CN103099602B - Based on the physical examinations method and system of optical identification - Google Patents
Based on the physical examinations method and system of optical identification Download PDFInfo
- Publication number
- CN103099602B CN103099602B CN201110357613.7A CN201110357613A CN103099602B CN 103099602 B CN103099602 B CN 103099602B CN 201110357613 A CN201110357613 A CN 201110357613A CN 103099602 B CN103099602 B CN 103099602B
- Authority
- CN
- China
- Prior art keywords
- data
- physical examinations
- sporter
- image
- depth
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 47
- 230000003287 optical effect Effects 0.000 title claims abstract description 46
- 230000033001 locomotion Effects 0.000 claims abstract description 20
- 230000008569 process Effects 0.000 claims abstract description 16
- 238000012545 processing Methods 0.000 claims abstract description 10
- 238000012360 testing method Methods 0.000 claims description 15
- 230000003187 abdominal effect Effects 0.000 claims description 7
- 238000001514 detection method Methods 0.000 claims description 6
- 230000005021 gait Effects 0.000 claims description 4
- 230000003993 interaction Effects 0.000 claims description 3
- 230000002452 interceptive effect Effects 0.000 claims description 3
- 230000015572 biosynthetic process Effects 0.000 claims description 2
- 210000003128 head Anatomy 0.000 description 30
- 239000000284 extract Substances 0.000 description 13
- 238000012549 training Methods 0.000 description 11
- 238000003384 imaging method Methods 0.000 description 8
- 238000002372 labelling Methods 0.000 description 8
- 238000005259 measurement Methods 0.000 description 6
- 230000001427 coherent effect Effects 0.000 description 5
- 235000019557 luminance Nutrition 0.000 description 5
- 230000008859 change Effects 0.000 description 4
- 238000013480 data collection Methods 0.000 description 4
- 238000012546 transfer Methods 0.000 description 4
- 230000000007 visual effect Effects 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 210000003141 lower extremity Anatomy 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000036772 blood pressure Effects 0.000 description 1
- 230000037237 body shape Effects 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000000205 computational method Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 210000003414 extremity Anatomy 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000009191 jumping Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000024159 perception of rate of movement Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Landscapes
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The present invention relates to a kind of physical examinations method and system based on optical identification.Should comprise based on the physical examinations method of optical identification: obtain the depth image including sporter; Obtain sporter's body data; The characteristic point coordinate adapted with sports events during obtaining sporter's motion start-stop; Calculate and form exercise data.Optical recognition system should be comprised based on the physical examinations system of optical identification, adopt depth camera identification sporter motion, and the depth image data formed; Data handling system, is connected with optical recognition system, processes this depth image data, and forms exercise data according to the type of sports date processing depth image data received.Physical examinations method and system based on optical identification of the present invention carries out the speed of physical examinations.
Description
Technical field
The present invention relates to a kind of physical examinations method, particularly adopt the detection method of optical identification.
The present invention relates to a kind of physical examinations system, particularly adopt the detection system of optical identification.
Background technology
From 2000, the Chinese government for the cycle with every 5 years, carried out national physique and detects.2010,10 departments such as General Administration of Sport, the Ministry of Education, the Department of Science and Technology started third time national physique detected activity in China.Testing index comprises body shape, physical function and body constitution three aspect, and the object of detection is the child of 3 to 6 years old, the children and youth of 7 to 19 years old, the adult of 20 to 59 years old, the old people of 60 to 69 years old.
In existing physical examinations, also adopt traditional detection method.The such as patent No. is 200520053518.8, patent name is the Chinese patent of " a kind of physical examinations or leg flexing structure of body builder ", it mainly detects the mechanism of leg power, effectively ensure bent lower limb action criteria, make test or body-building action standard more, and can different human bodies be adapted to, do not allow user utilize bent leg mechanism to carry out borrowing power, ensure that the accuracy of test result and the effectiveness of body-building.Although test result accurately with effective, still adopt manual record to the sampling of data, there will be certain personal error equally.Standing long jump in such as physical examinations one, needs the distance measuring out ski-jump and pick-up point with a yardstick, and hand-kept; For abdominal crunch, need manually to determine that in the scheduled time, sporter plays the number of seat.Equally to the reading of data by human eye, record is also will through unusually.That is, in existing physical examinations equipment, need artificial participation to measure and record data, use inconvenience.In use, just there will be some artificial errors simultaneously, objective, working specification, the result target accurately of sampling cannot be reached.
For the problems referred to above, although there have been some patents, these patents can not solve the problem completely.Local at other, also likely occur that personnel to be measured act as fraudulent substitute for a person, score personnel change the phenomenons such as total marks of the examination privately, adopt traditional technological means to be all difficult to solve, also make that physical examinations result is reliable not, degree of belief is relatively low.So for the deficiencies in the prior art, urgent need will change.
Summary of the invention
The object of the present invention is to provide a kind of physical examinations method and system based on optical identification.The present invention solves human body constitution datamation measurement, record problem, can process the multiple projects comprising long-jump, step-on testing, abdominal crunch, sitting body anteflexion, high jump, rope skipping etc.
Technical scheme of the present invention is as follows:
Based on a physical examinations method for optical identification, comprise the depth image obtaining and include sporter; Obtain sporter's body data; The characteristic point coordinate adapted with sports events during obtaining sporter's motion start-stop; Calculate and form exercise data.
In another preferred implementation of physical examinations method of the present invention, this sporter's body data comprises at least one in height, brachium and foot length.
In another preferred implementation of physical examinations method of the present invention, this physical examinations method also comprises at least one in the physiological data and appearance data processing sporter.
In another preferred implementation of physical examinations method of the present invention, this physical examinations method also comprises encryption body data, exercise data, physiological data and appearance data.
Based on a physical examinations system for optical identification, comprise optical recognition system, adopt depth camera identification sporter motion, and the depth image data formed; Data handling system, is connected with optical recognition system, processes this depth image data, and forms exercise data according to the type of sports date processing depth image data received.
In another preferred implementation of physical examinations system of the present invention, these type of sports data are external setting-up.
In another preferred implementation of physical examinations system of the present invention, this physical examinations system also comprises movement recognition system, be connected with data handling system with this optical recognition system, give data handling system for depth image data identification type of sports Data Concurrent.
In another preferred implementation of physical examinations system of the present invention, this optical recognition system obtains in very first time point the first view data including this sporter, obtain the second view data including this sporter in the second time point, and determine to represent motion characteristics point coordinates according to image.
In another preferred implementation of physical examinations system of the present invention, this characteristic point coordinate comprises the characteristic point coordinate obtaining very first time point and the characteristic point coordinate obtaining the second time point.
In another preferred implementation of physical examinations system of the present invention, this characteristic point coordinate is space coordinates.
In another preferred implementation of physical examinations system of the present invention, this physical examinations system also comprises a receiving system, can receive the third party's data transmitted outside.
In another preferred implementation of physical examinations system of the present invention, this physical examinations system also comprises a Camcording system, to all physical examinations process video recordings.
In another preferred implementation of physical examinations system of the present invention, this physical examinations system also comprises an encryption system, is connected with this data handling system, is encrypted video recording, third party's data, depth image data and exercise data.
In another preferred implementation of physical examinations system of the present invention, this type of sports is at least one among long-jump, step-on testing, abdominal crunch, sitting body anteflexion, high jump, rope skipping, shuttlecock kicking.
In another preferred implementation of physical examinations system of the present invention, this physical examinations system also comprises user log-in block, and this user log-in block logs according to one or more authentication of users in the build image of user, face-image, password, sound and gait.
In another preferred implementation of physical examinations system of the present invention, this physical examinations system also comprises exercise data comparation and assessment module, this exercise data comparation and assessment module preset standard data, this exercise data comparation and assessment module contrasts according to this normal data and the exercise data collected, and exports comparing result.
In another preferred implementation of physical examinations system of the present invention, this physical examinations system also comprises exercise data and transmits receiver module, this exercise data transmits receiver module and is used for exercise data to be sent to server end, and from received server-side data processed result.
In another preferred implementation of physical examinations system of the present invention, this physical examinations system also comprises camera module, the face-image of this camera module record sporter.
In another preferred implementation of physical examinations system of the present invention, this physical examinations system also comprises a personal-machine interactive device, and this human-computer interaction interface comprises at least one in a display device, keyboard, mouse.
In another preferred implementation of physical examinations system of the present invention, this physical examinations system comprises a data outputting module, whether this data outputting module detects has external memory to be connected to this physical examinations system, having detected that data are transferred in this external memory according to preset format when being connected to this physical examinations system by external memory.
In another preferred implementation of physical examinations system of the present invention, this physical examinations system also comprises a loudspeaker module, and after this exercise data is disposed, this loudspeaker module plays this exercise data result.
In another preferred implementation of physical examinations system of the present invention, this optical recognition system is one or more in single camera visible images acquisition system, single camera infrared light image acquisition system and depth camera device.
In another preferred implementation of physical examinations system of the present invention, this single camera visible images acquisition system is for obtaining visible images data; This single camera infrared light image acquisition system is for obtaining infrared light image data; This visible images forms the first depth image data through data prediction, and this infrared light image forms the second depth image through data prediction; This depth camera device is for obtaining the 3rd depth image data; This data handling system according at least one in these the first depth image data of the exercise data type of process received, the second depth image data and the 3rd depth image data to form exercise data or body data.
In another preferred implementation of physical examinations system of the present invention, this data handling system is also according at least one in these visible images data and these infrared light image data, carry out date processing in conjunction with at least one in these the first depth image data, the second depth image data and the 3rd depth image data, form exercise data or body data.
Relative to prior art, a kind of physical examinations method and system based on optical identification provided by the present invention, and there is good real-time and extended capability.
Relative to prior art, because the physical examinations method and system based on optical identification of the present invention adopts optical instrument automatically to identify body data and the exercise data of sporter, realize unmanned measurement and score.Thus make that the speed of test job is promoted, time shorten, human cost reduce.
Meanwhile, because data processing method used and standard all preset, can strictly unification, thus various interference factors when avoiding manual measurement.Thus ensure the standard comparatively unification of sporter's physical examinations, more there is fairness.
Meanwhile, owing to substantially not needing the participation of people, also stop various potential doing wrong to serve one's friends or relatives.Make that physical examinations result is relatively reliable, degree of belief is relatively high.
Detailed description of the invention
Below preferred embodiment of the present invention is described in detail.
Basic thought of the present invention is, adopts depth camera to gather scene image, in scene image, identifies the human body relevant to project survey content, adopts the computational methods be associated with project to calculate human body constitution data and record.In order to prevent the problem such as personal error from occurring, physical examinations process whole process adopts automatization's record data also to data encryption, adopts shooting record simultaneously, retains the appearance data of tester, and to appearance data and physical examinations Data Matching and encryption.
A kind of preferred implementation that the present invention is based on the physical examinations method of optical identification comprises step:
S1, obtains the depth image including sporter;
In definition world coordinate system, represent left and right, represent upper and lower with y-axis direction with x-axis direction, z-axis direction represents far and near.A usual width two dimensional image can only represent the position of pixel by the data in x-axis and y-axis direction, on image all pixels z-axis direction on value all identical.Depth image refers to that pixel has three-dimensional information, particularly includes the image of the data of z-axis direction distance.Namely, in depth image, the three-dimensional coordinate of pixel in world coordinate system can be specified.For each pixel, except represent its position in three dimensions by x, y, z except, general also have some other parameters, such as, also comprises R, G, B value for representing color and the brightness of pixel, or represent color and the brightness of pixel with C, M, Y, K.
The acquisition of depth image can adopt depth camera device to gather.This depth camera device can be 3 D visual image acquisition system, structure light image acquisition system or flight light image acquisition system.
3 D visual image acquisition system, mainly based on Binocular Vision Principle, usually adopts at least two video cameras to carry out image acquisition to target object or target scene, and calculates Formation Depth figure by data.Particularly, to have the 3 D visual image acquisition system of two video cameras, these two video cameras are commissarial two eyes usually, have overlapping image acquisition region, are called common scene.In a certain moment, two video cameras carry out image acquisition to this common scene, obtain the image that two width have 2-D data information.By the correlation formula of Binocular Vision Principle, and the parameter such as relative position, angle between these two video cameras, thus determine the three-dimensional coordinate of each pixel in this common scene, thus form a width depth map.
Certainly, in concrete data handling procedure, also only can carry out data calculating to the subregion in common scene, only obtain the three-dimensional coordinate of each pixel in this subregion.The demand of this subregion demand fulfillment subsequent treatment of the present invention.
Structure light image acquisition system mainly comprises a coherent source and a video camera.This coherent source launches the pattern with characteristic speckle in space.The pattern that video camera shooting object under test reflects.The characteristic such as shape, size of the pattern of this reflection and object under test distance coherent source distance or with the distance of video camera distance, there is certain rule.Such as, the shape of pattern that reflects of object under test A place or size are directly proportional to the distance in the z-axis direction of coherent source to this A place.Calculated shape or the size of the pattern of A place reflection by follow-up data process, just instead can release the distance in this distance coherent source z-axis direction, A place.Thus calculate the three-dimensional coordinate at this A place.
Flight light image acquisition system comprises impulse sender, detector and timer.Impulse sender is to target object emission of light, and the light of detector detection of a target object reflection, timer sends pulse and rises to this detector for writing down this impulse sender and detect time difference between reflection ray.By certain data rectification etc., calculate the concrete flight time of light pulse.The flight time of foundation light and the light velocity calculate the distance in the z-axis direction of target object.
When sporter motion in a region of specifying, or specify range of motion time, the acquisition of depth image can also be gathered by two dimensional image acquisition device.Particularly, (only need to obtain the data of sporter in a dimension) when sporter carries out jigging separation, the moving region of sporter can be defined as a bar-shaped zone.For a video camera, sporter is equivalent to move in a two dimensional surface.At this moment, image acquisition can be carried out with two dimensional image acquisition device.Two dimensional image is converted into depth image by the z-axis direction distance afterwards by presetting, and (depth image now can be understood as and the two dimensional image of original acquisition is carried out enlarged image to z-axis direction apart from relevant ratio according to one, in this image, the size in each pixel z-axis direction is all identical), thus calculate the exercise data of sporter.
Such as adopt single camera visible images collecting device, obtain the visible images of target scene.Or use single camera infrared light image collecting device, obtain the image of the infrared band of target.The feature of visible images is abundant information, but disturbs more, and date processing difficulty is large.To the image of infrared band, be characterized in that interference is less.When adopting infrared light image collecting device, infrared fileter can be added before photographic head.Can also set up an infrared light supply, user irradiates target scene.Target scene refers to the scene at target object place.
Obtain at this step S1 the depth image including sporter and refer to that sporter is in target scene, can catch by depth image acquisition device.Or sporter has in the region of limited boundary at one and moves, sporter catch by two dimensional image acquisition device, and respective image is converted into depth image.
Step S2, obtains sporter's body data;
Sporter's body data comprises the data that sporter's height, sporter's brachium (length of both arms horizontal development), lower limb length, foot long (length of sole, the i.e. size of foot) etc. represent human body limb length.In addition the information that human body face view data etc. can verify sporter's identity can also be comprised.
Particularly, in the present invention, obtaining sporter's body data is adopt optical recognition system to obtain.Human Height data instance is obtained with single camera visible images acquisition system:
When measuring height, need the human body identified to be head and foot, then the mathematic interpolation of foundation human body head and foot bottom goes out the height of human body.Calculating Human Height and only need a two field picture, is optimization data, reduces error and also can calculate multiple image, scheme selection numerical value that averaged or foundation are preset.To calculate height from a two field picture, concrete identifying comprises:
The head grader identification head of A1, use training in advance;
With the gradient orientation histogram (hereinafter referred to as HOG, HistogramofGradient) of head zone in human body visible images for features training grader, gather abundant sample information training, this is the training stage.Through the process of step S1, human body visible images be herein depth image in the present embodiment, this human body visible images may not be depth image.
After training head grader, from human body visible images, extract the HOG feature of head zone, call the head grader trained, judge the head that whether there is human body in image-region.When image-region comprises the head of human body, record the region that this comprises head.
To each region in image, the gradient direction angle of each pixel in domain.The gradient direction angle of each pixel is combined to form the rectangular histogram of gradient direction angle, and this is the HOG feature in region.
In HOG feature, the process of asking for of gradient direction angle is: judge that the image type gathered is coloured image or gray level image; As then image being converted into gray level image for coloured image, as then do not processed for gray level image.
Afterwards, represent that in piece image, coordinate is the brightness of the pixel of (i, j) if gray level image is I (i, j), I (i, j).Gradient direction angle is:
In follow-up visible images, it is basically identical that method is stated in the acquiring method of HOG feature and this place.
The grader identification foot of foot of B1, use training in advance
With the HOG features training foot grader of foot areas in human body visible images.During identification, first from human body visible images, extract HOG feature, then call the foot's grader trained, judge whether there is human foot in image-region.When there is human foot, record comprises the coordinate information etc. in the region of foot.
After identification head and foot, just can calculate the height data of human body.
To the info class of human body brachium, face-image, lower limb is long, foot is long acquisition and above-mentioned acquisition head, foot seemingly.Particularly, train corresponding grader exactly, during use, calling classification device identifies.
For single camera infrared light image acquisition system, introduce the present invention and how to identify calculating sporter body data.Infrared light is that image information is few relative to the feature of visible ray, but extracts simple and stable.The basic thought of infrared light image identification sporter body data is: paste on the position of the human body relevant with measure the item or wear infrared light labelling, and described infrared light labelling is the reflective object stronger to Infrared reflectance; Launch Infrared with infrared light supply and irradiate infrared light labelling; By the video camera photographic subjects scene with infrared fileter, and obtain depth image.
In the depth map obtained, except infrared light gauge point is brighter, other regional luminances are relatively little.The difference in brightness in infrared light gauge point and other regions is larger.Infrared light gauge point position in the picture can be extracted by binary conversion treatment.Described binary conversion treatment refers to and presets a luminance threshold, judges whether the brightness value of a certain pixel is greater than this luminance threshold.If the brightness value of this pixel is greater than this luminance threshold, then think that this pixel is infrared light gauge point imaging.If the brightness value of this pixel is less than this luminance threshold, then think that this pixel is background pixel.
When adopting single camera infrared light image acquisition system to carry out height measurement, measuring process comprises: paste or wear a first infrared light labelling human foot, pastes or wear a second infrared light labelling at head; Extract this first infrared light labelling and this second infrared light and be marked at position on image, even if Human Height.
For the introduction identification of depth camera device and calculating sporter body data, relative theory is: in depth image, each pixel is the three-dimensional coordinate of target scene mid point, identifies the coordinate that corresponding site just can obtain corresponding site.
Be measured as example with height equally, comprise with the process that depth camera device carries out identifying:
The degree of depth head grader identification head that A2, use preset
dimension coordinate.
After training degree of depth head grader, from depth image, extract the HOG feature of head zone, call the head grader trained, judge the head that whether there is human body in image-region.When image-region comprises the head of human body, record the region that this comprises head, obtain the three-dimensional coordinate of head.
HOG feature also refers to the rectangular histogram that the gradient direction angle of each pixel in image forms herein.When depth camera imaging, the information due to each pixel is the three-dimensional coordinate of scene point, therefore asks during HOG feature and only needs to use z-axis coordinate information.
Particularly, represent that xy direction coordinate is the z-axis coordinate figure of the pixel of (i, j) if z coordinate image is Z (i, j), Z (i, j), corresponding gradient direction angle is:
In subsequent calculations depth image, the method for HOG feature is basic consistent with preceding method.
The grader identification foot portion of degree of depth foot that B2, use preset;
With the HOG features training degree of depth foot grader of foot areas in depth image.During identification, first from depth image, extract HOG feature, then call the foot's grader trained, judge whether there is human foot in image-region.When there is human foot, record comprises the coordinate information etc. in the region of foot.
After identification head and foot, just can calculate the height data of human body.
Step S3, the characteristic point coordinate adapted with sports events during obtaining sporter's motion start-stop;
Step S4, calculates and forms exercise data.
The motion that the sports events of general physical examinations has long-jump, high jump, abdominal crunch, step-on testing, sitting body anteflexion, rope skipping and shuttlecock kicking etc. can carry out in a relatively limited space.Such as, the motion can carried out in the space of length, width and height within 5m*5m*3.5m.
To each sports events, the characteristic point related to all has their own characteristics each.For long-jump, long-jump need to measure take-off constantly tiptoe coordinate and land time heel coordinate, tiptoe when corresponding characteristic point is respectively take-off and heel when landing.For high jump, the maximum of assigned address and ground distance during needing to obtain motion start-stop, the such as maximum on waist assigned address distance ground.For abdominal crunch, human body head can be set to characteristic point, when portion is positioned at the primary importance preset, think that people lies flat right overhead; When head is positioned at the second position preset within first fixed time, think that people reaches rapidly sitting posture head near foot portion; When head returns to again this primary importance within second fixed time, think that people completes an abdominal crunch action.For rope skipping, respectively in rope skipping and on human body settle an index point, when the index point on skipping rope in the given time continuous print motion one week time and index point on human body is beated once within the identical time, can think that rope skipping rotation is once.For shuttlecock kicking, can think that foot is a characteristic point, shuttlecock is another characteristic point; When the coordinate of shuttlecock and foot is less than a preset value, think that shuttlecock contacts with foot; When shuttlecock speeds away foot in a Preset Time, think that shuttlecock is kicked up; With kick up add contact for one-period, each computation of Period be shuttlecock kicking once.To sitting body anteflexion and step-on testing, also characteristic point setting can be carried out according to the principle described in previous example.
For long-jump project, describe and adopt aforementioned a few class camera head to carry out long-jump method for measuring, particularly:
1, to single camera visible images acquisition system
In A3, image when take-off, identify tiptoe position;
With the HOG features training foot grader of foot areas in human body visible images.During identification, analyze image during take-off, from image during take-off, extract HOG feature, call the foot's grader trained, judge the foot that whether there is human body in image-region.As existed, then record the positional information of foot areas.
Suppose that single camera visible images acquisition system is arranged with horizontal direction, and hypothesis transfer direction is the direction that x-axis becomes large, then assert the position of tiptoe when take-off in image, some position that in foot areas, x coordinate is maximum.
In B3, image when landing, identify heel position.
Analyze image when landing, from image when landing, extract HOG feature, call the foot's grader trained, judge the foot that whether there is human body in image-region.As existed, then record the positional information of foot areas.
Suppose that single camera visible images acquisition system is arranged with horizontal direction, and hypothesis transfer direction is the direction that x-axis becomes large, then assert the position of heel when take-off in image, some position that in foot areas, x coordinate is minimum.
Obtain take-off time tiptoe position and land time heel position after, just can calculate the long-jump distance of sporter.
Certainly, such scheme is one and simply identifies the remote scheme of jumping, also can have some change.In such as B3, identification be also the position of tiptoe, now only need the size data result calculated being deducted again sporter's foot.The size of sporter's foot can preset before motion.
2, to single camera infrared light image acquisition system
For long-jump, due to heel and tiptoe will be measured, then paste respectively on heel and tiptoe or wear an infrared light labelling.
The photographed scene imaging when take-off, extracts tiptoe image coordinate; Land the imaging of thing photographed scene, extract heel image coordinate; Suppose that camera horizon is placed, people jumps from left to right, then starting point or pick-up point imaging, in two labellings, left side be heel, right side be tiptoe.The picture position of record ski-jump tiptoe and the picture position of pick-up point heel.Thus obtain long-jump data.
3, to depth camera device
In A4, image when take-off, identify tiptoe position;
With the HOG features training degree of depth foot grader of foot areas in depth image.During identification, analyze image during take-off, from image during take-off, extract HOG feature, call the degree of depth foot grader trained, judge the foot that whether there is human body in image-region.As existed, then record the positional information of foot areas.
Suppose that depth camera device is arranged with horizontal direction, and hypothesis transfer direction is the direction that x-axis becomes large, then assert the position of tiptoe when take-off in image, some position that in foot areas, x coordinate is maximum.
In B3, image when landing, identify heel position.
Analyze depth image when landing, from image when landing, extract HOG feature, call the degree of depth foot grader trained, judge the foot that whether there is human body in image-region.As existed, then record the positional information of foot areas.
Suppose that depth camera device is arranged with horizontal direction, and hypothesis transfer direction is the direction that x-axis becomes large, then assert the position of heel when take-off in image, some position that in foot areas, x coordinate is minimum.
Obtain take-off time tiptoe position and land time heel position after, just can calculate the long-jump distance of sporter.
Relative to prior art, because the physical examinations method and system based on optical identification of the present invention adopts optical instrument automatically to identify body data and the exercise data of sporter, realize unmanned measurement and score.Thus make that the speed of test job is promoted, time shorten, human cost reduce.
Meanwhile, because data processing method used and standard all preset, can strictly unification, thus various interference factors when avoiding manual measurement.Thus ensure the standard comparatively unification of sporter's physical examinations, more there is fairness.
Meanwhile, owing to not needing the participation of people, also stop various potential doing wrong to serve one's friends or relatives.Make that physical examinations result is relatively reliable, degree of belief is relatively high.
Some preferred embodiments of physical examinations method and system based on optical identification of the present invention, in other embodiment, during as employing depth camera device, calculate the characteristic point position of sports events and also can arrange index point, to facilitate date processing; During as adopted single camera visible images acquisition system, the characteristic point position of sports events being calculated and also can not index point be set.
It is pointed out that when adopting single camera visible images acquisition system and single camera infrared light image acquisition system, needing the range of activity limiting human body.
When calculating the data of body data or sporter's all the sports, also by two in single camera visible images acquisition system, single camera infrared light image acquisition system and depth camera device three or all can combine, thus the precision of the data result of calculation of raising body data, sports events.Such as, the depth image that depth camera device obtains can combine with the visible images of single camera visible images acquisition system collection, in the lump for the result of the data of the body data result or sports events that calculate sporter; Also depth image can be adopted to calculate use as a result, adopt visible images to verify use as a result, alternatively.
The advantage of the dissimilar imaging system of two classes is used for the depth camera device of structure based light and single camera visible images acquisition system description taken in conjunction:
The three-dimensional information of scene that what the depth camera device of structure based light obtained is, the color information of scene that what single camera visible images acquisition system obtained is, the information attribute that two kinds of imaging system acquires arrive is different, message complementary sense.
In structured-light system, when an object A blocks another object B in scene, three-dimensional coordinate can not be obtained near the intersection of object A and object B.
In single camera visible images acquisition system, the visible images of camera acquisition scene, if the brightness of object A and object B or color information, can obtain the border of the two in visible images.
The prior coordinate corresponding relation having demarcated structured-light system and VISIBLE LIGHT SYSTEM, can obtain object A and the real three-dimensional boundaries of object B according to the two-dimentional border obtained in visible images.
What more than say is that the combination of two type images is carried out calculating the precision that can improve operation result.
In sports events, need the visual light imaging of user to be fused in fantasy sport scene, be subject to the more real kinesthesia of people, therefore, people can be identified in depth image, then extract human body pixel from visible images.
Physical examinations system based on optical identification of the present invention also comprises some other modules, is rich in greater functionality to make the present invention.
Particularly, the physical examinations system based on optical identification of the present invention also comprises a physiological data collection module, this physiological data collection module may be used for gather people heart rate, blood pressure, vital capacity etc. in one or more.This physiological data collection module can be mated with some sports events and used, such as with step-on testing with the use of.This physiological data collection module and physical examinations system can adopt wired or wireless mode to transmit data.
Physical examinations system based on optical identification of the present invention also comprises a data processing module.Apparently, two dimensional image, for the treatment of image, is converted into depth image by this data processing module, calculates the characteristic point information related in sports events, forms exercise data, and controls whole system work etc.
Physical examinations system based on optical identification of the present invention also comprises a sports events identification or setting module.This sports events identification or setting module can preset the type of sports events, or can automatically identify according to the motor process of sporter the sports events that sporter carries out.
Physical examinations system based on optical identification of the present invention also comprises a receiver module and a data memory module.This receiver module receives the data flow, control instruction stream etc. of external transmission by wired or wireless mode.The data flow transmitted outside is stored into memory module.All view data that memory module also produces for storing moving person physical examinations device and data processed result.
Also picture recording module is comprised, the motor process image of this picture recording module record sporter in physical examinations system of the present invention.
Physical examinations system based on optical identification of the present invention also comprises a camera module, and the face-image of sporter is recorded by this camera module.This face-image and this sporter's data one_to_one corresponding, finally can utilize outside print data, by the face-image of sporter and physical examinations data all together with print.
Physical examinations system based on optical identification of the present invention also comprises an encrypting module, can be encrypted data all in native system.
The present embodiment best can comprehensively above camera module, picture recording module and enciphered data.Utilize face-image, motion video recording monitors, prevent someone from top of assuming another's name pasting.Simultaneously to all data encryptions, prevent people's altered data.
Physical examinations system based on optical identification of the present invention also comprises an accessory module.The stage property of this accessory module required for all kinds of special sports events is such as the markd shuttlecock of tool, rope skipping, step etc.
Physical examinations system based on optical identification of the present invention also comprises an appearance collection and contrast module, may be used for the profiling information gathering sporter, and this profiling information collection is stored, contrast gathering gained appearance data, for the authentication of sporter with the initial data prestored.
Physical examinations system based on optical identification of the present invention also comprises a user log-in block.This user log-in block logs according to one or more authentication of users in the build image of user, face-image, password, sound and gait.The information such as the build image of user, face-image, password, sound and gait can be stored in this memory module.Now, because sporter's identity does not adopt manual verification, therefore can reduce the appearance of physical examinations replacer, that invigorates health effectively and detect is fair and reasonable.
Physical examinations system based on optical identification of the present invention also comprises an exercise data comparation and assessment module, this exercise data comparation and assessment module preset standard data, this exercise data comparation and assessment module contrasts according to this normal data and the exercise data collected, and exports comparing result.Certainly, this exercise data comparation and assessment module also can not preset standard data, this exercise data comparation and assessment module comprises data sending terminal, exercise data is sent in cloud server, preset standard data are set in cloud server, and after exercise data and preset standard data being compared, comparison result is issued these transport data comparation and assessment module.
Certainly, the physical examinations system based on optical identification of the present invention can also comprise a personal-machine interactive device and a data outputting module.This human-computer interaction device comprises at least one in display device, keyboard, mouse and a speaker.Whether this data outputting module detects has external memory to be connected to this physical examinations system, having detected that data are transferred in this external memory according to preset format when being connected to this physical examinations system by external memory.
Above the physical examinations method based on optical identification provided by the present invention and the physical examinations system based on optical identification are described in detail, apply specific case herein to set forth principle of the present invention and embodiment, the explanation of above embodiment just understands core concept of the present invention for helping; Meanwhile, for one of ordinary skill in the art, according to thought of the present invention, all will change in specific embodiments and applications, and all these improve and convert the protection domain that all should belong to claims of the present invention.In sum, this description should not be construed as limitation of the present invention.
Claims (21)
1., based on a physical examinations method for optical identification, it is characterized in that, it comprises:
Obtained the depth image including sporter by depth camera device, obtained brightness or the color information of scene by single camera visible images acquisition system;
Obtain sporter's body data, wherein, when depth camera device cannot obtain the three-dimensional coordinate on border, the brightness got by single camera visible images acquisition system or color information obtain two-dimentional border, and then obtain three-dimensional boundaries according to the coordinate corresponding relation of the depth camera device demarcated in advance and single camera visible images acquisition system;
The characteristic point coordinate adapted with sports events during obtaining sporter's motion start-stop;
Calculate and form exercise data.
2. physical examinations method according to claim 1, is characterized in that, this sporter's body data comprises at least one in height, brachium and foot length.
3. physical examinations method according to claim 2, is characterized in that, this physical examinations method also comprises at least one in the physiological data and appearance data processing sporter.
4. physical examinations method according to claim 2, is characterized in that, the body data of this physical examinations method also comprises encrypted motion person and exercise data.
5. based on a physical examinations system for optical identification, it is characterized in that, this detection system comprises
Optical recognition system, adopts depth camera identification sporter motion, Formation Depth view data, and the brightness or the color information that are obtained scene by single camera visible images acquisition system;
Data handling system, be connected with optical recognition system, process this depth image data, when depth image data cannot obtain the three-dimensional coordinate on border, the brightness got according to single camera visible images acquisition system or color information obtain two-dimentional border, and then the coordinate corresponding relation combining depth camera and the single camera visible images acquisition system of having demarcated in advance obtains three-dimensional boundaries, forms exercise data thus according to the type of sports date processing depth image data received.
6. physical examinations system according to claim 5, it is characterized in that, these type of sports data are external setting-up.
7. physical examinations system according to claim 5, it is characterized in that, this physical examinations system also comprises movement recognition system, is connected with data handling system with this optical recognition system, gives data handling system for depth image data identification type of sports Data Concurrent.
8. physical examinations system according to claim 5, it is characterized in that, this optical recognition system obtains in very first time point the first view data including this sporter, obtain the second view data including this sporter in the second time point, and determine to represent motion characteristics point coordinates according to image.
9. physical examinations system according to claim 8, is characterized in that, this characteristic point coordinate comprises the characteristic point coordinate obtaining very first time point and the characteristic point coordinate obtaining the second time point.
10. physical examinations system according to claim 8 or claim 9, it is characterized in that, this characteristic point coordinate is space coordinates.
11. physical examinations systems according to claim 5, is characterized in that, also comprise a receiving system, can receive the third party's data transmitted outside.
12., according to physical examinations system described in claim 11, is characterized in that, also comprise a Camcording system, to all physical examinations process video recordings.
13. physical examinations systems according to claim 12, is characterized in that, also comprise an encryption system, be connected with this data handling system, are encrypted Video data, third party's data, depth image data and exercise data.
14. physical examinations systems according to claim 5, is characterized in that, this type of sports is at least one among long-jump, step-on testing, abdominal crunch, sitting body anteflexion, high jump, rope skipping, shuttlecock kicking.
15. physical examinations systems according to claim 5, it is characterized in that, this physical examinations system also comprises user log-in block, and this user log-in block logs according to one or more authentication of users in the build image of user, face-image, password, sound and gait.
16. physical examinations systems according to claim 5, it is characterized in that, this physical examinations system also comprises exercise data comparation and assessment module, this exercise data comparation and assessment module preset standard data, this exercise data comparation and assessment module contrasts according to this normal data and the exercise data collected, and exports comparing result.
17. physical examinations systems according to claim 5, it is characterized in that, this physical examinations system also comprises exercise data and transmits receiver module, and this exercise data transmits receiver module and is used for exercise data to be sent to server end, and from received server-side data processed result.
18. physical examinations systems according to claim 5, is characterized in that, this physical examinations system also comprises camera module, the face-image of this camera module record sporter.
19. physical examinations systems according to claim 5, it is characterized in that, this physical examinations system also comprises a personal-machine interactive device, this human-computer interaction device comprises at least one in a display device, keyboard, mouse.
20. physical examinations systems according to claim 5, it is characterized in that, this physical examinations system comprises a data outputting module, whether this data outputting module detects has external memory to be connected to this physical examinations system, having detected that data are transferred in this external memory according to preset format when being connected to this physical examinations system by external memory.
21. physical examinations systems according to claim 5, it is characterized in that, this physical examinations system also comprises a loudspeaker module, after this exercise data is disposed, this loudspeaker module plays this exercise data result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110357613.7A CN103099602B (en) | 2011-11-10 | 2011-11-10 | Based on the physical examinations method and system of optical identification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110357613.7A CN103099602B (en) | 2011-11-10 | 2011-11-10 | Based on the physical examinations method and system of optical identification |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103099602A CN103099602A (en) | 2013-05-15 |
| CN103099602B true CN103099602B (en) | 2016-04-06 |
Family
ID=48308038
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110357613.7A Active CN103099602B (en) | 2011-11-10 | 2011-11-10 | Based on the physical examinations method and system of optical identification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103099602B (en) |
Families Citing this family (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104394386A (en) * | 2014-12-11 | 2015-03-04 | 盐城工学院 | System and method for evaluating knee joint motion normativity through multi-view sequence images |
| CN104537670B (en) * | 2014-12-31 | 2017-08-04 | 深圳泰山体育科技股份有限公司 | A kind of pliability method of testing and system |
| CN104586404A (en) * | 2015-01-27 | 2015-05-06 | 深圳泰山在线科技有限公司 | Method and system for identifying posture of fitness and health monitoring |
| CN104688237B (en) * | 2015-02-11 | 2017-06-20 | 深圳泰山体育科技股份有限公司 | The time study method and system of physical examinations |
| CN104665838B (en) * | 2015-02-11 | 2017-08-04 | 深圳泰山体育科技股份有限公司 | Physical examinations method and system |
| CN104720817A (en) * | 2015-04-02 | 2015-06-24 | 吴爱好 | Height measuring instrument |
| JP6482102B2 (en) * | 2015-06-26 | 2019-03-13 | Ｎｅｃソリューションイノベータ株式会社 | Measuring apparatus and measuring method |
| JP6888250B2 (en) * | 2015-12-28 | 2021-06-16 | カシオ計算機株式会社 | Image processing equipment, measuring equipment, image processing system, image processing method and program |
| TWI598075B (en) * | 2016-08-09 | 2017-09-11 | 崑山科技大學 | Sit and reach detector system and method thereof |
| CN106175693A (en) * | 2016-08-30 | 2016-12-07 | 昆明东方清瑶美容有限公司 | The processing method of tcm constitution data and system |
| CN106454039B (en) * | 2016-10-20 | 2020-01-14 | 广东技术师范大学 | Seat body anteflexion detector based on thing networking and machine vision |
| CN106621284B (en) * | 2016-12-28 | 2018-08-17 | 湖南科技大学 | A kind of intelligent physical examinations platform |
| KR102002279B1 (en) * | 2017-04-06 | 2019-07-23 | 한국한의학연구원 | Apparatus for diaagnosing three dimensional face |
| CN109200563A (en) * | 2017-06-30 | 2019-01-15 | 深圳泰山体育科技股份有限公司 | The optical calibrating and measurement method of sit-ups test |
| CN107480612A (en) * | 2017-07-31 | 2017-12-15 | 广东欧珀移动通信有限公司 | Recognition methods, device and the terminal device of figure action |
| CN108122614A (en) * | 2018-02-28 | 2018-06-05 | 张喆 | A kind of national physique detection and exercise management system and method |
| CN109876416B (en) * | 2019-03-26 | 2020-06-19 | 浙江大学 | A skipping rope counting method based on image information |
| CN110136818A (en) * | 2019-05-23 | 2019-08-16 | 深圳市恒康佳业科技有限公司 | A kind of novel physique test instrument smart host |
| CN110251925B (en) * | 2019-05-27 | 2020-09-25 | 安徽康岁健康科技有限公司 | Physique detection system and working method thereof |
| CN111012357A (en) * | 2019-12-06 | 2020-04-17 | 西南交通大学 | A device and method for detecting forward flexion of sitting body based on image recognition |
| CN111700624B (en) * | 2020-07-27 | 2024-03-12 | 中国科学院合肥物质科学研究院 | Pattern recognition method and system for detecting motion gesture by intelligent bracelet |
| US20220313526A1 (en) * | 2021-03-30 | 2022-10-06 | Hill-Rom Services, Inc. | Bed and mattress auto settings for intubation |
| CN113198166B (en) * | 2021-04-30 | 2022-06-28 | 光彻科技(杭州)有限公司 | Long jump detection method and detection device |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101237522A (en) * | 2007-02-02 | 2008-08-06 | 华为技术有限公司 | Motion detection method and device |
| CN102194105A (en) * | 2010-03-19 | 2011-09-21 | 微软公司 | Proxy training data for human body tracking |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4007899B2 (en) * | 2002-11-07 | 2007-11-14 | オリンパス株式会社 | Motion detection device |
| US7782358B2 (en) * | 2007-06-08 | 2010-08-24 | Nokia Corporation | Measuring human movements—method and apparatus |
-
2011
- 2011-11-10 CN CN201110357613.7A patent/CN103099602B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101237522A (en) * | 2007-02-02 | 2008-08-06 | 华为技术有限公司 | Motion detection method and device |
| CN102194105A (en) * | 2010-03-19 | 2011-09-21 | 微软公司 | Proxy training data for human body tracking |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103099602A (en) | 2013-05-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103099602B (en) | Based on the physical examinations method and system of optical identification | |
| Bartol et al. | A review of body measurement using 3D scanning | |
| CN109815907B (en) | Sit-up posture detection and guidance method based on computer vision technology | |
| CN111263953B (en) | Action state evaluation system, device, server, method and storage medium thereof | |
| CN110321754B (en) | Human motion posture correction method and system based on computer vision | |
| CN106037651B (en) | A kind of heart rate detection method and system | |
| CN108156450A (en) | For the method for calibration camera, calibrator (-ter) unit, calibration system and machine readable storage medium | |
| CN106471523B (en) | Color identification using infrared imaging | |
| CN105956586A (en) | Intelligent tracking system based on TOF 3D camera | |
| KR101555863B1 (en) | Device and method for taekwondo poomsae readout and education | |
| CN107930048B (en) | Space somatosensory recognition motion analysis system and motion analysis method | |
| CN108259887B (en) | Gaze point calibration method and device, gaze point calibration method and device | |
| CN113239797B (en) | Human body action recognition method, device and system | |
| CN112668549B (en) | Pedestrian attitude analysis method, system, terminal and storage medium | |
| CN106529399A (en) | Human body information acquisition method, device and system | |
| CN112933581A (en) | Sports action scoring method and device based on virtual reality technology | |
| CN106504283A (en) | Information broadcasting method, apparatus and system | |
| Yang et al. | 3-D markerless tracking of human gait by geometric trilateration of multiple Kinects | |
| Hesse et al. | Concurrent validity of a custom method for markerless 3D full-body motion tracking of children and young adults based on a single RGB-D camera | |
| CN115568823B (en) | A method, system and device for evaluating human balance ability | |
| CN206819290U (en) | A kind of system of virtual reality multi-person interactive | |
| KR102475470B1 (en) | System of obtaining athletics video utilizing drone and method of obtainig athletics video utilizing drone | |
| CN106056578A (en) | Method for utilizing video to precisely estimating height and bodily form of human body | |
| Chong et al. | A photogrammetric application in virtual sport training | |
| CN116309712A (en) | A functional motion analysis system and method based on a depth camera |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C56 | Change in the name or address of the patentee | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 518057 Shenzhen, Nanshan District Province, science and technology, South Road, the building of the big square, building 02, room 4, Patentee after: SHENZHEN TAISHAN SPORTS TECHNOLOGY CORP., LTD. Address before: 518057 Shenzhen, Nanshan District Province, science and technology, South Road, the building of the big square, building 02, room 4, Patentee before: Shenzhen Tol Technology Co., Ltd. |
|
| CP01 | Change in the name or title of a patent holder |
Address after: 518057, room 4, building 02, Fangda building, South twelve Road, Nanshan District science and technology, Shenzhen, Guangdong Patentee after: Shenzhen Taishan Sports Technology Co.,Ltd. Address before: 518057, room 4, building 02, Fangda building, South twelve Road, Nanshan District science and technology, Shenzhen, Guangdong Patentee before: SHENZHEN TAISHAN SPORTS TECHNOLOGY Corp.,Ltd. |
|
| CP01 | Change in the name or title of a patent holder |