CN102175172B - Image-identification-based system for detecting contour dimension of motor vehicle - Google Patents
Image-identification-based system for detecting contour dimension of motor vehicle Download PDFInfo
- Publication number
- CN102175172B CN102175172B CN2010106226573A CN201010622657A CN102175172B CN 102175172 B CN102175172 B CN 102175172B CN 2010106226573 A CN2010106226573 A CN 2010106226573A CN 201010622657 A CN201010622657 A CN 201010622657A CN 102175172 B CN102175172 B CN 102175172B
- Authority
- CN
- China
- Prior art keywords
- digital camera
- black
- motor vehicles
- portal frame
- image
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention relates to a system for detecting the contour dimension of a motor vehicle, in particular to an image-identification-based system for detecting the contour dimension of a motor vehicle. The system comprises a digital camera set, a front-end controller, a switcher and a computer, wherein the digital camera set is connected with the front-end controller, and the front-end controller is connected with the computer through the switcher. The system is characterized by also comprising a black and white barcode grade rod, a double portal frame, single portal frames and a double-face frame with two rectangular portal frames The black and white barcode grade rod is arranged on the outer side of the rectangular double-face frame, the single portable frames are respectively arranged at the position 2-5 meters in right ahead of each portal frame face at the outer side of the rectangular double-face frame, wherein each single portable frame is a plan-shaped single-face frame, the digital camera set is arranged right above the plan-shaped single-face frames, and the photographing direction of the digital camera set is symmetrical with the direction of the black and white barcode grade rod arranged on the double-face frame with two portal frames. By the system, contour dimension and other parameters of various motor vehicles can be measured quickly and efficiently.
Description
One, technical field
The present invention relates to a kind of motor vehicles physical dimension detection system, particularly a kind of motor vehicles physical dimension detection system based on image recognition.
Two, background technology
Along with developing of automobile industry, Motor Vehicles Detection Technique is also improving constantly.But at present China vehicle management department detects for motor vehicles physical dimension and mainly still relies on manual detection, promptly through the artificial visually examine or utilize survey instrument to measure.Not enough below this metering system exists: 1. the testing result error is bigger, and precision can't guarantee; 2. detection time longer, inefficiency, and take more human and material resources; 3. automaticity is low, can not satisfy the development need of motor vehicle management department to the management of motor vehicles implementation information.
" based on the vehicle overall size detecting system of Flame Image Process " that the patented claim CN03114546.9 of Northwestern Polytechnical University proposes; This technical scheme is to obtain two width of cloth contour of the vehicle images through image acquisition units; Get into storer; Digital processing unit reads in two width of cloth contour of the vehicle view data successively from storer, height, width and the length data that can obtain motor vehicles through the graphical analysis to the motor vehicles profile are the physical dimension of motor vehicles.Though this technical scheme can access motor vehicles physical dimension, the same deficiency that exists: 1. owing to receive the optimal viewing angle of video camera and the restriction of coverage, the motor vehicles outside drawing of picked-up looks like to exist error; 2. look like to carry out the parameter correction because of lacking clear and definite reference substance in this technical scheme, make testing result have error; 3. can't measure this important formal parameter of wheelspan of motor vehicles.
" obtaining the full-automatic device of image measurement vehicle contour dimension according to vehicle in the motion " that " obtaining the Automatic Measuring Apparatus of vehicle image metering vehicle contour dimension in the motion " that the sincere car insurance checkout equipment of Chengdu will company limited patented claim 200920079045.7 proposes and patented claim 200820140878.5 propose; Though these two kinds of technical schemes can both realize the measurement to motor vehicles contour dimension; But the obvious deficiency that exists is: the vehicle cross-sectional image that 1. ought obtain exists distortion or distortion then can't realize seamless spliced between the image, thereby can not accurately calculate vehicle contour dimension; 2. because surface mark and facade sign are indeterminate, complicated to the processing of vehicle cross-sectional image, the analytic process of obtaining, thus the difficulty that detects increased; 3. surface mark and facade sign and tested vehicle section when calculating vehicle contour dimension, also need solve depth of field problem not in same plane, have enlarged the error that detects data result.
Present technical scheme has further proposed to utilize the vehicle overall size detecting system based on Flame Image Process of shifting sledge picture pick-up device realization.This technical scheme proposes on shifting sledge, to be provided with the industrial digital video camera; Utilize slide rail to drive industrial camera the continuous high speed candid photograph is carried out in the front of motor vehicles and the car image of side; Then the image that obtains is carried out a series of relevant treatment, thereby obtain motor vehicles physical dimension.Though more aforementioned each technical scheme of this technical scheme makes progress; But still exist obviously not enough: have tangible mechanical motion error when 1. utilizing shifting sledge to drive industrial digital camera acquisition image; And this error is inevitable; If will reduce error must be prerequisite to increase equipment cost, the existing difficulty of historical facts or anecdotes; 2. since industrial digital video camera and shifting sledge repeat move, the picture steadiness that causes obtaining is not high, so image is carried out having error when a frame one frame splices, directly influences the precision of testing result; 3. the maintaining work of shifting sledge and industrial digital video camera is complicated, and is very high to equipment operator's requirement; 4. the cost of big, the shifting sledge of hold facility site areas such as length, shifting sledge and industrial digital video camera is high and repeat mobile this service life of equipment shortening etc. that causes time of needing of testing process.
The deficiency that how to overcome above prior art has become the great difficult problem that current motor vehicles physical dimension detection range needs to be resolved hurrily.
Three, summary of the invention
The objective of the invention is for the deficiency that overcomes prior art provides a kind of motor vehicles physical dimension detection system based on image recognition, this system can realize measuring various motor vehicles physical dimension are isoparametric fast, efficient.
A kind of motor vehicles physical dimension detection system that proposes according to the present invention based on image recognition; Comprise digital camera group, front controller, switch, computing machine; The digital camera group is connected with front controller, and front controller is connected with computing machine through switch, it is characterized in that also comprising black and white bar yardstick, two portal frame, single portal frame; Two portal frames are two-door framework of square; Two-door framework lateral surface of its square is provided with black and white bar yardstick, and 2～5 meters, dead ahead of each shop front are provided with single portal frame respectively in two-door the framework outside of its square, and single portal frame is a planar shaped simple gate face framework; Above its planar shaped simple gate face framework, be provided with the digital camera group, the black and white bar yardstick orientation that is provided with on the photograph orientation of digital camera group and the two portal frame shop front frameworks is symmetrical.
Principle of work of the present invention is: at first tested motor vehicles are rested in the inboard tested position of two portal frame shop front frameworks; Black and white bar yardstick that is provided with on the two portal frame shop front frameworks of assurance and motor vehicles tested surface are basically at grade; Front controller sends the instruction of startup work to the digital camera group then; Each group digital camera will be taken the topography that obtains and feed back to computing machine through front controller, switch, the motor vehicles part image that obtains spliced, cuts apart, will analyze and calculates motor vehicles length, width, height and wheelbase realizing.According to topology layout of the present invention; In camera motor vehicle image, also taken image as the black and white bar yardstick of reference field; And the size of each striped is known on the black and white bar yardstick; As long as guarantee black and white bar yardstick and motor vehicles tested surface basically at grade; Just can utilize the proportionate relationship of bar yardstick and motor vehicles image to obtain the physical dimension of motor vehicles, in the process of image mosaic, can utilize black and white bar yardstick that the motor vehicles image that obtains is carried out non-linear correction simultaneously.
The present invention is its remarkable advantage compared with prior art: the first, and the present invention does not need complicated rail set, and is easy to install; The occupied ground area is little, and plant factor is high, long service life; Equipment cost is low, and corrective maintenance of the present invention simultaneously is simple and convenient with maintenance, and its operating cost is very low.The second, the present invention is arranged on the tested surface of motor vehicles on the same plane of black and white bar yardstick of reference field, has simplified the complexity that the tested surface depth of field is handled, and has saved the plenty of time, has improved detection speed.The 3rd; The present invention utilizes the edge of the bar code block that obtains after the black and white bar yardstick rim detection to come the position of positioning strip yardstick in image; Utilize bar yardstick and motor vehicles image scaled relation after the rim detection can obtain the physical dimension of motor vehicles simultaneously; Both improve accuracy of detection, improved the detection operational efficiency again.The 4th, the present invention is high through the degree of stability of the vehicle topography that fixing digital camera group is obtained, and in the process of image mosaic, can utilize black and white bar yardstick that the image that obtains is carried out non-linear correction simultaneously, has improved accuracy of detection.The present invention is adaptable across measuring various motor vehicles physical dimension are isoparametric fast, efficient.
Four, description of drawings
Fig. 1 is the structural representation of a kind of motor vehicles physical dimension detection system based on image recognition of proposing of the present invention.
Fig. 2 is the structural representation of a kind of motor vehicles physical dimension detection system based on image recognition that is provided with tested motor vehicles that proposes of the present invention.
Five, embodiment
Below in conjunction with accompanying drawing and embodiment the present invention is made further detailed description.
In conjunction with accompanying drawing 1, embodiment of the present invention is following: the digital camera group (2) that the present invention mainly is made up of digital camera (1), black and white bar yardstick (3), two portal frame (4), single portal frame (5), front controller (6), switch (7) and computing machine (8) are formed.Two portal frames (4) are two-door framework of square; Its specification can be confirmed according to the factors such as physical dimension of tested motor vehicles; Generally be of a size of with the shop front of selecting the positive shop front: highly * length=5 * 5m, the shop front of side door face is of a size of: highly * length=5 * 18m is good; Two portal frames (4) lateral surface framework is provided with the black and white bar yardstick (3) that equates with its size; In the black and white bar yardstick (3) every black or informal voucher line size equates; The big I of the area of its black streaking or informal voucher line waits to confirm according to the physical dimension and the measuring accuracy requirement of tested motor vehicles; Be that 30cm is good to select the every length of side generally, its manufacturing materials is not limited, but black or the informal voucher line is clear and legible; Position with two portal frames (4) is a benchmark; 2～5 meters at two two-door the frameworks of portal frame (4) square of distance axially are provided with single portal frame (5) respectively; Single portal frame (5) is a planar shaped simple gate face framework; Its specification can be confirmed with the factors such as specification of the digital camera (1) that is adopted according to the physical dimension of tested motor vehicles; Generally be of a size of with the shop front of selecting the positive corresponding single portal frame of the shop front (5) of two portal frames (4): highly * length=2 * 5m, the shop front of the corresponding single portal frame of two portal frames (4) side door face (5) is of a size of: highly * length=2 * 18m; Top at each single portal frame (5) shop front framework is provided with digital camera group (2); The quantity of the digital camera (1) in the digital camera group (2) can be confirmed according to the factors such as physical dimension size of tested motor vehicles; The digital camera group (2) of camera motor vehicle width face figure generally becomes to be advisable with 2 digital cameras (1) parallel-group; The digital camera group (2) of camera motor vehicle length face figure is generally with 2～6 digital cameras (1) parallel-group to become to be advisable; Black and white bar yardstick (3) orientation that is provided with on the photograph orientation of digital camera group (2) and two-door portal frame (4) shop front framework is symmetrical; Digital camera group (2) links to each other with front controller (6) through data line, and front controller (6) links to each other with computing machine (8) through switch (7).
In conjunction with accompanying drawing 2, concrete detection step of the present invention is following: the first step, prepare.Tested motor vehicles (9) are rested in the inboard tested position of two portal frame (4) shop front frameworks, and black and white bar yardstick (3) that is provided with on two portal frame (4) shop front frameworks of assurance and motor vehicles (9) tested surface are basically in one plane; In second step, contour of the vehicle information is obtained.Send the instruction of startup work to digital camera group (2) by front controller (6); Two digital camera set (2) will be taken the topography that obtains and fed back to computing machine (8) through front controller (6) and switch (7), with the realization IMAQ with and be that benchmark becomes complete motor vehicles (9) front view (FV) and outboard profile with two width of cloth image mosaics with the bar yardstick; The 3rd step, the contour of the vehicle parameter extraction.From the front of the complete motor vehicles that obtain and outboard profile, be partitioned into motor vehicles and black and white bar yardstick; And this is carried out rim detection and histogram analysis extract the edge contour of vehicle, the histogram of wheel correspondence and the edge detection graph of corresponding bar yardstick; Then these are analyzed, calculate, draw the pixel number of length, width, height and the wheelbase of motor vehicles (9); The 4th step, the parameter correction.Utilize the bar yardstick to realize the conversion of the pixel and the size of reality, thereby obtain length, width, height and the wheelbase of motor vehicles; The 5th step, interpretation of result.According to the relevant regulations of motor vehicles (9) physical dimension detection national standard, for its tolerance allowed band of tested motor vehicles (9) be respectively: automobile product (comprising M class, N class and O class) is ± 1%, and the agricultural transportation vehicle product is ± 3%.
The present invention is an example with 07 section of Buick land honor version GL83.0L type commercial vehicle, its detected value and the contrast of vehicle overall size standard value, and the detection error of the regulation that is up to state standards sees following table for details.
The testing result of the present invention and the vehicle overall size standard value table of comparisons
| Standard value (mm) | Detected value (mm) | Detect error (%) | |
| Length of wagon | 5213 | 5186 | 0.5 |
| Bodywork height | 1745 | 1757 | -0.7 |
| Body width | 1847 | 1858 | -0.6 |
| Vehicle wheelbase | 3079 | 3054 | 0.8 |
The present invention has obtained satisfied effect through validation trial.
Claims (3)
1. motor vehicles physical dimension detection system based on image recognition; Comprise digital camera group (2), front controller (6), switch (7) and computing machine (8); Digital camera group (2) is connected with front controller (6); Front controller (6) is connected with computing machine (8) through switch (7); It is characterized in that also comprising black and white bar yardstick (3), two portal frame (4) and single portal frame (5), two portal frames (4) are two-door framework of square, and two-door framework lateral surface of its square is provided with black and white bar yardstick (3); 2～5 meters, dead ahead of each shop front are provided with single portal frame (5) respectively in two-door the framework outside of its square; Single portal frame (5) is a planar shaped simple gate face framework, above its planar shaped simple gate face framework, is provided with digital camera group (2), and black and white bar yardstick (3) orientation that is provided with on the photograph orientation of digital camera group (2) and two portal frame (4) shop front frameworks is symmetrical.
2. motor vehicles physical dimension detection system according to claim 1 is characterized in that digital camera group (2) is by 2～6 digital cameras (1) parallel composition.
3. motor vehicles physical dimension detection system according to claim 1 and 2 is characterized in that in the black and white bar yardstick (3) every black or fine flour striped size equates.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010106226573A CN102175172B (en) | 2010-12-30 | 2010-12-30 | Image-identification-based system for detecting contour dimension of motor vehicle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010106226573A CN102175172B (en) | 2010-12-30 | 2010-12-30 | Image-identification-based system for detecting contour dimension of motor vehicle |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102175172A CN102175172A (en) | 2011-09-07 |
| CN102175172B true CN102175172B (en) | 2012-07-04 |
Family
ID=44518383
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010106226573A Expired - Fee Related CN102175172B (en) | 2010-12-30 | 2010-12-30 | Image-identification-based system for detecting contour dimension of motor vehicle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102175172B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103000067B (en) * | 2012-12-28 | 2014-12-17 | 苏州苏迪智能系统有限公司 | Right angle turning detection system and detecting method thereof |
| CN103954234B (en) * | 2014-04-29 | 2018-01-05 | 南京理工大学 | A kind of vehicle's contour and wheelbase self-calibration Measurement Algorithm |
| CN105783750B (en) * | 2015-11-06 | 2018-11-13 | 佛山市南华仪器股份有限公司 | A method of measuring vehicle wheelbase and front and back outstanding |
| KR20230017365A (en) * | 2016-06-27 | 2023-02-03 | 모빌아이 비젼 테크놀로지스 엘티디. | controlling host vehicle based on detected spacing between stationary vehicles |
| CN106352803A (en) * | 2016-09-28 | 2017-01-25 | 北京信路威科技股份有限公司 | Calibration distance measurement method and system based on video stitching image |
| CN107145810A (en) * | 2017-04-26 | 2017-09-08 | 南京理工大学 | A kind of comprehensive bar code identifying device and method |
| CN112349129A (en) * | 2020-11-07 | 2021-02-09 | 杭州拼便宜网络科技有限公司 | Parking space navigation method and system |
| CN112734671A (en) * | 2021-01-11 | 2021-04-30 | 哈尔滨市科佳通用机电股份有限公司 | Vehicle image deformation repairing method based on deep learning |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3206414B2 (en) * | 1996-01-10 | 2001-09-10 | トヨタ自動車株式会社 | Vehicle type identification device |
| CN1532091A (en) * | 2003-03-18 | 2004-09-29 | 西北工业大学 | Vehicle overall size detection system based on image processing |
| EP1492996A1 (en) * | 2002-04-05 | 2005-01-05 | Ivp Integrated Vision Products Ab | Arrangement in a measuring system |
| CN1955976A (en) * | 2005-10-26 | 2007-05-02 | 电装波动株式会社 | Optical information verifying device |
| CN101373135A (en) * | 2008-07-01 | 2009-02-25 | 南京航空航天大学 | Method and apparatus for measuring three-dimensional curved surface based on instantaneous accidental light irradiation |
| CN201269748Y (en) * | 2008-10-22 | 2009-07-08 | 成都志诚汽车保修检测设备有限公司 | Full-automatic apparatus for measuring vehicle contour dimension according to moving vehicle image |
| CN101646919A (en) * | 2007-03-29 | 2010-02-10 | 学校法人福冈工业大学 | Three-dimensional image measurement apparatus, three-dimensional image measurement method, and three-dimensional image measurement program of non-static object |
| CN201417141Y (en) * | 2009-07-06 | 2010-03-03 | 交通部公路科学研究院 | Outline, height difference and F/R suspension dynamic inspection system for automobiles and trains |
-
2010
- 2010-12-30 CN CN2010106226573A patent/CN102175172B/en not_active Expired - Fee Related
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3206414B2 (en) * | 1996-01-10 | 2001-09-10 | トヨタ自動車株式会社 | Vehicle type identification device |
| EP1492996A1 (en) * | 2002-04-05 | 2005-01-05 | Ivp Integrated Vision Products Ab | Arrangement in a measuring system |
| CN1532091A (en) * | 2003-03-18 | 2004-09-29 | 西北工业大学 | Vehicle overall size detection system based on image processing |
| CN1955976A (en) * | 2005-10-26 | 2007-05-02 | 电装波动株式会社 | Optical information verifying device |
| CN101646919A (en) * | 2007-03-29 | 2010-02-10 | 学校法人福冈工业大学 | Three-dimensional image measurement apparatus, three-dimensional image measurement method, and three-dimensional image measurement program of non-static object |
| CN101373135A (en) * | 2008-07-01 | 2009-02-25 | 南京航空航天大学 | Method and apparatus for measuring three-dimensional curved surface based on instantaneous accidental light irradiation |
| CN201269748Y (en) * | 2008-10-22 | 2009-07-08 | 成都志诚汽车保修检测设备有限公司 | Full-automatic apparatus for measuring vehicle contour dimension according to moving vehicle image |
| CN201417141Y (en) * | 2009-07-06 | 2010-03-03 | 交通部公路科学研究院 | Outline, height difference and F/R suspension dynamic inspection system for automobiles and trains |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102175172A (en) | 2011-09-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102175172B (en) | Image-identification-based system for detecting contour dimension of motor vehicle | |
| CN102901457B (en) | Dynamic measurement method and system for train wheel diameter | |
| CN103712555B (en) | Automotive frame pilot hole vision on-line measurement system and method thereof | |
| CN102519400B (en) | Large slenderness ratio shaft part straightness error detection method based on machine vision | |
| CN103499303B (en) | A kind of wool fineness method for automatic measurement | |
| CN101982727B (en) | Truck volumetric measurement method based on laser triangulation | |
| CN201269748Y (en) | Full-automatic apparatus for measuring vehicle contour dimension according to moving vehicle image | |
| CN105574845B (en) | A kind of polyphaser array cigarette-brand lamination quantity measuring method and device | |
| CN107703513B (en) | Non-contact net relative position detection method based on image processing | |
| CN103593678A (en) | Long-span bridge vehicle dynamic load distribution detection method | |
| CN103954458A (en) | Non-contacting type four-wheel aligner and detecting method thereof | |
| CN109855531B (en) | Dimension measuring system for large-format panel material and measuring method thereof | |
| CN205655802U (en) | Detect equipment in car battery cap clearance | |
| CN205403728U (en) | Laser scanning vehicle outline size measurement device | |
| CN109590952B (en) | Intelligent detection method and detection workbench for complete process assembly plate | |
| CN102798349A (en) | Three-dimensional surface extraction method based on equal-gray line search | |
| CN104634592A (en) | Train running gear fault diagnosis method and train running gear fault diagnosis device | |
| CN107796826A (en) | Miniature duplicate gear broken teeth defect inspection method based on the analysis of tooth heart distance Curve | |
| CN107316330A (en) | Underground hydraulic support frame group pose and verticality measuring method based on many image sequences | |
| CN112858321A (en) | Steel plate surface defect detection system and method based on linear array CCD | |
| JP2022155303A (en) | Measurement method of damage of road surface | |
| CN103175851A (en) | Calibrating tool and calibrating method for multi-camera scanning system | |
| CN103810676A (en) | Method for monitoring running speed of steel pipe | |
| CN106012778A (en) | Digital image collection and analysis method for freeway pavement strain measurement | |
| CN102306382A (en) | Image correcting method of accident scene based on inherent characteristics of road environment |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20120704 Termination date: 20141230 |
|
| EXPY | Termination of patent right or utility model |