CN101021948A - Automatic identifying device and method for joint in human body symmetric motion image - Google Patents
Automatic identifying device and method for joint in human body symmetric motion image Download PDFInfo
- Publication number
- CN101021948A CN101021948A CN 200710027147 CN200710027147A CN101021948A CN 101021948 A CN101021948 A CN 101021948A CN 200710027147 CN200710027147 CN 200710027147 CN 200710027147 A CN200710027147 A CN 200710027147A CN 101021948 A CN101021948 A CN 101021948A
- Authority
- CN
- China
- Prior art keywords
- module
- image
- monumented point
- joint
- processing module
- 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
Images
Abstract
This invention provides an automatic identifying device and a method for arthrosis in human body symmetrical moving images, in which, the device includes a high speed vidicon, a file converter, an image processor and a computer, in which, the high speed vidicon, the file converter and the image processor are connected with the computer, and the image processor includes orderly connected image segment module, an automatic identification module and an automatic tracking module, the file converter is connected with the image segment module, the converter and the automatic tracking module are connected with the storage of the computer separately, which can realize automatic identification of arthrosis point of human bodies.
Description
Technical field
The present invention relates to the human motion image processing techniques, specifically be meant the automatic identification equipment and the method in joint in the human body symmetrical moving image.
Background technology
Analyze the collection that human motion at first will realize human body movement data, human body movement data has mainly comprised kinematics amount and dynamics amount, the method that directly obtains the dynamics amount mainly is an ergograph, and the dynamics amount is that the inertial parameter in conjunction with human body calculates on the basis of kinematics amount indirectly.The kinematics amount is by motion-captured acquisition, and human body motion capture can be divided into advantages such as electromagnetic type is motion-captured, electromechanical is motion-captured and optical motion capture, and the optical motion capture system is quick, safe, accurate with it, make things convenient for, and is widely used.
It is comparatively accurate a kind of catching mode that optical motion is caught.Optical motion capture method commonly used is that the gauge point on human joint points is a kind of spheroidite that scribbles special reflectorized material, utilize video camera to take from different perspectives, utilize the image coordinate of gauge point on the software analysis image then, utilize principle of computer vision to carry out three-dimensional reconstruction, draw the exercise data of gauge point.Optical motion is caught main finger high-speed photography treatment of picture and analysis.
Mainly comprise image segmentation, gauge point identification and gauge point tracking, gauge point kinematics dynamics calculation and motion rendition etc. for human motion high-speed camera treatment of picture and analysis.
Edge of image is the key character that image segmentation relies on, and image edge processing mainly contains the differential method (as LOG operator and DOG operator method), fitting process (Prewitt fitting process, Haralick fitting process and Huechel fitting process), analysis of neural network method (as Poggio method and Hopfield method) and becomes yardstick human body rim detection (method of scales manually is set, based on the calculating method of scales of knowledge and definite automatically filter scale method).For the Image Edge-Detection in joint in the symmetry human motion, existing method exists efficient low relatively poor with real-time, is difficult to realize the human synovial requirement of test automatically in the symmetry human motion.
The identification of human motion is the identification of gauge point, and recognizer has method (HMM modelling, dynamically Bayes's method) based on statistics, template matching method (DTW method) and based on the method (FSM method, STS method) of grammer.Stick on the human body plane motion of the gauge point of joint for symmetry, constant this feature of distance between its adjacent segment point is arranged, existing method does not relate to human body symmetrical plane this feature of moving, and this has brought adverse influence to recognition effect.
Existing human synovial gauge point tracking has based on sports ground estimation approach, rectangular area predicting tracing method, human synovial freedom of motion restriction tracking and single order prediction local search approach.Do not having under constraint or the conservative force effect, human synovial can only be done around the near-end link and rotate, on according to the zone, should be and the interval of human synovial axis symmetry that rotation characteristics and organization of human body that existing tracking is not followed human synovial fully limit anatomic characteristic down.
Summary of the invention
The objective of the invention is to overcome the shortcoming and defect of above-mentioned prior art, provide a kind of, the automatic identification equipment in joint in the high efficiency human body symmetry moving image in conjunction with the organization of human body characteristics and the characteristics of motion.
The present invention also aims to provide the automatic identifying method that adopts the automatic identification equipment in joint in the above-mentioned human body symmetrical moving image.
Purpose of the present invention is achieved through the following technical solutions: the automatic identification equipment in joint in this human body symmetry moving image, comprise high-speed camera, file converteractivemil builder activemil builder, image processor, computing machine, described high-speed camera, file converteractivemil builder activemil builder, image processor are connected with computing machine respectively, described image processor comprises the image segmentation module that connects successively, automatic identification module and automatic tracking module, described file converteractivemil builder activemil builder is connected with the image segmentation module, and file converteractivemil builder activemil builder, automatic tracking module are connected with the storer of computing machine respectively.
Described image segmentation module comprises normalized image processing module, layering processing module, thresholding processing module, the edge fitting processing module that connects successively, and the normalized image processing module is connected with described file converteractivemil builder activemil builder;
Described automatic identification module comprises the artificial set handling module of first frame, monumented point edge coordinate processing module, the monumented point position computation module that connects successively, the artificial set handling module of first frame is connected with described edge fitting processing module, and monumented point edge coordinate processing module, monumented point position computation module are connected with described automatic tracking module respectively;
Described automatic tracking module comprises joint monumented point search module, joint position and the displacement computing module that connects successively, and described joint position and displacement computing module are connected with the storer of computing machine.
Adopt the automatic identifying method of the automatic identification equipment in joint in the above-mentioned human body symmetrical moving image, its step comprises:
(1) set up the human motion checkout area comprise that high-speed camera is demarcated and survey mark is demarcated, and guarantee that human motion finishes in described human motion checkout area, human synovial position subsides monumented point, its specific requirement is carried out according to the agreement in the anthropological measuring; In described human motion checkout area, human motion is by the high-speed camera record, and the A/D device that the analog picture signal of record carries by high-speed camera is converted to data image signal and sends computing machine to;
(2) described data image signal is converted to the video format file earlier by file converteractivemil builder activemil builder, be converted to the static frames image again, video format file and static frames image file be by after the memory stores of computing machine, and computing machine sends the static frames image file to image processor with the form of frame;
(3) after described static frames image file handles accordingly through the image segmentation module of image processor, automatic identification module, automatic tracking module, the data of gained are stream-oriented file by the memory stores that automatic tracking module is transported to computing machine, thereby realize the automatic identification in joint in the human body symmetrical moving image.
The processing of described image segmentation module, its step comprises:
(1) by the normalized image processing module described static frames image file is carried out normalized, normalized at first demarcates to determine the unchangeability of " motion " according to setting photogrammetric sign, also with regard to the stationarity of high-speed camera coordinate, in order to accelerate image processing velocity, the normalized image processing module is unified the physical dimension of every two field picture simultaneously;
(2) carry out layering by the static frames image of layering processing module after to above-mentioned normalized and handle, obtain image layer R layer, G layer, B layer, on each image layer, the color dot of gray level image has only 1/65535 quantity of information of coloured image;
(3) in order to guarantee accurately cutting apart of monumented point, the passing threshold processing module is carried out thresholding to image layer R layer, G layer, B layer respectively and is handled, the thresholding processing module is on the basis of system initialization value, total colour according to image is carried out disposable adjustment to controlled variable, this process is only done once, reason be we to suppose to take ready-made environment be constant relatively (generally the actuation time that will analyze is within 1 second);
(4) in order to ensure the automatic identification of monumented point and from motion tracking, the image after above-mentioned thresholding processing module is cut apart carries out an edge fitting by the edge fitting processing module and handles after extreme value is judged, obtains static monumented point image.
The processing of described automatic identification module, its step comprises:
(1) obtains static monumented point image after described image segmentation processing module is handled, by the artificial set handling module of described first frame first two field picture manually is provided with, described artificial setting is meant centre of form title and the position that provides monumented point, determine link between two monumented points with straight line, putting on sequence number for simultaneously first two field picture is 1;
(2) according to the position of first two field picture monumented point, and the position of the two field picture monumented point of all the other sequence numbers that obtain according to the controlled variable of described automatic tracking module, by monumented point edge coordinate processing module the edge of monumented point is carried out profile and handle;
(3) described monumented point position computation module is according to the planimetric coordinates of described monumented point continuous edge point, obtain the position of monumented point and the monumented point distance on same link in conjunction with the resultant moment principle, the data of the monumented point distance on position that comprises picture frame sequence number, monumented point title, monumented point and the same link are sent to the joint position and the displacement computing module of described automatic tracking module.
The processing of described automatic tracking module, its step comprises:
(1) accepts described automatic identification module by described joint monumented point search module and handle the data that obtain, position according to n flag of frame point, and the position of n-1 frame, predict the position of monumented point in the image of n+1 frame, in conjunction with the principle of far-end link around the rotation of near-end link, determine the region of search of monumented point in the n+1 frame, the predicted value of monumented point and region of search is sent to automatic identification module as controlled variable;
(2) joint position and displacement computing module be according to above-mentioned controlled variable, in conjunction with the human innertial parameter, and is true origin with the hip joint, according to the constant characteristic of body segment's length, obtains joint position and angle; The definition that joint position and displacement computing module quicken according to linear velocity, linear acceleration, angular velocity, angle, the position that comprises picture frame sequence number, monumented point title, monumented point of transmitting to described joint position and angle and by described monumented point position computation module and the data of the monumented point distance on the same link are handled, obtain corresponding joint parameter, all joint parameters are transported to described storer and store with stream-oriented file.
The data structure of described joint parameter comprises: the linear acceleration of the position of two field picture sequence number, monumented point title, each monumented point, the linear velocity of each monumented point, each monumented point, the angle of link (comprising hip angle, knee angle, ankle angle), the angular velocity of link, the angular acceleration of link.
The present invention compared with prior art, have following advantage and beneficial effect: the present invention is according to the characteristics of human body symmetrical motion, adopted image partition method based on layer, monumented point can be extracted from movement background and human body fully, monumented point is discerned automatically and is adopted the position of continuous boundary profile coordinate to calculate, when guaranteeing accuracy, accelerated recognition speed, automatic tracking method combines the characteristics of anatomic characteristic and body segment's forms of motion of human body on the basis of conventional motion estimation, the movement tendency of monumented point in the predictive frame, guaranteed the efficient of automatic identification with accurately, calculated all kinematics amounts in the human body symmetrical motion according to the needs of human motion analysis.The present invention can realize the automatic identification of human joint points, and sick and wounded research and rehabilitation, analyzing and diagnosing Motion Technology and human engineering etc. are had great significance.
Description of drawings
Fig. 1 is the fundamental diagram of the automatic identification equipment in joint in the human body symmetry moving image of the present invention;
Fig. 2 is the fundamental diagram of image segmentation module shown in Figure 1;
Fig. 3 is the fundamental diagram of automatic identification module shown in Figure 1;
Fig. 4 is the fundamental diagram of automatic tracking module shown in Figure 1.
Embodiment
Below in conjunction with embodiment and accompanying drawing, the present invention is described in further detail, but embodiments of the present invention are not limited thereto.
Embodiment one
As shown in Figure 1, the automatic identification equipment in joint in this human body symmetry moving image, comprise high-speed camera, file converteractivemil builder activemil builder, image processor, computing machine, described high-speed camera, file converteractivemil builder activemil builder, image processor are connected with computing machine respectively, described image processor comprises the image segmentation module that connects successively, automatic identification module and automatic tracking module, described file converteractivemil builder activemil builder is connected with the image segmentation module, and file converteractivemil builder activemil builder, automatic tracking module are connected with the storer of computing machine respectively.
As shown in Figure 2, described image segmentation module comprises normalized image processing module, layering processing module, thresholding processing module, the edge fitting processing module that connects successively, and the normalized image processing module is connected with described file converteractivemil builder activemil builder;
As shown in Figure 3, described automatic identification module comprises the artificial set handling module of first frame, monumented point edge coordinate processing module, the monumented point position computation module that connects successively, the artificial set handling module of first frame is connected with described edge fitting processing module, and monumented point edge coordinate processing module, monumented point position computation module are connected with described automatic tracking module respectively;
As shown in Figure 4, described automatic tracking module comprises joint monumented point search module, joint position and the displacement computing module that connects successively, and described joint position and displacement computing module are connected with the storer of computing machine.
Shown in Fig. 1～4, the automatic identification equipment in joint is performed such the automatic identification in joint in the human body symmetrical moving image in this human body symmetry moving image:
(1) set up the human motion checkout area comprise that high-speed camera is demarcated and survey mark is demarcated, and guarantee that human motion finishes in described human motion checkout area, human synovial position subsides monumented point, its specific requirement is carried out according to the agreement in the anthropological measuring; In described human motion checkout area, human motion is by the high-speed camera record, and the A/D device that the analog picture signal of record carries by high-speed camera is converted to data image signal and sends computing machine to;
(2) described data image signal is converted to the video format file earlier by file converteractivemil builder activemil builder, be converted to the static frames image again, video format file and static frames image file be by after the memory stores of computing machine, and computing machine sends the static frames image file to image processor with the form of frame;
(3) after described static frames image file handles accordingly through the image segmentation module of image processor, automatic identification module, automatic tracking module, the data of gained are stream-oriented file by the memory stores that automatic tracking module is transported to computing machine, thereby realize the automatic identification in joint in the human body symmetrical moving image.
As shown in Figure 2, the processing procedure of the described image segmentation module of step (3) is:
1, by the normalized image processing module described static frames image file is carried out normalized, normalized at first demarcates to determine the unchangeability of " motion " according to setting photogrammetric sign, also with regard to the stationarity of high-speed camera coordinate, in order to accelerate image processing velocity, the normalized image processing module is unified the physical dimension of every two field picture simultaneously;
2, carry out layering by the static frames image of layering processing module after to above-mentioned normalized and handle, obtain image layer R layer, G layer, B layer, on each image layer, the color dot of gray level image has only 1/65535 quantity of information of coloured image;
3, in order to guarantee accurately cutting apart of monumented point, the passing threshold processing module is carried out thresholding to image layer R layer, G layer, B layer respectively and is handled, the thresholding processing module is on the basis of system initialization value, total colour according to image is carried out disposable adjustment to controlled variable, this process is only done once, reason be we to suppose to take ready-made environment be constant relatively (generally the actuation time that will analyze is within 1 second);
4, in order to ensure the automatic identification of monumented point and from motion tracking, the image after above-mentioned thresholding processing module is cut apart carries out an edge fitting by the edge fitting processing module and handles after extreme value is judged, obtains static monumented point image.
As shown in Figure 3, the processing procedure of the described automatic identification module of step (3) is:
1, obtains static monumented point image after described image segmentation processing module is handled, by the artificial set handling module of described first frame first two field picture manually is provided with, described artificial setting is meant centre of form title and the position that provides monumented point, determine link between two monumented points with straight line, putting on sequence number for simultaneously first two field picture is 1;
2, according to the position of first two field picture monumented point, and the position of the two field picture monumented point of all the other sequence numbers that obtain according to the controlled variable of described automatic tracking module, by monumented point edge coordinate processing module the edge of monumented point is carried out profile and handle;
3, described monumented point position computation module is according to the planimetric coordinates of described monumented point continuous edge point, obtain the position of monumented point and the monumented point distance on same link in conjunction with the resultant moment principle, the data of the monumented point distance on position that comprises picture frame sequence number, monumented point title, monumented point and the same link are sent to the joint position and the displacement computing module of described automatic tracking module.
As shown in Figure 4, the processing procedure of the described automatic tracking module of step (3) is:
1, accepts described automatic identification module by described joint monumented point search module and handle the data that obtain, position according to n flag of frame point, and the position of n-1 frame, predict the position of monumented point in the image of n+1 frame, in conjunction with the principle of far-end link around the rotation of near-end link, determine the region of search of monumented point in the n+1 frame, the predicted value of monumented point and region of search is sent to automatic identification module as controlled variable;
2, joint position and displacement computing module be according to above-mentioned controlled variable, in conjunction with the human innertial parameter, and is true origin with the hip joint, according to the constant characteristic of body segment's length, obtains joint position and angle; The definition that joint position and displacement computing module quicken according to linear velocity, linear acceleration, angular velocity, angle, the position that comprises picture frame sequence number, monumented point title, monumented point of transmitting to described joint position and angle and by described monumented point position computation module and the data of the monumented point distance on the same link are handled, obtain corresponding joint parameter, all joint parameters are transported to described storer and store with stream-oriented file.
The data structure of described joint parameter comprises: the linear acceleration of the position of two field picture sequence number, monumented point title, each monumented point, the linear velocity of each monumented point, each monumented point, the angle of link (comprising hip angle, knee angle, ankle angle), the angular velocity of link, the angular acceleration of link.The data structure of described joint parameter comprises: the linear acceleration of the position of picture frame sequence number, monumented point title, each monumented point, the linear velocity of each monumented point, each monumented point, the angle of link (comprising hip angle, knee angle, ankle angle), the angular velocity of link, the angular acceleration of link.
By experiment and theoretical analysis, the present invention is according to the human body distance of human body adjacent segment in plane motion remains unchanged in the symmetry action feature and human synovial laws of motion (the anatomy scope of activities and the kinematics character in joint), thresholding Flame Image Process, the articulation point of pass through human synovial (in joint decals will) that proposes to the high-speed camera image discern automatically and articulation point from motion tracking, thereby the method that realizes the automatic identification in joint in the human body symmetric motion image is complete feasible.
As mentioned above, just can realize the present invention preferably.
Claims (5)
1, the automatic identification equipment in joint in the human body symmetrical moving image, it is characterized in that: comprise high-speed camera, file converteractivemil builder activemil builder, image processor, computing machine, described high-speed camera, file converteractivemil builder activemil builder, image processor are connected with computing machine respectively, described image processor comprises the image segmentation module that connects successively, automatic identification module and automatic tracking module, described file converteractivemil builder activemil builder is connected with the image segmentation module, and file converteractivemil builder activemil builder, automatic tracking module are connected with the storer of computing machine respectively.
2, according to the automatic identification equipment in joint in the described human body symmetrical moving image of claim 1, it is characterized in that: described image segmentation module comprises normalized image processing module, layering processing module, thresholding processing module, the edge fitting processing module that connects successively, and the normalized image processing module is connected with described file converteractivemil builder activemil builder;
Described automatic identification module comprises the artificial set handling module of first frame, monumented point edge coordinate processing module, the monumented point position computation module that connects successively, the artificial set handling module of first frame is connected with described edge fitting processing module, and monumented point edge coordinate processing module, monumented point position computation module are connected with described automatic tracking module respectively;
Described automatic tracking module comprises joint monumented point search module, joint position and the displacement computing module that connects successively, and described joint position and displacement computing module are connected with the storer of computing machine.
3, adopt the automatic identifying method in joint in the human body symmetrical moving image of the automatic identification equipment in joint in the described human body symmetrical moving image of claim 1, it is characterized in that may further comprise the steps:
(1) set up the human motion checkout area comprise that high-speed camera is demarcated and survey mark is demarcated, and guarantee that human motion finishes in described human motion checkout area, human synovial position subsides monumented point, its specific requirement is carried out according to the agreement in the anthropological measuring; In described human motion checkout area, human motion is by the high-speed camera record, and the A/D device that the analog picture signal of record carries by high-speed camera is converted to data image signal and sends computing machine to;
(2) described data image signal is converted to the video format file earlier by file converteractivemil builder activemil builder, be converted to the static frames image again, video format file and static frames image file be by after the memory stores of computing machine, and computing machine sends the static frames image file to image processor with the form of frame;
(3) after described static frames image file handles accordingly through the image segmentation module of image processor, automatic identification module, automatic tracking module, the data of gained are stream-oriented file by the memory stores that automatic tracking module is transported to computing machine, thereby realize the automatic identification in joint in the human body symmetrical moving image.
4, according to the automatic identifying method in joint in the described human body symmetrical moving image of claim 3, it is characterized in that:
The processing of described image segmentation module may further comprise the steps:
(1) by the normalized image processing module described static frames image file is carried out normalized, normalized at first demarcates to determine the unchangeability of " motion " according to setting photogrammetric sign, also with regard to the stationarity of high-speed camera coordinate, simultaneously, the normalized image processing module is unified the physical dimension of every two field picture;
(2) carry out layering by the static frames image of layering processing module after to above-mentioned normalized and handle, obtain image layer R layer, G layer, B layer, on each image layer, the color dot of gray level image has only 1/65535 quantity of information of coloured image;
(3) the passing threshold processing module is carried out the thresholding processing to image layer R layer, G layer, B layer respectively, and the thresholding processing module is carried out disposable adjustment according to total colour of image to controlled variable on the basis of system initialization value;
(4) image after above-mentioned thresholding processing module is cut apart carries out an edge fitting by the edge fitting processing module and handles after extreme value is judged, obtains static monumented point image;
The processing of described automatic identification module may further comprise the steps:
(1) obtains static monumented point image after described image segmentation processing module is handled, by the artificial set handling module of described first frame first two field picture manually is provided with, described artificial setting is meant centre of form title and the position that provides monumented point, determine link between two monumented points with straight line, putting on sequence number for simultaneously first two field picture is 1;
(2) according to the position of first two field picture monumented point, and the position of the two field picture monumented point of all the other sequence numbers that obtain according to the controlled variable of described automatic tracking module, by monumented point edge coordinate processing module the edge of monumented point is carried out profile and handle;
(3) described monumented point position computation module is according to the planimetric coordinates of described monumented point continuous edge point, obtain the position of monumented point and the monumented point distance on same link, the data of the monumented point distance on position that comprises picture frame sequence number, monumented point title, monumented point and the same link are sent to the joint position and the displacement computing module of described automatic tracking module;
The processing of described automatic tracking module may further comprise the steps:
(1) accepts described automatic identification module by described joint monumented point search module and handle the data that obtain, position according to n flag of frame point, and the position of n-1 frame, predict the position of monumented point in the image of n+1 frame, determine the region of search of monumented point in the n+1 frame, the predicted value of monumented point and region of search is sent to automatic identification module as controlled variable;
(2) joint position and displacement computing module be according to above-mentioned controlled variable, in conjunction with the human innertial parameter, and is true origin with the hip joint, according to the constant characteristic of body segment's length, obtains joint position and angle; The position that comprises picture frame sequence number, monumented point title, monumented point that joint position and displacement computing module transmit to described joint position and angle and by described monumented point position computation module and the data of the monumented point distance on the same link are handled, obtain corresponding joint parameter, all joint parameters are transported to described storer and store with stream-oriented file.
5, according to the automatic identifying method in joint in the described human body symmetrical moving image of claim 4, it is characterized in that the data structure of described joint parameter comprises: the linear acceleration of the position of two field picture sequence number, monumented point title, each monumented point, the linear velocity of each monumented point, each monumented point, the angle of link, the angular velocity of link, the angular acceleration of link; The angle of described link comprises hip angle, knee angle, ankle angle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200710027147 CN101021948A (en) | 2007-03-14 | 2007-03-14 | Automatic identifying device and method for joint in human body symmetric motion image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200710027147 CN101021948A (en) | 2007-03-14 | 2007-03-14 | Automatic identifying device and method for joint in human body symmetric motion image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101021948A true CN101021948A (en) | 2007-08-22 |
Family
ID=38709704
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200710027147 Pending CN101021948A (en) | 2007-03-14 | 2007-03-14 | Automatic identifying device and method for joint in human body symmetric motion image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101021948A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102668572A (en) * | 2009-12-22 | 2012-09-12 | 汤姆逊许可证公司 | Method and apparatus for optimal motion reproduction in stereoscopic digital cinema |
| CN103533242A (en) * | 2013-10-15 | 2014-01-22 | 中国科学院深圳先进技术研究院 | Method and system for extracting and tracking cursor point in out-of-focus video |
| CN104898837A (en) * | 2015-05-22 | 2015-09-09 | 燕山大学 | Portable hand virtual rehabilitation experiment box based on gesture interaction and method |
| CN104887238A (en) * | 2015-06-10 | 2015-09-09 | 上海大学 | Hand rehabilitation training evaluation system and method based on motion capture |
| CN105832343A (en) * | 2016-05-22 | 2016-08-10 | 上海大学 | Multi-dimensional vision hand function rehabilitation quantitative evaluation system and evaluation method |
| CN109945594A (en) * | 2019-03-22 | 2019-06-28 | 上海五宝网络科技有限公司 | Intelligent vision refrigerator based on dynamic video detection |
-
2007
- 2007-03-14 CN CN 200710027147 patent/CN101021948A/en active Pending
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102668572A (en) * | 2009-12-22 | 2012-09-12 | 汤姆逊许可证公司 | Method and apparatus for optimal motion reproduction in stereoscopic digital cinema |
| CN102668572B (en) * | 2009-12-22 | 2015-04-29 | 汤姆逊许可证公司 | Method and apparatus for optimal motion reproduction in stereoscopic digital cinema |
| US9030525B2 (en) | 2009-12-22 | 2015-05-12 | Thomson Licensing | Method and apparatus for optimal motion reproduction in stereoscopic digital cinema |
| CN103533242A (en) * | 2013-10-15 | 2014-01-22 | 中国科学院深圳先进技术研究院 | Method and system for extracting and tracking cursor point in out-of-focus video |
| CN103533242B (en) * | 2013-10-15 | 2016-08-10 | 中国科学院深圳先进技术研究院 | The method and system with tracking cursor point are extracted in video out of focus |
| CN104898837A (en) * | 2015-05-22 | 2015-09-09 | 燕山大学 | Portable hand virtual rehabilitation experiment box based on gesture interaction and method |
| CN104887238A (en) * | 2015-06-10 | 2015-09-09 | 上海大学 | Hand rehabilitation training evaluation system and method based on motion capture |
| CN105832343A (en) * | 2016-05-22 | 2016-08-10 | 上海大学 | Multi-dimensional vision hand function rehabilitation quantitative evaluation system and evaluation method |
| CN109945594A (en) * | 2019-03-22 | 2019-06-28 | 上海五宝网络科技有限公司 | Intelligent vision refrigerator based on dynamic video detection |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111340797B (en) | Laser radar and binocular camera data fusion detection method and system | |
| CN104183127B (en) | Traffic surveillance video detection method and device | |
| CN108416321A (en) | For predicting that target object moves method, control method for vehicle and the device of direction | |
| CN102243687A (en) | Physical education teaching auxiliary system based on motion identification technology and implementation method of physical education teaching auxiliary system | |
| CN104794737B (en) | A kind of depth information Auxiliary Particle Filter tracking | |
| CN101021948A (en) | Automatic identifying device and method for joint in human body symmetric motion image | |
| CN109758756B (en) | Gymnastics video analysis method and system based on 3D camera | |
| CN101726498B (en) | Intelligent detector and method of copper strip surface quality on basis of vision bionics | |
| CN111968048B (en) | Method and system for enhancing image data of less power inspection samples | |
| CN112270310A (en) | Cross-camera pedestrian multi-target tracking method and device based on deep learning | |
| CN107238534A (en) | The method and device of on-line monitoring plate stretch performance is calculated based on image | |
| CN108961330A (en) | The long measuring method of pig body and system based on image | |
| CN108805016A (en) | A kind of head and shoulder method for detecting area and device | |
| US11948344B2 (en) | Method, system, medium, equipment and terminal for inland vessel identification and depth estimation for smart maritime | |
| CN108471497A (en) | A kind of ship target real-time detection method based on monopod video camera | |
| CN113706579A (en) | Prawn multi-target tracking system and method based on industrial culture | |
| CN114037703A (en) | Subway valve state detection method based on two-dimensional positioning and three-dimensional attitude calculation | |
| CN116152697A (en) | Three-dimensional model measuring method and related device for concrete structure cracks | |
| CN108305242A (en) | A kind of intelligent visual detection method, system and device for industrial production line | |
| CN110349209A (en) | Vibrating spear localization method based on binocular vision | |
| Kwon et al. | Recursive bayesian filtering for multiple human pose tracking from multiple cameras | |
| CN100580372C (en) | System for testing life buoy for military use on controlling pose of human body in water | |
| CN201017346Y (en) | Arthrosis automatic recognition device in human body symmetrical characteristic movement image | |
| CN104615987B (en) | A kind of the wreckage of an plane intelligent identification Method and system based on error-duration model neutral net | |
| CN116152928A (en) | Drowning prevention early warning method and system based on lightweight human body posture estimation model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |