wherein, the first-hand toggle point is
It is located at the wrist node of the first human body homonymy with the first-hand carpopodium point；
Step 3：Determine whether the first-hand carpopodium point is located at the head zone, if so, determining first human body
Posture is different from the preset posture, otherwise determines that the posture of first human body is the preset posture.
5. the method as described in any one of claim 1-4, which is characterized in that according to the wrist node respectively with it is described
Relative position relation between elbow node, the shoulder node and head zone, before the posture for determining first human body,
Further include：The position coordinates system for determining the multiple human body node location is established according to the multiple human body node, or, will
The image pixel coordinates system of the human body image is as the position coordinates system；
According to the wrist node, the relative position between the elbow node, the shoulder node and head zone is closed respectively
System, determines the posture of first human body, including：According to the wrist node, the elbow node, the shoulder node and institute
Coordinate position of the multiple nodes of head and neck in the position coordinates system is stated, determines the relative position relation, and root
The posture of first human body is determined according to the relative position relation.
6. a kind of limbs posture judgment means, which is characterized in that including：
Node determining module, multiple human body nodes for determining the first human body in human body image, the multiple human body node
Including at least multiple nodes on wrist node, elbow node, shoulder node and head and neck；
Posture determining module, for according to the wrist node respectively with the elbow node, the shoulder node and header area
Relative position relation between domain determines the posture of first human body, wherein the head zone be according to the head with
The region for characterizing first human body head that multiple nodes of neck determine.
7. device as claimed in claim 6, which is characterized in that the posture determining module is used for：Condition is determined according to following
One or more of determine condition, determine whether the posture of first human body is preset posture：
(1) determine whether the wrist node is higher than or is less than within the scope of preset height difference the shoulder node；
(2) determine whether the wrist node is higher than the elbow node；
(3) determine whether the wrist node is located at the head zone.
8. device as claimed in claim 7, which is characterized in that the posture determining module is specifically used for executing following steps：
Step 1：Determine the first-hand carpopodium point in the wrist node whether higher than the first-hand toggle in the elbow node
Otherwise point determines that the posture of first human body is different from the preset posture if so, thening follow the steps 2；Wherein, described
Proficiency carpopodium point is any wrist node in the wrist node, and the first-hand toggle point is and the first-hand carpopodium point
Positioned at the wrist node of the first human body homonymy；
Step 2：Determine whether the first-hand carpopodium point is higher than or within the scope of preset height difference less than in the shoulder node
The first shoulder node otherwise determine that the posture of first human body is different from the preset posture if so, then follow the steps 3；
Wherein, the first shoulder node is the shoulder node for being located at the first human body homonymy with the first-hand carpopodium point；
Step 3：Determine whether the first-hand carpopodium point is located at the head zone, if so, determining first human body
Posture is different from the preset posture, otherwise determines that the posture of first human body is the preset posture.
9. device as claimed in claim 7, which is characterized in that the posture determining module is specifically used for executing following steps：
Step 1：Determine whether the first-hand carpopodium point in the wrist node is higher than or is less than within the scope of preset height difference institute
The first shoulder node in shoulder node is stated, if so, thening follow the steps 2, otherwise determines that the posture of first human body is different from
The preset posture；Wherein, the first-hand carpopodium point is any wrist node in the wrist node, first shoulder
Node is the shoulder node for being located at the first human body homonymy with the first-hand carpopodium point；
Step 2：The first-hand carpopodium point is determined whether higher than the first-hand toggle point in the elbow node, if so, holding
Otherwise row step 3 determines that the posture of first human body is different from the preset posture；Wherein, the first-hand toggle point is
It is located at the wrist node of the first human body homonymy with the first-hand carpopodium point；
Step 3：Determine whether the first-hand carpopodium point is located at the head zone, if so, determining first human body
Posture is different from the preset posture, otherwise determines that the posture of first human body is the preset posture.
10. the device as described in any one of claim 6-9, which is characterized in that described device further includes that coordinate system determines mould
Block is used for：Opposite between the elbow node, the shoulder node and head zone respectively according to the wrist node
Position relationship before the posture for determining first human body, is established the multiple for determining according to the multiple human body node
The position coordinates system of human body node location, or, using the image pixel coordinates system of the human body image as the position coordinates system；
The posture determining module is specifically used for：According to the wrist node, the elbow node, the shoulder node and described
Coordinate position of the multiple nodes of head and neck in the position coordinates system, determines the relative position relation, and according to
The relative position relation determines the posture of first human body.
11. a kind of computer equipment, which is characterized in that including：
At least one processor, and
The memory being connect at least one processor；
Wherein, the memory is stored with the instruction that can be executed by least one processor, at least one processor
By executing the instruction of the memory storage, the method as described in any one of claim 1-5 is executed.
12. a kind of computer readable storage medium, it is characterised in that：
The computer-readable recording medium storage has computer instruction, when the computer instruction is run on computers,
So that computer executes the method as described in any one of claim 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810229630.4A CN108573216A (en) | 2018-03-20 | 2018-03-20 | A kind of limbs posture judgment method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810229630.4A CN108573216A (en) | 2018-03-20 | 2018-03-20 | A kind of limbs posture judgment method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108573216A true CN108573216A (en) | 2018-09-25 |
Family
ID=63574479
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810229630.4A Pending CN108573216A (en) | 2018-03-20 | 2018-03-20 | A kind of limbs posture judgment method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108573216A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109470263A (en) * | 2018-09-30 | 2019-03-15 | 北京诺亦腾科技有限公司 | Motion capture method, electronic equipment and computer storage medium |
| CN111242004A (en) * | 2020-01-10 | 2020-06-05 | 厦门瑞为信息技术有限公司 | Automatic alarm method and system based on elevator monitoring data processing |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104157107A (en) * | 2014-07-24 | 2014-11-19 | 燕山大学 | Human body posture correction device based on Kinect sensor |
| CN105224090A (en) * | 2010-12-27 | 2016-01-06 | 日立麦克赛尔株式会社 | Image display device |
| CN105608467A (en) * | 2015-12-16 | 2016-05-25 | 西北工业大学 | Kinect-based non-contact type student physical fitness evaluation method |
| CN107392086A (en) * | 2017-05-26 | 2017-11-24 | 深圳奥比中光科技有限公司 | Apparatus for evaluating, system and the storage device of human body attitude |
| CN107609517A (en) * | 2017-09-15 | 2018-01-19 | 华中科技大学 | A kind of classroom behavior detecting system based on computer vision |
-
2018
- 2018-03-20 CN CN201810229630.4A patent/CN108573216A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105224090A (en) * | 2010-12-27 | 2016-01-06 | 日立麦克赛尔株式会社 | Image display device |
| CN104157107A (en) * | 2014-07-24 | 2014-11-19 | 燕山大学 | Human body posture correction device based on Kinect sensor |
| CN105608467A (en) * | 2015-12-16 | 2016-05-25 | 西北工业大学 | Kinect-based non-contact type student physical fitness evaluation method |
| CN107392086A (en) * | 2017-05-26 | 2017-11-24 | 深圳奥比中光科技有限公司 | Apparatus for evaluating, system and the storage device of human body attitude |
| CN107609517A (en) * | 2017-09-15 | 2018-01-19 | 华中科技大学 | A kind of classroom behavior detecting system based on computer vision |
Non-Patent Citations (2)
| Title |
|---|
| 张丹等: ""基于Kinect骨骼预定义的体态识别算法"", 《计算机应用》 * |
| 张鸿宇: ""课堂学习行为测量系统的设计与实现"", 《中国优秀硕士学位论文全文数据库·社会科学Ⅱ辑》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109470263A (en) * | 2018-09-30 | 2019-03-15 | 北京诺亦腾科技有限公司 | Motion capture method, electronic equipment and computer storage medium |
| CN109470263B (en) * | 2018-09-30 | 2020-03-20 | 北京诺亦腾科技有限公司 | Motion capture method, electronic device, and computer storage medium |
| CN111242004A (en) * | 2020-01-10 | 2020-06-05 | 厦门瑞为信息技术有限公司 | Automatic alarm method and system based on elevator monitoring data processing |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105740780B (en) | Method and device for detecting living human face | |
| JP4149213B2 (en) | Pointed position detection device and autonomous robot | |
| CN109859305B (en) | Three-dimensional face modeling and recognizing method and device based on multi-angle two-dimensional face | |
| CN109117893A (en) | A kind of action identification method and device based on human body attitude | |
| Oprisescu et al. | Automatic static hand gesture recognition using tof cameras | |
| US20050201594A1 (en) | Movement evaluation apparatus and method | |
| Maisto et al. | An accurate algorithm for the identification of fingertips using an RGB-D camera | |
| JP2012518236A (en) | Method and system for gesture recognition | |
| CN107767335A (en) | A kind of image interfusion method and system based on face recognition features' point location | |
| CN112070782B (en) | Method, device, computer readable medium and electronic equipment for identifying scene contour | |
| CN112487964B (en) | Gesture detection and recognition method, gesture detection and recognition equipment and computer-readable storage medium | |
| CN111488775B (en) | Device and method for judging degree of visibility | |
| JP2000251078A (en) | Method and device for estimating three-dimensional posture of person, and method and device for estimating position of elbow of person | |
| CN106778574A (en) | For the detection method and device of facial image | |
| Bei et al. | Sitting posture detection using adaptively fused 3D features | |
| CN108573216A (en) | A kind of limbs posture judgment method and device | |
| CN109558797B (en) | Method for distinguishing human body balance disorder based on gravity center area model under visual stimulation | |
| CN111784660B (en) | Method and system for analyzing frontal face degree of face image | |
| JP5111321B2 (en) | 瞼 Likelihood calculation device and program | |
| CN110598647B (en) | Head posture recognition method based on image recognition | |
| CN110826495A (en) | Body left and right limb consistency tracking and distinguishing method and system based on face orientation | |
| CN115019396A (en) | Learning state monitoring method, device, equipment and medium | |
| CN113470073A (en) | Animal center tracking method based on deep learning | |
| Liu et al. | Gesture recognition based on Kinect | |
| CN112183155B (en) | Method and device for establishing action posture library, generating action posture and identifying action posture |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180925 |