CN110298814A - A kind of stage Adjustable head lamp sports coordination analysis method based on intelligent depth vision - Google Patents
A kind of stage Adjustable head lamp sports coordination analysis method based on intelligent depth vision Download PDFInfo
- Publication number
- CN110298814A CN110298814A CN201810242996.5A CN201810242996A CN110298814A CN 110298814 A CN110298814 A CN 110298814A CN 201810242996 A CN201810242996 A CN 201810242996A CN 110298814 A CN110298814 A CN 110298814A
- Authority
- CN
- China
- Prior art keywords
- depth
- adjustable head
- drift angle
- head lamp
- lamp
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10028—Range image; Depth image; 3D point clouds
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The stage Adjustable head lamp sports coordination analysis method based on intelligent depth vision that the invention discloses a kind of, hardware device includes: one for capturing the depth camera and a computer of range information of each point to depth transducer plane on equipment under test in the method；The method are as follows: the depth image comprising each equipment under test of depth camera input finally, obtains movement drift angle of the lamp cap relative to pedestal by target detection deep neural network, image segmentation, plane monitoring-network and feature extraction algorithm.The present invention can track and detect simultaneously more stage Adjustable head lamps by depth camera combination computer vision technique, test and record its kinematic parameter, generate accurately and reliably examining report；And performance test and comparison of the support to multiple batches of product.
Description
Technical field
The present invention relates to stage lighting detection technique field, specially a kind of stage Adjustable head lamp fortune based on intelligent depth vision
Dynamic Coordination Analysis method.
Background technique
Motion synchronicity can be important one of the technical indicator of stage Adjustable head lamp.For same style, with a batch of dance
Platform Adjustable head lamp needs to complete completely the same movement in the case where executing identical multiple groups movement instruction.And it is researched and developed in equipment
In assembly, the problems such as due to algorithm, component, assembly method, easily causes lamp cap and move not quite identical problem.It is this
Movement inconsistence problems can directly result in lamps and lanterns and the problems such as light is uncoordinated occur at performance scene.In the process researched and developed and assembled
In, the movement for making the lamp cap of each stage Adjustable head lamp required as far as possible is consistent.It is therefore desirable to have suitable movement synchronizes
Property detection and analysis method comes auxiliary development and assembly.
Currently, mainly detecting it by the encoder being mounted on each kinematic axis in research and development and experimentation
Kinematic parameter, including position, velocity and acceleration data.This method needs add new detection part, nothing inside equipment
Test assessment of the method suitable for a large amount of assembly.
It in assembly, is presently mainly visually inspected, can not be accurately checked with a batch of net synchronization capability by assembly crewman,
It is even more impossible to detect the net synchronization capability of different assembly batches.
The present invention can be tracked and be detected simultaneously more stages by depth camera combination computer vision technique and be shaken the head
Its kinematic parameter is tested and recorded to lamp, generates accurately and reliably examining report.And support to the performance test of multiple batches of product with
Comparison.
Summary of the invention
The purpose of the present invention is to provide a kind of stage Adjustable head lamp sports coordination analysis side based on intelligent depth vision
Method, solve existing method for testing performance it is inconvenient, inaccurate the problems such as.With at low cost, precision is high, non-contact, lossless
The features such as material；Be conducive to improve its design and producing precision in the research and development and production of stage Adjustable head lamp.
To achieve the above object, the invention provides the following technical scheme:
A kind of stage Adjustable head lamp sports coordination analysis method based on intelligent depth vision, hardware device in the method
It include: one for capturing the depth camera and one of range information of each point to depth transducer plane on equipment under test
Platform computer；The method are as follows: the depth image comprising each equipment under test of depth camera input passes through target detection depth
Neural network, image segmentation, plane monitoring-network and feature extraction algorithm are spent, finally, it is inclined relative to the movement of pedestal to obtain lamp cap
Angle.
As a further solution of the present invention: specifically comprising the following steps:
1) data acquire: acquiring more tested stage Adjustable head lamps by depth camera, obtain the reception of stage Adjustable head lamp
Video when movement after control instruction；
2) position detection: deep neural network is examined by target to detect every stage and shake the head the position of lamp apparatus, and right
Each stage Adjustable head lamp equipment is individually created a depth map；
3) pedestal drift angle identify: to each stage shake the head lamp apparatus depth map carry out image segmentation, believed by depth
Breath finds base front surface, then, plane monitoring-network is carried out to it, and the drift angle of pedestal is calculated by plane equation；
4) lamp cap drift angle identify: to each stage shake the head lamp apparatus depth map carry out image segmentation, believed by depth
Breath finds the position of lamp cap, and does plane monitoring-network to entire lamp cap, then does feature extraction to each plane detected, finds
For calculating the crucial plane of drift angle, and go out by screen equation calculation the drift angle of lamp cap；
5) it obtains lamp cap drift angle: being subtracted each other by the drift angle of pedestal in the drift angle of lamp cap in step 3) and step 4), obtain lamp
Drift angle of the head relative to pedestal, analyzes the kinematic parameter of stage Adjustable head lamp, facilitates comparison and storage.
As a further solution of the present invention: the depth camera uses the high frame per second depth camera of high-resolution.
As a further solution of the present invention: the target detection deep neural network using Fast-RCNN neural network,
Faster-RCNN neural network or YOLO neural network.
As a further solution of the present invention: image partition method is used in the step 3) and step 4) is rolled up based on depth
The depth image segmentation method of product network, or use the depth image segmentation method based on traditional computer vision technique.
As a further solution of the present invention: plane monitoring-network method, which uses, in the step 3) and step 4) is based on depth map
Plane monitoring-network method, plane equation indicate are as follows: ax+by+cz+d=0 can calculate folder by simple geometric knowledge
Angle.
Compared with prior art, the beneficial effects of the present invention are:
(1) the advanced computer vision technique based on depth map has been used；
(2) non-contact detection does not need installation and removal detection device；
(3) what the movenent performance parameter analyzed can be convenient compares with other equipment；
(4) single detection device can more lamp apparatus of shaking the head of detection and analysis simultaneously.
Detailed description of the invention
Fig. 1 is the schematic diagram of the method for the present invention usage scenario；
Fig. 2 is the corresponding system flow chart of the method for the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
Fig. 1~2 are please referred to, the present invention provides a kind of technical solution: a kind of stage Adjustable head lamp based on intelligent depth vision
Sports coordination analysis method, hardware device includes: one for capturing on equipment under test each point to depth in the method
The depth camera of the range information of sensor plane and a computer；The method are as follows: the packet of depth camera input
Depth image containing each equipment under test is mentioned by target detection deep neural network, image segmentation, plane monitoring-network and feature
Algorithm is taken, finally, obtains movement drift angle of the lamp cap relative to pedestal.
Specific implementation includes the following steps:
1) data acquire: acquiring more tested stage Adjustable head lamps by depth camera, obtain the reception of stage Adjustable head lamp
Video when movement after control instruction；
2) position detection: deep neural network is examined by target to detect every stage and shake the head the position of lamp apparatus, and right
Each stage Adjustable head lamp equipment is individually created a depth map；
3) pedestal drift angle identify: to each stage shake the head lamp apparatus depth map carry out image segmentation, believed by depth
Breath finds base front surface, then, plane monitoring-network is carried out to it, and the drift angle of pedestal is calculated by plane equation；
4) lamp cap drift angle identify: to each stage shake the head lamp apparatus depth map carry out image segmentation, believed by depth
Breath finds the position of lamp cap, and does plane monitoring-network to entire lamp cap, then does feature extraction to each plane detected, finds
For calculating the crucial plane of drift angle, and go out by screen equation calculation the drift angle of lamp cap；
5) it obtains lamp cap drift angle: being subtracted each other by the drift angle of pedestal in the drift angle of lamp cap in step 3) and step 4), obtain lamp
Drift angle of the head relative to pedestal, analyzes the kinematic parameter of stage Adjustable head lamp, facilitates comparison and storage.
In the method step 1), the depth camera uses the high frame per second depth camera of high-resolution.Depth camera
The accuracy and resolution ratio of head sensor influence whether that final precision, higher precision and resolution ratio can obtain higher analysis
Precision.Suitable camera can be selected according to the actual situation.
In the method step 2), the type for the target detection neural network that can be used has: Fast-RCNN network (ginseng
It examines: Girshick R.Fast R-CNN [C] //Computer Vision (ICCV), 2015IEEE International
Conference on.IEEE, 2015:1440-1448.), (reference: Ren S, He K, Girshick of Faster-RCNN network
R,et al.Faster R-CNN:towards real-time object detection with region proposal
networks[J].IEEE transactions on pattern analysis and machine intelligence,
2017,39 (6): 1137-1149.), (reference: Redmon J, Divvala S, Girshick R, et al.You of YOLO network
only look once:Unified,real-time object detection[C]//Proceedings of the IEEE
Conference on computer vision and pattern recognition.2016:779-788.) etc. nerve nets
Network method.
In the method step 2), since input data is depth map, the model to texture maps training can not be directly used,
Need to resurvey and mark training sample.
In the method step 3) and step 4), the depth map segmentation technology based on depth convolutional network, example can be used
Such as document: Long J, Shelhamer E, Darrell T.Fully convolutional networks for semantic
segmentation[C]//Proceedings of the IEEE conference on computer vision and
pattern recognition.2015:3431-3440.Also the depth map based on traditional computer vision technique point can be used
Cut technology, such as document: Silberman N, Fergus R.Indoor scene segmentation using a
structured light sensor[C]//Computer Vision Workshops(ICCV Workshops),
2011IEEE International Conference on.IEEE,2011:601-608.
In the method step 3) and step 4), using the plane monitoring-network method based on depth map, such as document: Jin Z,
Tillo T,Cheng F.Planar surfaces detection on depth map using patch based
approach[C]//Consumer Electronics(GCCE),2014IEEE3rd Global Conference
on.IEEE,2014:227-229.
In the method step 3) and step 4), plane equation can be expressed as ax+by+cz+d=0, by simple several
What knowledge, can calculate angle.
By this technology, the parameter that may finally be tested and analyze has: the instruction execution delay time of Adjustable head lamp lamp cap,
Lamp cap curve movement, lamp cap movement velocity curve, lamp cap acceleration of motion curve etc..Exploitation and test for lamp performance are all
Play a significant role.
It is obvious to a person skilled in the art that invention is not limited to the details of the above exemplary embodiments, Er Qie
In the case where without departing substantially from spirit or essential attributes of the invention, the present invention can be realized in other specific forms.Therefore, no matter
From the point of view of which point, the present embodiments are to be considered as illustrative and not restrictive, and the scope of the present invention is by appended power
Benefit requires rather than above description limits, it is intended that all by what is fallen within the meaning and scope of the equivalent elements of the claims
Variation is included within the present invention.Any reference signs in the claims should not be construed as limiting the involved claims.
Claims (6)
1. a kind of stage Adjustable head lamp sports coordination analysis method based on intelligent depth vision, it is characterised in that: the method
Middle hardware device includes: one and takes the photograph for capturing the depth of range information of each point to depth transducer plane on equipment under test
Camera and a computer；The method are as follows: the depth image comprising each equipment under test of depth camera input passes through
Target detection deep neural network, image segmentation, plane monitoring-network and feature extraction algorithm finally obtain lamp cap the bottom of relative to
The movement drift angle of seat.
2. a kind of stage Adjustable head lamp sports coordination analysis method based on intelligent depth vision according to claim 1,
It is characterized by: specifically comprising the following steps:
1) data acquire: acquiring more tested stage Adjustable head lamps by depth camera, obtain stage Adjustable head lamp and receive control
Video when movement after instruction；
2) position detection: deep neural network is examined by target to detect every stage and shake the head the position of lamp apparatus, and to each
Platform stage Adjustable head lamp equipment is individually created a depth map；
3) pedestal drift angle identify: to each stage shake the head lamp apparatus depth map carry out image segmentation, looked for by depth information
To base front surface, then, plane monitoring-network is carried out to it, and the drift angle of pedestal is calculated by plane equation；
4) lamp cap drift angle identify: to each stage shake the head lamp apparatus depth map carry out image segmentation, looked for by depth information
Plane monitoring-network is done to the position of lamp cap, and to entire lamp cap, feature extraction then is done to each plane detected, finds and is used for
The crucial plane of drift angle is calculated, and goes out the drift angle of lamp cap by screen equation calculation；
5) it obtains lamp cap drift angle: being subtracted each other by the drift angle of pedestal in the drift angle of lamp cap in step 3) and step 4), obtain lamp cap phase
For the drift angle of pedestal, the kinematic parameter of stage Adjustable head lamp is analyzed, facilitates comparison and storage.
3. a kind of stage Adjustable head lamp sports coordination analysis method based on intelligent depth vision according to claim 2,
It is characterized by: the depth camera uses the high frame per second depth camera of high-resolution.
4. a kind of stage Adjustable head lamp sports coordination analysis method based on intelligent depth vision according to claim 2,
It is characterized by: target detection deep neural network uses Fast-RCNN neural network, Faster-RCNN in the step 2)
Neural network or YOLO neural network.
5. a kind of stage Adjustable head lamp sports coordination analysis method based on intelligent depth vision according to claim 2,
It is characterized by: image partition method uses the depth map segmentation side based on depth convolutional network in the step 3) and step 4)
Method, or use the depth image segmentation method based on traditional computer vision technique.
6. a kind of stage Adjustable head lamp sports coordination analysis method based on intelligent depth vision according to claim 2,
It is characterized by: plane monitoring-network method uses the plane monitoring-network method based on depth map, plane in the step 3) and step 4)
Equation indicates are as follows: ax+by+cz+d=0 can calculate angle by simple geometric knowledge.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810242996.5A CN110298814A (en) | 2018-03-23 | 2018-03-23 | A kind of stage Adjustable head lamp sports coordination analysis method based on intelligent depth vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810242996.5A CN110298814A (en) | 2018-03-23 | 2018-03-23 | A kind of stage Adjustable head lamp sports coordination analysis method based on intelligent depth vision |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110298814A true CN110298814A (en) | 2019-10-01 |
Family
ID=68025807
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810242996.5A Pending CN110298814A (en) | 2018-03-23 | 2018-03-23 | A kind of stage Adjustable head lamp sports coordination analysis method based on intelligent depth vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110298814A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104236866A (en) * | 2014-09-01 | 2014-12-24 | 南京林业大学 | Automobile headlamp test data error correction method based on driving direction |
| US20170068237A1 (en) * | 2015-09-04 | 2017-03-09 | Minebea Co., Ltd. | Device control system |
| CN107567169A (en) * | 2017-09-30 | 2018-01-09 | 广州市浩洋电子股份有限公司 | Automatic stage lighting tracking system and control method thereof |
| CN107563333A (en) * | 2017-09-05 | 2018-01-09 | 广州大学 | A kind of binocular vision gesture identification method and device based on ranging auxiliary |
-
2018
- 2018-03-23 CN CN201810242996.5A patent/CN110298814A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104236866A (en) * | 2014-09-01 | 2014-12-24 | 南京林业大学 | Automobile headlamp test data error correction method based on driving direction |
| US20170068237A1 (en) * | 2015-09-04 | 2017-03-09 | Minebea Co., Ltd. | Device control system |
| CN107563333A (en) * | 2017-09-05 | 2018-01-09 | 广州大学 | A kind of binocular vision gesture identification method and device based on ranging auxiliary |
| CN107567169A (en) * | 2017-09-30 | 2018-01-09 | 广州市浩洋电子股份有限公司 | Automatic stage lighting tracking system and control method thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109887020B (en) | Plant organ separation method and system | |
| Barekatain et al. | Okutama-action: An aerial view video dataset for concurrent human action detection | |
| Marshall et al. | Modelling the Galactic interstellar extinction distribution in three dimensions | |
| Gourgari et al. | Thetis: Three dimensional tennis shots a human action dataset | |
| CN105701798B (en) | The data reduction method and device of columnar object | |
| CN109190508A (en) | A kind of multi-cam data fusion method based on space coordinates | |
| Bo et al. | Particle pollution estimation from images using convolutional neural network and weather features | |
| CN109636927B (en) | System and method for training and identifying aircraft attitude measurement algorithm | |
| CN103105924B (en) | Man-machine interaction method and device | |
| CN107247926B (en) | A kind of human body detecting method and device | |
| CN109492618A (en) | Object detection method and device based on grouping expansion convolutional neural networks model | |
| CN111209882A (en) | Gait information acquisition method and system of omnidirectional movement device and readable storage medium | |
| CN103438834A (en) | Hierarchy-type rapid three-dimensional measuring device and method based on structured light projection | |
| CN104680188A (en) | Method for constructing human body posture reference image library | |
| CN103759724B (en) | A kind of indoor navigation method based on lamp decoration feature and system | |
| CN109840905A (en) | Power equipment rusty stain detection method and system | |
| CN110290311B (en) | Method, device and system for generating jitter track for evaluating anti-jitter performance of video | |
| CN105631849B (en) | The change detecting method and device of target polygon | |
| CN108873363A (en) | 3 d light fields imaging system and method based on structure signal | |
| CN110298814A (en) | A kind of stage Adjustable head lamp sports coordination analysis method based on intelligent depth vision | |
| CN113076781A (en) | Method, device and equipment for detecting falling and storage medium | |
| CN110298880A (en) | A kind of stage Adjustable head lamp sports coordination analysis method based on intelligent monocular vision | |
| Enuş et al. | Three-dimensional measurements of tree crown movement using an infrared time-of-flight camera | |
| CN105203045A (en) | System and method for detecting product shape integrity based on asynchronous time domain vision sensor | |
| WO2018032628A1 (en) | Teaching toy kit and colored card identification method thereof |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20191001 |