CN109063536A - A kind of reflective stage personnel identity recognition methods - Google Patents
A kind of reflective stage personnel identity recognition methods Download PDFInfo
- Publication number
- CN109063536A CN109063536A CN201810556557.1A CN201810556557A CN109063536A CN 109063536 A CN109063536 A CN 109063536A CN 201810556557 A CN201810556557 A CN 201810556557A CN 109063536 A CN109063536 A CN 109063536A
- Authority
- CN
- China
- Prior art keywords
- stage personnel
- identity
- label
- image
- reflective
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
- G06V10/12—Details of acquisition arrangements; Constructional details thereof
- G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
- G06V10/141—Control of illumination
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of reflective stage personnel identity recognition methods, comprising steps of 1) making reflective light label 2) infrared light compensating lamp is configured by video camera, and the identity label that can reflect infrared ray is worn by stage personnel；2) after the realtime graphic for collecting stage personnel, determine that human body key point reduces sliding window range；3) identity label character is extracted in the region that primary segmentation goes out, and carries out character recognition；4) present invention is greatly reduced the difficulty of label and background separation, using human body critical point detection and its accuracy rate and efficiency is greatly improved, so as to shorten the time needed for identification using the identity label of infrared light compensating lamp and reflection infrared ray.
Description
Technical field
The present invention relates to a kind of reflective stage personnel identity recognition methods.
Background technique
At present, novel identity recognizing technology mainly has following seed type:
1. remote finger scan
Most of fingerprint scanner requires the physical contact of finger and scanner, however such contact also can usually make
At pollution and failure together.Therefore now with researcher be try to develop it is a kind of can be remote about 20 feet (being roughly equal to 6 meters)
Place carries out the instrument of finger scan.
The principle of its work is to receive different image datas using the camera of two 130 everythings elements: a level is inclined
Shake light, another vertical polarised light.When in use, light beam being had to be mapped on finger, reflected light subsequently enters lens systems,
Sharp image is synthesized into the polarized wave wherein separated.
2. ear
Everyone ear is different, by some features of ear when the fingerprint that conduct oneself like that come be subject to using.?
2010, light beam was beaten the ear in people using the technology of a kind of entitled " conversion of image radio frequency " by one group of Britain research personnel, with
After repeat this process, utilize a kind of algorithm generate human ear tubular area image.The fringe region of this ear is for everyone
It is all different, and be also the most apparent part of difference.Then, researcher converts the image into a series of number and stores
Get up, so, later when the ear that scanning is consistent to data again, system just can by scanning result in database
There is sample to be compared.
3. gait
The difficult point done so is that mankind's walking is less difficult to capture.As long as putting on a briefcase in this manpower
Or lame leg can all influence capture of the system to clear gait information.Therefore it watches the walking of a people less and can not provide
Specific biometric information.In addition to this, a people might have several different gaits, complete when walking and running
It is different.
4. face identification system
Existing major part face identification system requires to obtain good illumination, and clearly face-image can just work.
However light is also not enough to rise to the challenge in this way.Several factors can all interfere with recognition result, such as the variation of illumination condition, distance
Variation etc. can all influence capture of the equipment to image, especially when a people is among movement.
Complexity in stage environment due to light causes some of the above method that can not realize personal identification well
Effect, and the gait of stage personnel is typically all to pass through specialized training, is not in a kind of naturally state, Yi Xieyi
Method by gesture recognition will be affected.
Summary of the invention
Object of the present invention is to provide a kind of reflective stage personnel identity recognition methods in view of the defects existing in the prior art,
In view of the complexity of light on stage, we wear special reflecting formula label using stage personnel to regard characteristic point, and lead to
The mode for crossing infrared light filling carries out the identification of personnel.
The present invention to achieve the above object, adopts the following technical scheme that a kind of reflective stage personnel identity recognition methods,
The following steps are included:
1) the rectangle alphanumeric tag of 4.5CM*9CM, is fabricated to using reflecting material；
2) video camera and infrared light compensating lamp, have been set up, light compensating lamp angle and camera angle are not less than 120 degree；
3), ensure the optical axis of video camera and the optical axis keeping parallelism of infrared light compensating lamp and in adjacent locations；
4), identity label is worn near waist by stage personnel；
5) coordinate of stage personnel's artis, is extracted by video camera, and then the position of identity label slightly determine
Position；
6) the coarse positioning image of identity label is split from original image, and obtained image is filtered, two-value
The processing such as change, to obtain the exact position of identification tag；
7) tag characters, are partitioned into from image obtained after processing；
8), the tag characters being partitioned into are identified using the method for template matching, to identify stage personnel's
Identity.
Further, the identity label is made of the material that can reflect infrared ray.
Further, the concrete mode of the coarse positioning in the step 5) are as follows: extract the hip joint and knee of stage personnel
Joint coordinates, and as unit of the distance between joint, the approximate region of primary segmentation outgoing label.
Beneficial effects of the present invention: the present invention is substantially reduced using the identity label of infrared light compensating lamp and reflection infrared ray
The difficulty of label and background separation, using human body critical point detection and greatly improves its accuracy rate and efficiency, to contract
Time needed for short identification.
Detailed description of the invention
Fig. 1 is present invention reflection, scatters, the schematic diagram of reflex reflection；
Fig. 2 is identification work flow diagram of the present invention；
Fig. 3 is human body key point schematic diagram of the present invention；
Fig. 4 is human body critical point detection schematic network structure of the present invention；
Fig. 5 is that the reflective stage personnel identity of the present invention identifies the first schematic diagram；
Fig. 6 is that the reflective stage personnel identity of the present invention identifies the second schematic diagram.
Specific embodiment
Shown in Fig. 1 to Fig. 6, a kind of reflective stage personnel identity recognition methods is disclosed, including the following steps:
1), stage personnel wear the square alphanumeric tag of the 4.5CM*9CM made by reflecting material in loins.
2), start video camera and infrared light compensating lamp, adjust infrared light compensating lamp angle, the optical axis of video camera and infrared light filling
The optical axis of lamp is in the best reflecting effect to obtain the square reflective light label of 4.5CM*9CM in parallel.
3), video camera obtains the realtime graphic of stage personnel.
4) it, since the identity label of stage personnel is worn near waist, is detected by human joint points and determines waist
The position in portion reduces target labels region of search.
5), since label uses reflecting material, brightness of image of the label in video camera will be much higher than surrounding enviroment, therefore
Gamma transformation, which can first be carried out, enhances image, so that character picture becomes apparent relative to background image, thus more sharp
In the extraction of character.After finishing image enhancement, connected domain analysis is done to image, to remove noise.By setting in the present invention
Determine the range removal noise of the area of connected domain.
6), due to the angle problem of shooting, label image, in this way can be to Character segmentation result it is possible that inclination conditions
It affects greatly.Therefore, it is necessary to carry out slant correction to image.The present invention is come out lines detection using Hough transform, and
Straight line and angle both horizontally and vertically are calculated, original image rotation is then completed into slant correction.After the completion of slant correction,
Thresholding processing is carried out to image using Otsu method again, is come out character as foreground extraction.After being partitioned into character picture,
Further micronization processes are done to character picture, are handled convenient for subsequent character recognition.
7) it is normalized after, being corrected to tag characters, character recognition is carried out, to differentiate to identity.
The present invention is directed to the feature of light of stage complexity, eliminates complicated light using the principle of infrared light filling and reflex reflection
Interference.Reduce the candidate region of sliding window using the method for first detecting human body key point when sliding window detects square tags.
Human body critical point detection network structure is as shown in Figure 4.One group of prediction is two-dimensional simultaneously using a feedforward network for the network structure
Human joint points position S and one group of affine domain L in two dimension part, next refines continuous stage forecast, in each rank
Section has intermediate supervision.
Network structure is specially first to be analyzed (as at the beginning of first 10 layers of VGG-19 as convolutional neural networks input picture
Beginningization is simultaneously finely adjusted), a Feature Mapping F is obtained, and be inputted each branch of first stage.Net in the first stage
Network generates the confidence map S of detection artis1With the affine vector field L in part1, each small forecast period, Liang Ge branch exist later
The prediction in previous stage is used to the accurate prediction in the stage together with original feature F together.
A loss L2 function is added in preferably prediction in each branch at the end in each stage, and spatially
Loss function is weighted.Artis confidence map and associate field are solved finally by greedy algorithm, to export institute in detection image
The 2D key point of someone.
The foregoing is merely presently preferred embodiments of the present invention, is not intended to limit the invention, it is all in spirit of the invention and
Within principle, any modification, equivalent replacement, improvement and so on be should all be included in the protection scope of the present invention.
Claims (3)
1. a kind of reflective stage personnel identity recognition methods, which comprises the following steps:
1) the rectangle alphanumeric tag of 4.5CM*9CM, is fabricated to using reflecting material；
2) video camera and infrared light compensating lamp, have been set up, light compensating lamp angle and camera angle are not less than 120 degree；
3), ensure the optical axis of video camera and the optical axis keeping parallelism of infrared light compensating lamp and in adjacent locations；
4), identity label is worn near waist by stage personnel；
5) coordinate of stage personnel's artis, is extracted by video camera, and then coarse positioning is carried out to the position of identity label；
6) the coarse positioning image of identity label is split from original image, and obtained image is filtered, at binaryzation
Reason, to obtain the exact position of identification tag；
7) tag characters, are partitioned into from image obtained after processing；
8), the tag characters being partitioned into are identified using the method for template matching, to identify the identity of stage personnel.
2. the reflective stage personnel identity recognition positioning method as described in claim 1 based on binocular positioning, feature exist
In the identity label is made of the material that can reflect infrared ray.
3. the reflective stage personnel identity recognition positioning method as described in claim 1 based on binocular positioning, feature exist
In the concrete mode of the coarse positioning in the step 5) are as follows: extract left and right two hip joint coordinates of stage personnel, and with
The distance between two hip joints are unit, the approximate region of primary segmentation outgoing label.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810556557.1A CN109063536A (en) | 2018-06-01 | 2018-06-01 | A kind of reflective stage personnel identity recognition methods |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810556557.1A CN109063536A (en) | 2018-06-01 | 2018-06-01 | A kind of reflective stage personnel identity recognition methods |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109063536A true CN109063536A (en) | 2018-12-21 |
Family
ID=64819884
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810556557.1A Pending CN109063536A (en) | 2018-06-01 | 2018-06-01 | A kind of reflective stage personnel identity recognition methods |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109063536A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112032686A (en) * | 2020-08-05 | 2020-12-04 | 浙江大丰实业股份有限公司 | Brightness adjusting system and method for stage performance lamplight |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2078287A2 (en) * | 2006-11-01 | 2009-07-15 | Sony Corporation | Capturing surface in motion picture |
| CN201454059U (en) * | 2009-06-26 | 2010-05-12 | 上海久娱网络科技有限公司 | Character identity identification system in card game console |
| CN101999127A (en) * | 2008-02-19 | 2011-03-30 | 比尔凯科技新加坡有限公司 | A reading device for identifying a tag or an object adapted to be identified, related methods and systems |
| CN102024157A (en) * | 2009-09-09 | 2011-04-20 | 索尼公司 | Information processing apparatus, information processing method,and information processing program |
| CN106156770A (en) * | 2015-04-09 | 2016-11-23 | 宋怀淳 | A kind of theater audiences demographic method |
| CN107609108A (en) * | 2017-09-13 | 2018-01-19 | 杭州景联文科技有限公司 | A kind of sportsman's photo method for sorting based on number slip identification and recognition of face |
| CN107708273A (en) * | 2017-10-19 | 2018-02-16 | 北京工业大学 | A kind of stage light control system, polishing lamp lamp holder driving method, tag location method |
| CN108009468A (en) * | 2017-10-23 | 2018-05-08 | 广东数相智能科技有限公司 | A kind of marathon race anti-cheat method, electronic equipment and storage medium |
-
2018
- 2018-06-01 CN CN201810556557.1A patent/CN109063536A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2078287A2 (en) * | 2006-11-01 | 2009-07-15 | Sony Corporation | Capturing surface in motion picture |
| CN101999127A (en) * | 2008-02-19 | 2011-03-30 | 比尔凯科技新加坡有限公司 | A reading device for identifying a tag or an object adapted to be identified, related methods and systems |
| CN201454059U (en) * | 2009-06-26 | 2010-05-12 | 上海久娱网络科技有限公司 | Character identity identification system in card game console |
| CN102024157A (en) * | 2009-09-09 | 2011-04-20 | 索尼公司 | Information processing apparatus, information processing method,and information processing program |
| CN106156770A (en) * | 2015-04-09 | 2016-11-23 | 宋怀淳 | A kind of theater audiences demographic method |
| CN107609108A (en) * | 2017-09-13 | 2018-01-19 | 杭州景联文科技有限公司 | A kind of sportsman's photo method for sorting based on number slip identification and recognition of face |
| CN107708273A (en) * | 2017-10-19 | 2018-02-16 | 北京工业大学 | A kind of stage light control system, polishing lamp lamp holder driving method, tag location method |
| CN108009468A (en) * | 2017-10-23 | 2018-05-08 | 广东数相智能科技有限公司 | A kind of marathon race anti-cheat method, electronic equipment and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| 林强，田双良: "《行为识别与智能计算》", 30 November 2016 * |
| 赵丽科等: ""田径运动员号码牌图像的号码识别"", 《华东师范大学学报（ 自 然科学版 ）》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112032686A (en) * | 2020-08-05 | 2020-12-04 | 浙江大丰实业股份有限公司 | Brightness adjusting system and method for stage performance lamplight |
| CN112032686B (en) * | 2020-08-05 | 2022-11-04 | 浙江大丰实业股份有限公司 | Brightness adjusting system and method for stage performance lamplight |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105373135B (en) | A kind of method and system of aircraft docking guidance and plane type recognition based on machine vision | |
| CN102214291B (en) | Method for quickly and accurately detecting and tracking human face based on video sequence | |
| CN101246544B (en) | Iris positioning method based on boundary point search and minimum kernel value similarity region edge detection | |
| CN110008909B (en) | Real-name system business real-time auditing system based on AI | |
| CN102822865A (en) | Face recognition device and face recognition method | |
| KR20050025927A (en) | The pupil detection method and shape descriptor extraction method for a iris recognition, iris feature extraction apparatus and method, and iris recognition system and method using its | |
| CN103729614A (en) | People recognition method and device based on video images | |
| CN102289672A (en) | Infrared gait identification method adopting double-channel feature fusion | |
| CN111027432B (en) | Gait feature-based visual following robot method | |
| US11651624B2 (en) | Iris authentication device, iris authentication method, and recording medium | |
| CN107038400A (en) | Face identification device and method and utilize its target person tracks of device and method | |
| Chen et al. | Eyes localization algorithm based on prior MTCNN face detection | |
| CN109063536A (en) | A kind of reflective stage personnel identity recognition methods | |
| CN113920591A (en) | Medium-distance and long-distance identity authentication method and device based on multi-mode biological feature recognition | |
| CN107145820B (en) | Binocular positioning method based on HOG characteristics and FAST algorithm | |
| Hilario et al. | Pedestrian detection for intelligent vehicles based on active contour models and stereo vision | |
| CN109543582A (en) | Human body foreign body detection method based on millimeter-wave image | |
| CN112633222A (en) | Gait recognition method, device, equipment and medium based on confrontation network | |
| KR100397750B1 (en) | Realtime pupil detecting method for iris recognition | |
| Zaim et al. | A robust and accurate segmentation of iris images using optimal partitioning | |
| Cao et al. | Localizing the iris center by region growing search | |
| Sheshadri et al. | Kinect based Frontal Gait Recognition using skeleton and depth derived features | |
| Park | Face Recognition: face in video, age invariance, and facial marks | |
| CN114373091A (en) | Gait recognition method based on deep learning fusion SVM | |
| Ishikawa et al. | Pupil Center Detection by Abstracted Contour Graph Analysis for Iris Detection |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181221 |
|
| RJ01 | Rejection of invention patent application after publication |