CN105488789A - Grading damage assessment method for automobile part - Google Patents
Grading damage assessment method for automobile part Download PDFInfo
- Publication number
- CN105488789A CN105488789A CN201510829433.2A CN201510829433A CN105488789A CN 105488789 A CN105488789 A CN 105488789A CN 201510829433 A CN201510829433 A CN 201510829433A CN 105488789 A CN105488789 A CN 105488789A
- Authority
- CN
- China
- Prior art keywords
- components
- auto parts
- information
- parts
- identification method
- 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.)
- Granted
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
- G06T7/0004—Industrial image inspection
- G06T7/0006—Industrial image inspection using a design-rule based approach
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30164—Workpiece; Machine component
Abstract
The invention discloses a grading damage assessment method for an automobile part, which belongs to the field of automobile damage assessment and is used for solving the problem that extracted information cannot truly reflect an automobile damage degree due to incomplete extracted feature information in the prior art. The technical key point is that the method comprises the steps of information acquisition, information processing and part damage assessment. The step of acquiring information of the automobile part comprises a sub-step of extracting a multi-dimensional feature of the automobile part, wherein the multi-dimensional feature comprises texture information and contour information. The method has the effects that the contour information and the texture information can fully reflect the deformation of the part, and the extraction of the multi-dimensional feature can effectively measure the damage condition of the part.
Description
Technical field
The invention belongs to automobile setting loss field, particularly relate to a kind of auto parts and components classification damage identification method.
Background technology
The current world is in the epoch of development of Mobile Internet technology develop rapidly, and intelligent terminal has penetrated in the life of people gradually.At automotive field, also constantly update performance, automotive system becomes increasingly complex.Release wearable intelligent glasses from Google, to the intelligent glasses for automobile that Bayerische Motorne Werke Aktiengeellschaft proposes, invariably show following development trend.Although the intelligent glasses in automobile, do not occur in the market, popularize, the concept proposed is quite abundant, makes it the every demand meeting user.Such as, the augmented reality glasses that Bayerische Motorne Werke Aktiengeellschaft releases receive much concern at present.Comprise service: navigation, service for life recommendation, prompting message etc.; And maintenance: fault diagnosis, speech recognition, Operating Guideline etc., make auto repair become intelligent, directly perceived, easy.
There is certain deficiency in the realization for technology: intelligent glasses is just for specific vehicle, and in maintenance process, glasses can only identify the parts of specific model; Control mode is single, only has Voice command and sensor to control; Function imperfection, drives and the function of maintenance two aspects although contain, the function such as setting loss before still lacking maintenance.
Summary of the invention
In order to solve in prior art, characteristic information extraction imperfection, the information extraction caused truly cannot reflect the problem of degree of injury, the present invention proposes auto parts and components classification damage identification method, improve vehicle degree of injury information extracting step further, ensure the correctness of setting loss.
To achieve these goals, the technical scheme that the present invention uses is: a kind of auto parts and components classification damage identification method, there is the step of information acquisition, information processing and parts setting loss, during auto parts and components information acquisition, have the step of the multidimensional characteristic extracting auto parts and components, described multidimensional characteristic comprises texture information and profile information.
Further, in described information acquisition step, information collecting device is provided with light source, grating and binocular stereo vision camera, opens light source, gathers the data message of auto parts and components to be measured, by connecing non-contact sensor transmissions in computing machine.
Further, described information handling step, uses CATIA software to carry out the reverse paving of a cloud, modeling, obtains the 3 D stereo information of auto parts and components to be measured in computing machine, comprise each limit size, curvature, some cloud any point to the distance of corresponding flat.
Further, extract the LBP feature of auto parts and components view data, at the initial operator selecting a certain window size, with the center pixel of window for threshold value, adjacent pixel value and threshold value are made comparisons, if pixel value is around greater than the pixel value at center, then this position is marked as 1, otherwise is labeled as 0, pixel around all relatively after, a binary coding is obtained, i.e. the LBP value of central threshold present position according to clockwise or counterclockwise order.
Further, carry out elementary setting loss: the normal auto parts and components model stored in the reconstruction model of auto parts and components to be measured and computing machine is carried out similarity measurement, as follows:
ρ'＝1-ρ(2)
Wherein, ρ is the similarity of auto parts and components to be measured and normal automotive parts, p is the one-dimensional vector of each size composition of normal automotive parts, q is the one-dimensional vector of each size composition of auto parts and components to be measured, and make normalized respectively, weigh the degree of injury of auto parts and components to be measured with ρ ': ρ ' value is less, the degree of injury of parts is less.
Further, carry out ultimate setting loss: extract under line LBP characteristic sum each limit size of auto parts and components view data, curvature, some cloud any point to the data of the distance of corresponding flat, set up SVM support vector cassification model, vehicle impairment scale is classified, when collecting the auto parts and components occurring damage, extract LBP characteristic sum each limit size of the view data of this damage vehicle component, curvature, some cloud any point to the data of the distance of corresponding flat, and obtain corresponding disaggregated model, to judge the degree of injury of this vehicle.
The invention still further relates to the application of auto parts and components classification damage identification method on intelligent glasses used in technique scheme, intelligent glasses is as information collecting device.
Beneficial effect:
1. add the function of the intelligent glasses for auto repair, utilize the 3-D scanning function of glasses, classification setting loss is carried out to vehicle, especially more helpful for defining of slight damage; This invention can provide effective foundation to the damage of vehicle, avoids the wasting of resources, can alleviate the dispute of car insurance to a certain extent.
2. achieve machinery and pattern-recognition fusion interdisciplinary: the CATIA software that make use of Machine Design carries out the reverse Three-dimension Reconstruction Model of a cloud, obtain supplemental characteristic, binding pattern knows method for distinguishing, realizes failure modes, reaches the object of injury scale.
3. the present invention is directed to the auto parts and components that profile is impaired, profile information and texture information fully can reflect the deformation of parts, extract multidimensional characteristic, effectively can weigh the damage situations of parts.
4. parts setting loss classification realizes double insurance: be made up of elementary setting loss and ultimate setting loss.Twice Modling model, the bigness scale amount of the existing parts based on the reverse three-dimensional modeling of a cloud is selected, and contains again the calculating of support vector cassification model, enhances the accuracy of classification setting loss.
Accompanying drawing explanation
Fig. 1 car damage identification overall design block diagram;
Fig. 2 intelligent glasses 3-D scanning schematic diagram.
Embodiment
Embodiment 1:
A kind of auto parts and components classification damage identification method, have the step of information acquisition, information processing and parts setting loss, during auto parts and components information acquisition, have the step of the multidimensional characteristic extracting auto parts and components, described multidimensional characteristic comprises texture information and profile information.
In described information acquisition step, information collecting device is provided with light source, grating and binocular stereo vision camera, opens light source, gathers the data message of auto parts and components to be measured, by connecing non-contact sensor transmissions in computing machine.
Described information handling step, in computing machine, use CATIA software to carry out the reverse paving of a cloud, modeling, obtain the 3 D stereo information of auto parts and components to be measured, comprise each limit size, curvature, some cloud any point to the distance of corresponding flat, above-mentioned information is profile information.
Extract the LBP feature of auto parts and components view data, at the initial operator selecting a certain window size, with the center pixel of window for threshold value, adjacent pixel value and threshold value are made comparisons, if pixel value is around greater than the pixel value at center, then this position is marked as 1, otherwise be labeled as 0, pixel around all relatively after, obtain a binary coding according to clockwise or counterclockwise order, the i.e. LBP value of central threshold present position, it is texture information.
Carry out elementary setting loss: the normal auto parts and components model stored in the reconstruction model of auto parts and components to be measured and computing machine is carried out similarity measurement, as follows:
ρ'＝1-ρ(2)
Wherein, ρ is the similarity of auto parts and components to be measured and normal automotive parts, p is the one-dimensional vector that each size (curvature or distance) of normal automotive parts forms, q is the one-dimensional vector that each size (curvature or distance) of auto parts and components to be measured forms, and make normalized respectively, weigh the degree of injury of auto parts and components to be measured with ρ ': ρ ' value is less, the degree of injury of parts is less.It combined with actual damage measurement qualification result, suggestion degree of injury is divided into following rank:
0≤ρ ' <0.1, UUT is normal condition;
0.1≤ρ ' <0.3, UUT is minor failure;
0.3≤ρ ' <0.7, UUT is moderate fault;
ρ ' >=0.7, UUT is severe fault condition.
Finally, according to the function setting loss of concrete parts, maintenance and replacing.
Carry out ultimate setting loss: extract under line LBP characteristic sum each limit size of auto parts and components view data, curvature, some cloud any point to the data of the distance of corresponding flat, set up SVM support vector cassification model, vehicle impairment scale is classified, when collecting the auto parts and components occurring damage, extract LBP characteristic sum each limit size of the view data of this damage vehicle component, curvature, some cloud any point to the data of the distance of corresponding flat, and obtain corresponding disaggregated model, to judge the degree of injury of this vehicle.
Embodiment 2:
The invention still further relates to the application of auto parts and components classification damage identification method on intelligent glasses used in technique scheme, intelligent glasses is as information collecting device.
Embodiment 3:
When carrying out ultimate setting loss, embodiment 1 extract under employing line LBP characteristic sum each limit size of auto parts and components view data, curvature, some cloud any point to the data of the distance of corresponding flat, set up SVM support vector cassification model, to the method for vehicle impairment scale classification, when collecting the auto parts and components occurring damage, extract LBP characteristic sum each limit size of the view data of this damage vehicle component, curvature, some cloud any point to the data of the distance of corresponding flat, and obtain corresponding disaggregated model, to judge the degree of injury of this vehicle.
Said method make use of the perfect car networked information of my company, obtains a large amount of training sample, and set up the grade separation that car damages fault, this ranking score is level Four by the present embodiment: normal, minor failure, moderate fault and catastrophic failure.For parts to be measured, directly scanned by intelligent glasses, the three-dimensional model rebuild by CATIA and the forecast model of foundation are comprehensively analyzed, and determine that the affiliated car of UUT damages classification.Support vector machine utilizes VC to tie up theoretical and structural risk minimization principle, based on Nonlinear separability, input vector is mapped to high-dimensional feature space through nonlinear transformation, make its linear separability, one and optimum interphase is found in space after the conversion, make its Generalization Ability best, thus carry out linear classification.
Original support vector machine is the classification problem of two classes, adopts support vector machine multi-class classification method here: a class is to all the other.Namely still solve two class problems each time, solve repeatedly.
A given training sample (x
(i), y
(i)), x is feature, and y is result label.I represents the i sample, defined function interval:
By solving weight w and off-set value b, obtaining largest interval, namely obtaining optimal classification surface.A class like this is to all the other, and car damages state and obtains effective classification.
Embodiment 4:
Application number is: the Chinese patent application of 2015106328103 describes a kind of intelligent glasses, and this intelligent glasses can assist the method realized in the above embodiment of the present invention.
Claims (7)
1. an auto parts and components classification damage identification method, there is the step of information acquisition, information processing and parts setting loss, it is characterized in that, during auto parts and components information acquisition, have the step of the multidimensional characteristic extracting auto parts and components, described multidimensional characteristic comprises texture information and profile information.
2. auto parts and components classification damage identification method as claimed in claim 1, it is characterized in that, in described information acquisition step, information collecting device is provided with light source, grating and binocular stereo vision camera, open light source, gather the data message of auto parts and components to be measured, by connecing non-contact sensor transmissions in computing machine.
3. auto parts and components classification damage identification method as claimed in claim 2, it is characterized in that, described information handling step, in computing machine, use CATIA software to carry out the reverse paving of a cloud, modeling, obtain the 3 D stereo information of auto parts and components to be measured, comprise each limit size, curvature, some cloud any point to the distance of corresponding flat.
4. auto parts and components classification damage identification method as claimed in claim 3, it is characterized in that, extract the LBP feature of auto parts and components view data, at the initial operator selecting a certain window size, with the center pixel of window for threshold value, adjacent pixel value and threshold value are made comparisons, if pixel value is around greater than the pixel value at center, then this position is marked as 1, otherwise be labeled as 0, pixel around all relatively after, obtain a binary coding according to clockwise or counterclockwise order, i.e. the LBP value of central threshold present position.
5. auto parts and components classification damage identification method as claimed in claim 4, is characterized in that, carry out elementary setting loss: the normal auto parts and components model stored in the reconstruction model of auto parts and components to be measured and computing machine is carried out similarity measurement, as follows:
ρ'＝1-ρ(2)
Wherein, ρ is the similarity of auto parts and components to be measured and normal automotive parts, p is the one-dimensional vector of each size composition of normal automotive parts, q is the one-dimensional vector of each size composition of auto parts and components to be measured, and make normalized respectively, weigh the degree of injury of auto parts and components to be measured with ρ ': ρ ' value is less, the degree of injury of parts is less.
6. auto parts and components classification damage identification method as claimed in claim 5, it is characterized in that, carry out ultimate setting loss: the LBP characteristic sum each limit size extracting auto parts and components view data under line, curvature, put the data of any point to the distance of corresponding flat of cloud, set up SVM support vector cassification model, vehicle impairment scale is classified, when collecting the auto parts and components occurring damage, extract LBP characteristic sum each limit size of the view data of this damage vehicle component, curvature, put the data of any point to the distance of corresponding flat of cloud, and obtain corresponding disaggregated model, to judge the degree of injury of this vehicle.
7. the application of a kind of auto parts and components classification damage identification method on intelligent glasses as described in claim 1-6, intelligent glasses is as information collecting device.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510829433.2A CN105488789B (en) | 2015-11-24 | Auto parts and components are classified damage identification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510829433.2A CN105488789B (en) | 2015-11-24 | Auto parts and components are classified damage identification method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105488789A true CN105488789A (en) | 2016-04-13 |
| CN105488789B CN105488789B (en) | 2018-08-31 |
Family
ID=
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105930680A (en) * | 2016-05-27 | 2016-09-07 | 大连楼兰科技股份有限公司 | Damage degree model verification method and system |
| CN106022387A (en) * | 2016-05-27 | 2016-10-12 | 大连楼兰科技股份有限公司 | Method and system for testing damage grade model |
| CN106066907A (en) * | 2016-05-27 | 2016-11-02 | 大连楼兰科技股份有限公司 | The setting loss grading method judged based on many parts multi-model |
| CN106092597A (en) * | 2016-05-27 | 2016-11-09 | 大连楼兰科技股份有限公司 | Based on mathematical model method of testing and the system of sharing formula |
| CN106504248A (en) * | 2016-12-06 | 2017-03-15 | 成都通甲优博科技有限责任公司 | Vehicle damage method of discrimination based on computer vision |
| CN106530410A (en) * | 2016-11-04 | 2017-03-22 | 大连文森特软件科技有限公司 | Automobile assembly demonstration system based on augmented reality technology |
| CN106530403A (en) * | 2016-11-04 | 2017-03-22 | 大连文森特软件科技有限公司 | Automobile component quality inspection system based on augmented reality technology |
| CN107403424A (en) * | 2017-04-11 | 2017-11-28 | 阿里巴巴集团控股有限公司 | A kind of car damage identification method based on image, device and electronic equipment |
| CN107608507A (en) * | 2017-09-05 | 2018-01-19 | 清华大学 | The method for selecting of locomotive component and locomotive auxiliary maintaining system under low light condition |
| CN108171708A (en) * | 2018-01-24 | 2018-06-15 | 北京威远图易数字科技有限公司 | Car damage identification method and system |
| CN108364253A (en) * | 2018-03-15 | 2018-08-03 | 北京威远图易数字科技有限公司 | Car damage identification method, system and electronic equipment |
| CN109146834A (en) * | 2017-06-13 | 2019-01-04 | 上海擎感智能科技有限公司 | Car damage identification method and device, computer readable storage medium, terminal |
| CN109271908A (en) * | 2018-09-03 | 2019-01-25 | 阿里巴巴集团控股有限公司 | Vehicle damages detection method, device and equipment |
| CN109614935A (en) * | 2018-12-12 | 2019-04-12 | 泰康保险集团股份有限公司 | Car damage identification method and device, storage medium and electronic equipment |
| CN110008823A (en) * | 2019-02-19 | 2019-07-12 | 阿里巴巴集团控股有限公司 | Car damage identification method and device, electronic equipment |
| WO2019169688A1 (en) * | 2018-03-09 | 2019-09-12 | 平安科技（深圳）有限公司 | Vehicle loss assessment method and apparatus, electronic device, and storage medium |
| US10817956B2 (en) | 2017-04-11 | 2020-10-27 | Alibaba Group Holding Limited | Image-based vehicle damage determining method and apparatus, and electronic device |
| US11544914B2 (en) | 2021-02-18 | 2023-01-03 | Inait Sa | Annotation of 3D models with signs of use visible in 2D images |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09297838A (en) * | 1996-05-08 | 1997-11-18 | Casio Comput Co Ltd | Image processor |
| US6185540B1 (en) * | 1994-12-28 | 2001-02-06 | Automatic Data Processing | Insurance estimating system |
| US20020161533A1 (en) * | 2000-02-15 | 2002-10-31 | Tateo Uegaki | System for recognizing damaged part of accident-involved car and computer-readable medium on which program is recorded |
| JP4057571B2 (en) * | 2004-09-14 | 2008-03-05 | 株式会社日立情報システムズ | Image analysis system and image analysis method |
| CN104268879A (en) * | 2014-09-28 | 2015-01-07 | 民政部国家减灾中心 | Physical building quantity damage evaluation method based on remote sensing multi-spectral images |
| JP2015191572A (en) * | 2014-03-28 | 2015-11-02 | 株式会社パスコ | Manhole detection device, manhole detection method and program |
| CN105182535A (en) * | 2015-09-28 | 2015-12-23 | 大连楼兰科技股份有限公司 | Method of using intelligent glasses for vehicle maintenance |
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6185540B1 (en) * | 1994-12-28 | 2001-02-06 | Automatic Data Processing | Insurance estimating system |
| JPH09297838A (en) * | 1996-05-08 | 1997-11-18 | Casio Comput Co Ltd | Image processor |
| US20020161533A1 (en) * | 2000-02-15 | 2002-10-31 | Tateo Uegaki | System for recognizing damaged part of accident-involved car and computer-readable medium on which program is recorded |
| JP4057571B2 (en) * | 2004-09-14 | 2008-03-05 | 株式会社日立情報システムズ | Image analysis system and image analysis method |
| JP2015191572A (en) * | 2014-03-28 | 2015-11-02 | 株式会社パスコ | Manhole detection device, manhole detection method and program |
| CN104268879A (en) * | 2014-09-28 | 2015-01-07 | 民政部国家减灾中心 | Physical building quantity damage evaluation method based on remote sensing multi-spectral images |
| CN105182535A (en) * | 2015-09-28 | 2015-12-23 | 大连楼兰科技股份有限公司 | Method of using intelligent glasses for vehicle maintenance |
Non-Patent Citations (3)
| Title |
|---|
| 张文建 等: "基于CATIAV5的汽车车身逆向设计", 《机械》 * |
| 王雪 等: "采用小波分析与支持向量机的车轮踏面擦伤识别方法", 《中国机械工程》 * |
| 陈果 等: "基于孔探图像纹理特征的航空发动机损伤识别方法", 《仪器仪表学报》 * |
Cited By (28)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106022387A (en) * | 2016-05-27 | 2016-10-12 | 大连楼兰科技股份有限公司 | Method and system for testing damage grade model |
| CN106066907A (en) * | 2016-05-27 | 2016-11-02 | 大连楼兰科技股份有限公司 | The setting loss grading method judged based on many parts multi-model |
| CN106092597A (en) * | 2016-05-27 | 2016-11-09 | 大连楼兰科技股份有限公司 | Based on mathematical model method of testing and the system of sharing formula |
| CN105930680A (en) * | 2016-05-27 | 2016-09-07 | 大连楼兰科技股份有限公司 | Damage degree model verification method and system |
| CN106066907B (en) * | 2016-05-27 | 2020-04-14 | 大连楼兰科技股份有限公司 | Loss assessment grading method based on multi-part multi-model judgment |
| CN106530410B (en) * | 2016-11-04 | 2019-03-12 | 快创科技（大连）有限公司 | A kind of automobile demo system based on augmented reality |
| CN106530410A (en) * | 2016-11-04 | 2017-03-22 | 大连文森特软件科技有限公司 | Automobile assembly demonstration system based on augmented reality technology |
| CN106530403A (en) * | 2016-11-04 | 2017-03-22 | 大连文森特软件科技有限公司 | Automobile component quality inspection system based on augmented reality technology |
| CN106530403B (en) * | 2016-11-04 | 2019-04-05 | 快创科技（大连）有限公司 | A kind of auto parts and components quality inspection system based on augmented reality |
| CN106504248A (en) * | 2016-12-06 | 2017-03-15 | 成都通甲优博科技有限责任公司 | Vehicle damage method of discrimination based on computer vision |
| CN106504248B (en) * | 2016-12-06 | 2021-02-26 | 成都通甲优博科技有限责任公司 | Vehicle damage judging method based on computer vision |
| CN107403424A (en) * | 2017-04-11 | 2017-11-28 | 阿里巴巴集团控股有限公司 | A kind of car damage identification method based on image, device and electronic equipment |
| CN107403424B (en) * | 2017-04-11 | 2020-09-18 | 阿里巴巴集团控股有限公司 | Vehicle loss assessment method and device based on image and electronic equipment |
| US10789786B2 (en) | 2017-04-11 | 2020-09-29 | Alibaba Group Holding Limited | Picture-based vehicle loss assessment |
| US10817956B2 (en) | 2017-04-11 | 2020-10-27 | Alibaba Group Holding Limited | Image-based vehicle damage determining method and apparatus, and electronic device |
| US11049334B2 (en) | 2017-04-11 | 2021-06-29 | Advanced New Technologies Co., Ltd. | Picture-based vehicle loss assessment |
| CN109146834A (en) * | 2017-06-13 | 2019-01-04 | 上海擎感智能科技有限公司 | Car damage identification method and device, computer readable storage medium, terminal |
| CN107608507A (en) * | 2017-09-05 | 2018-01-19 | 清华大学 | The method for selecting of locomotive component and locomotive auxiliary maintaining system under low light condition |
| CN108171708A (en) * | 2018-01-24 | 2018-06-15 | 北京威远图易数字科技有限公司 | Car damage identification method and system |
| WO2019169688A1 (en) * | 2018-03-09 | 2019-09-12 | 平安科技（深圳）有限公司 | Vehicle loss assessment method and apparatus, electronic device, and storage medium |
| CN108364253A (en) * | 2018-03-15 | 2018-08-03 | 北京威远图易数字科技有限公司 | Car damage identification method, system and electronic equipment |
| CN108364253B (en) * | 2018-03-15 | 2022-04-15 | 北京威远图易数字科技有限公司 | Vehicle damage assessment method and system and electronic equipment |
| CN109271908A (en) * | 2018-09-03 | 2019-01-25 | 阿里巴巴集团控股有限公司 | Vehicle damages detection method, device and equipment |
| CN109271908B (en) * | 2018-09-03 | 2022-05-13 | 创新先进技术有限公司 | Vehicle loss detection method, device and equipment |
| CN109614935B (en) * | 2018-12-12 | 2021-07-06 | 泰康保险集团股份有限公司 | Vehicle damage assessment method and device, storage medium and electronic equipment |
| CN109614935A (en) * | 2018-12-12 | 2019-04-12 | 泰康保险集团股份有限公司 | Car damage identification method and device, storage medium and electronic equipment |
| CN110008823A (en) * | 2019-02-19 | 2019-07-12 | 阿里巴巴集团控股有限公司 | Car damage identification method and device, electronic equipment |
| US11544914B2 (en) | 2021-02-18 | 2023-01-03 | Inait Sa | Annotation of 3D models with signs of use visible in 2D images |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113850825B (en) | Remote sensing image road segmentation method based on context information and multi-scale feature fusion | |
| CN109829891B (en) | Magnetic shoe surface defect detection method based on dense generation of antagonistic neural network | |
| CN105719188A (en) | Method and server for achieving insurance claim anti-fraud based on consistency of multiple pictures | |
| CN111161349B (en) | Object posture estimation method, device and equipment | |
| CN108875602A (en) | Monitor the face identification method based on deep learning under environment | |
| CN106370128A (en) | Automobile part damage assessment method | |
| CN105426905A (en) | Robot barrier identification method based on gradient histogram and support vector machine | |
| CN103761515A (en) | Human face feature extracting method and device based on LBP | |
| CN105282543A (en) | Total blindness three-dimensional image quality objective evaluation method based on three-dimensional visual perception | |
| CN105376563A (en) | No-reference three-dimensional image quality evaluation method based on binocular fusion feature similarity | |
| CN105160303A (en) | Fingerprint identification method based on mixed matching | |
| CN105357519A (en) | Quality objective evaluation method for three-dimensional image without reference based on self-similarity characteristic | |
| CN113435424A (en) | Method and system for identifying destroying granularity of confidential medium | |
| CN106780727B (en) | Vehicle head detection model reconstruction method and device | |
| CN113408321B (en) | Real-time target detection method and device for lightweight image and video data | |
| CN114332802A (en) | Road surface flatness semantic segmentation method and system based on binocular camera | |
| CN114694039A (en) | Remote sensing hyperspectral and laser radar image fusion classification method and device | |
| CN114693963A (en) | Recognition model training and recognition method and device based on electric power data feature extraction | |
| CN111797704B (en) | Action recognition method based on related object perception | |
| CN105069794A (en) | Binocular rivalry based totally blind stereo image quality evaluation method | |
| CN103761538A (en) | Traffic sign recognition method based on shape feature invariant subspace | |
| CN117037119A (en) | Road target detection method and system based on improved YOLOv8 | |
| CN105488789A (en) | Grading damage assessment method for automobile part | |
| KR101391667B1 (en) | A model learning and recognition method for object category recognition robust to scale changes | |
| CN105488789B (en) | Auto parts and components are classified damage identification method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |