CN109597991B - Automobile remote detection method - Google Patents
Automobile remote detection method Download PDFInfo
- Publication number
- CN109597991B CN109597991B CN201811425532.4A CN201811425532A CN109597991B CN 109597991 B CN109597991 B CN 109597991B CN 201811425532 A CN201811425532 A CN 201811425532A CN 109597991 B CN109597991 B CN 109597991B
- Authority
- CN
- China
- Prior art keywords
- detection
- vehicle
- defect
- semantic
- background server
- 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.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 233
- 230000007547 defect Effects 0.000 claims abstract description 82
- 238000000034 method Methods 0.000 claims abstract description 39
- 238000004458 analytical method Methods 0.000 claims abstract description 13
- 238000011156 evaluation Methods 0.000 claims abstract description 10
- 238000007689 inspection Methods 0.000 claims description 33
- 230000007246 mechanism Effects 0.000 claims description 24
- 238000012545 processing Methods 0.000 claims description 17
- 239000006096 absorbing agent Substances 0.000 claims description 11
- 230000035939 shock Effects 0.000 claims description 11
- 230000008569 process Effects 0.000 claims description 10
- 238000013507 mapping Methods 0.000 claims description 9
- 239000000284 extract Substances 0.000 claims description 8
- 238000000605 extraction Methods 0.000 claims description 6
- 238000012423 maintenance Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000013527 convolutional neural network Methods 0.000 claims description 3
- 239000002828 fuel tank Substances 0.000 claims description 3
- 230000004927 fusion Effects 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 2
- 238000007789 sealing Methods 0.000 claims 4
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims 3
- 238000005192 partition Methods 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 210000004709 eyebrow Anatomy 0.000 description 4
- 239000003973 paint Substances 0.000 description 4
- 238000010801 machine learning Methods 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 239000003086 colorant Substances 0.000 description 2
- 239000000428 dust Substances 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 210000003195 fascia Anatomy 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000003058 natural language processing Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000012550 audit Methods 0.000 description 1
- 230000004888 barrier function Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 238000013210 evaluation model Methods 0.000 description 1
- 210000003128 head Anatomy 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
Landscapes
- Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides an automobile remote detection method. The method comprises the following steps: the method comprises the steps of storing a detection target of a vehicle and a semantic matching rule corresponding to the detection target in a vehicle defect semantic recognition rule library: determining a detection item according to detection requirements, acquiring a detection target corresponding to the detection item, detecting the detection target of the vehicle through a detection tool, and submitting a preliminary defect detection report of the detected vehicle to a background server; the background server performs semantic analysis on text content in the defect preliminary detection report by utilizing semantic matching rules to obtain vehicle defect description key information, and obtains a vehicle evaluation report based on the vehicle defect description key information. According to the invention, through presetting the vehicle defect semantic recognition rule base, the semantic matching rule corresponding to the detection target is designed and utilized, so that the nonstandard voice and text input of the inspector is realized, the standard vehicle condition information is automatically converted, and the method has extremely high flexibility and convenience.
Description
Technical Field
The invention relates to the technical field of automobile detection, in particular to an automobile remote detection method.
Background
At present, the detection of the second hand vehicle on the market is thousands of people and thousands of people, and a recognized detection standard is lacked. The traditional second-hand vehicle detection tool has very high requirement on the service capability of an inspector, and needs professional detection training to detect the actual vehicle, and correspondingly, the cost of enterprises and the service popularization difficulty are increased. Many 4S store/establishment handling specialists do not have professional second hand vehicle detection capabilities and require third party detection establishment support.
The conventional second-hand vehicle detection tool does not have the flexible adjustment capability for the newly added service, and as a simple example, the financial estimation service only needs to shoot basic photos of the vehicle, has a requirement for rapid detection and evaluation on the vehicle condition, is not used for transaction, and is pursued for detection efficiency, the second-hand vehicle transaction service is very comprehensive in detecting the vehicle, and the vehicle condition information needs to be completely recorded for the basic photos and procedure information of the vehicle, and if the conventional detection tool is shared, a large number of photos need to be shot to meet the transaction service, but the time-efficiency requirement of the financial estimation service is not met.
NLP (Natural Language Processing, natural semantic recognition) is a sub-field of Artificial Intelligence (AI). Semantic recognition technology is equivalent to human brain and is responsible for thinking and information processing. Modern NLP algorithms are based on machine learning, in particular statistical machine learning. The machine learning paradigm is different from the generally prior attempted language processing. Implementation of language processing tasks typically involves large sets of regular codes directly by hand.
At present, the second-hand vehicle detection industry does not have a practical application scheme based on natural semantic recognition.
Disclosure of Invention
The embodiment of the invention provides an automobile remote detection method for overcoming the problems in the prior art.
In order to achieve the above purpose, the present invention adopts the following technical scheme.
A car remote detection method stores a detection target of a car and a semantic matching rule corresponding to the detection target in a car defect semantic recognition rule base, and the method comprises the following steps:
determining a detection item according to detection requirements, acquiring a detection target corresponding to the detection item, detecting the detection target of the vehicle through a detection tool, and submitting a preliminary defect detection report of the detected vehicle to a background server;
and the background server performs semantic analysis on text content in the defect preliminary detection report by utilizing semantic matching rules corresponding to detection targets stored in the vehicle defect semantic recognition rule base to obtain vehicle defect description key information, and obtains a vehicle evaluation report based on the vehicle defect description key information.
Further, the method further comprises the steps of:
the method comprises the steps that firstly, a detector registers in a detection mechanism, identity information of the detector is submitted to the detection mechanism, the detection mechanism transmits the received identity information of the detector to a background server, after the background server checks that the identity information of the detector passes, an identity checking passing message of the detector is returned to the detection mechanism, the detector registers successfully in the detection mechanism, and the detection mechanism stores and manages the identity information of the detector.
Further, the storing, in the vehicle defect semantic recognition rule base, the detection target of the vehicle and the semantic matching rule corresponding to the detection target includes:
a vehicle defect semantic recognition rule base is managed in a background server, and all detection targets of the vehicle, all possible detection results corresponding to each detection target and semantic matching rules corresponding to each detection target are stored in the vehicle defect semantic recognition rule base, wherein the semantic matching rules comprise vehicle condition semantic codes represented by voice, pictures, mathematical symbols, marks and/or texts.
Further, all the detection targets include: left front tire, left front hub, left front fender, left A pillar outer, left front door, left outside rear view mirror, left front door edge seal, left bottom large edge, left front door sill, left B pillar outer, left C pillar outer, left rear door edge seal, left rear door sill, left rear tire, left rear hub, fuel tank cap, left rear fender, front bumper, grille, right headlamp assembly, left front fog lamp, a right front fog light, a front windshield wiper, a roof, a rear windshield, a trunk lid edge, a rear windshield wiper, a right front hub, a right front tire, a right front fender, a right A pillar outer, a right outside rear view mirror, a right front door seal, a right bottom edge, a right front door sill, a right B pillar outer, a right rear door, a right C pillar outer, a right rear door seal, a right rear door sill, a right rear fender, a right rear hub, a right rear tire, a roof, a trunk, a left front door trim, a left A pillar trim, a left rear door trim, a left B pillar trim, a left C pillar trim, a steering wheel, a workbench, a front drain trim outlet, a glove box, an instrument, a center console, a handle bar dust cover, a armrest box, a main drive seat, a secondary drive seat, a left rear seat, a right rear seat, a left front door trim, a right A pillar trim, a right rear door trim, a right B pillar trim, a right C pillar trim, a front bumper beam, a left front shock absorber mount, a removable tank frame, a left headlamp frame, a tank upper bracket, a tank lower bracket, a left front fender liner, a left front rail, a left front A pillar inner side and edge seal, a left B pillar inner and edge seal, a left C pillar inner and edge seal, a front dash, a rear bumper beam, a rear dash panel, a left rear shock absorber mount, left tail lamp frame, vehicle body bottom plate, suitcase banding, suitcase baffle, left back longeron, right front shock absorber seat, right head lamp frame, right front fender inside lining, right front longeron, right B post inboard and banding, right C post inboard and banding, right A post inboard and banding, right back shock absorber seat, right tail lamp frame, right back longeron, left front door handle, left back door handle, right front door handle, right back door handle, front fog lamp, back fog lamp, left front wheel eyebrow, left back wheel eyebrow, right front wheel eyebrow, right back wheel eyebrow, front bumper lower plaque, back bumper lower plaque, front fog lamp grid and back fog lamp grid.
Further, the determining the detection item according to the detection requirement, obtaining a detection target corresponding to the detection item, detecting the detection target of the vehicle by a detection tool, and submitting a preliminary detection report of the detected defect of the vehicle to a background server, including:
a detection configuration database is managed in a background server, each detection item on the computer application program APP is configured in the detection configuration database according to the service requirement, and each detection item comprises a plurality of detection targets;
firstly, a inspector determines an inspection item according to actual inspection requirements, inquires an inspection configuration database according to the inspection item to obtain an inspection target corresponding to the inspection item, the inspector utilizes an APP to inspect the inspection target of the vehicle through various inspection tools, and the inspection target is inspected through voice, pictures and/or text description to obtain an inspection preliminary inspection report of the vehicle, and the inspection preliminary inspection report of the vehicle is submitted to a background server.
Further, the background server performs semantic analysis on text content in the defect preliminary detection report by using semantic matching rules corresponding to detection targets stored in the vehicle defect semantic recognition rule library to obtain vehicle defect description key information, and obtains a vehicle evaluation report based on the vehicle defect description key information, including:
after receiving the defect preliminary detection report of the vehicle, the background server extracts text content in the defect preliminary detection report, and performs semantic analysis on the text content by utilizing semantic matching rules to obtain vehicle defect description key information;
the background server extracts the picture content in the defect preliminary detection report, locates an image area where the license plate is located in the picture content, and performs blurring on the image area where the license plate is located;
the background server utilizes a vehicle model database to identify the style information of the vehicle according to the vehicle defect description key information, confirms the vehicle configuration information, inquires the vehicle history maintenance record and the insurance record through a maintenance inquiry interface of the background, synthesizes various detection data, and combines the vehicle defect description key information and a license plate blurring result to obtain a vehicle evaluation report;
the semantic analysis of the text content by using the semantic matching rule comprises the following processing steps:
step_1, performing Chinese special character processing on the text content, and converting full-angle characters in the text content into half-angle characters;
step_2, dividing sentences of the text content by English semicolon(s) to obtain a sentence set to be processed;
step 3, performing word segmentation processing on each sentence to obtain a vocabulary sequence of each sentence;
step 4, traversing the vocabulary sequence of the sentence, and replacing the mapping words with corresponding mapping content sets to obtain a set of mapped vocabulary sequences;
step_5, recombining the mapped vocabulary sequences to obtain a recombined sentence set, and segmenting the recombined sentences to obtain a new vocabulary sequence set;
step_6, mapping each vocabulary in the vocabulary sequence into a corresponding detection target stored in a vehicle defect semantic recognition rule base, and carrying out rule matching on each vocabulary sequence according to a semantic matching rule corresponding to the detection target stored in the vehicle defect semantic recognition rule base to obtain candidate vehicle defect detection information;
the key information extraction process comprises the following steps:
1. traversing a vocabulary sequence in the candidate vehicle defect detection information set;
2. finding a vocabulary C marked as a component;
3. traversing the online vocabulary forwards and backwards respectively based on the vocabulary C, traversing the window size of the context to be 5, and searching the vocabulary marked as position and description so as to form a key information combination;
4. when traversing the context, the maximum number of position and depth vocabularies found forward and backward is 2;
5. and obtaining the vehicle defect description key information.
Further, the background server extracts the picture content in the defect preliminary detection report, locates the image area where the license plate is located in the picture content, and performs blurring on the image area where the license plate is located, including:
the background server extracts picture content in a defect preliminary detection report, a deep convolutional neural network is used for extracting high-level features of license plates in images, a rectangular coordinate point and a rotation angle of the position of the license plates are obtained through forward calculation in a mode of fusion of shallow features and deep features, an image area of the license plates is framed according to the rectangular coordinate point and the rotation angle, and license plate blurring processing is carried out on the image area of the license plates by means of a mean value filtering process.
According to the technical scheme provided by the embodiment of the invention, the embodiment of the invention has the data learning function by customizing the semantic recognition rule base in advance, so that the more detectors are used, the more accurate the data is, the threshold of detection in the existing second-hand vehicle industry is greatly simplified, and the detection capability of the detectors is improved. By designing the vehicle condition semantic recognition method, non-standardized voice and text input of the inspector is realized, the method is automatically converted into standard vehicle condition information, the method has extremely high flexibility and convenience, the inspector is not required to shoot all vehicle condition pictures according to detection requirements, and only needs to shoot remarks of the vehicle conditions with problems.
Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic diagram of an implementation principle of an automobile remote detection method according to an embodiment of the present invention;
FIG. 2 is a process flow diagram of a method for remote detection of an automobile according to an embodiment of the present invention;
fig. 3 is a process flow diagram of semantic analysis according to an embodiment of the present invention.
Detailed Description
Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals refer to the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the drawings are exemplary only for explaining the present invention and are not to be construed as limiting the present invention.
As used herein, the singular forms "a", "an", "the" and "the" are intended to include the plural forms as well, unless expressly stated otherwise, as understood by those skilled in the art. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. It will be understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may also be present. Further, "connected" or "coupled" as used herein may include wirelessly connected or coupled. The term "and/or" as used herein includes any and all combinations of one or more of the associated listed items.
It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
For the purpose of facilitating an understanding of the embodiments of the invention, reference will now be made to the drawings of several specific embodiments illustrated in the drawings and in no way should be taken to limit the embodiments of the invention.
The embodiment of the invention provides an automobile remote detection method, which designs detection rules and processes and simple and easy-to-use detection functions according to experience of industrial intensive inspectors, so that inspectors with basic vehicle common sense can take specified photos according to requirements, import paint film instrument data and assist the inspectors to generate reliable second-hand vehicle detection reports in a dual mode of cloud identification and manual identification. According to the method, the detection capability of the automobile is enabled to users with different detection capabilities in a cloud mode, and the problem of the automobile is found in time.
The automobile remote detection method provided by the embodiment of the invention is matched with APP (Application).
The user roles in the automobile remote detection method provided by the embodiment of the invention are defined as follows:
inspector: all front end inspectors using the automobile remote detection method;
cloud: a background server of the automobile remote detection method, wherein a user of the background server comprises a professional evaluator team;
the detection mechanism comprises: all third parties using the external management background detect the demand clients;
the buyer: registering registered auction buyers, possibly car dealers, 4S shop groups, etc., in the millennium association platform;
the seller: the second-hand car sellers registered in the Qianliye correlation platform can be car sellers, 4S shop groups and markets;
identity authentication flow:
the inspector needs to submit the identity information real name authentication of the inspector, and the real name detection of the vehicle is carried out after the background audit of the automobile remote detection method is passed;
the detection mechanism can manage the inspector and conduct transaction only when the detection mechanism needs to make a contract with the automobile remote detection method platform;
buyers and sellers: the related auction platform in the automobile remote detection method is required to pay the guarantee deposit, so that the vehicle can be transacted
Detecting a business flow:
the seller and the detection mechanism initiate detection requirements, a inspector is arranged to initiate detection on the vehicle through a dispatching system and offline rules, different detection configurations can be obtained according to service requirements by the automobile remote detection method, and then detection information is perfected and submitted to the Qianli background.
The evaluator processes the detection data according to the system dispatch and the auxiliary functions of the system, generates a professional detection report and distributes the professional detection report to the inspector and the detection mechanism; if the detected data has problems, the detected data is refused to the inspector to request re-detection; if the detection data has no problem, the detection personnel or the detection mechanism can use the detection report to directly put on the shelf commodity for transaction.
The schematic implementation diagram of the method for remote detection of the automobile provided by the embodiment of the invention is shown in fig. 1, and the specific processing flow is shown in fig. 2, and the method comprises the following processing steps:
step 21, firstly, the inspector needs to register in the inspection mechanism, and the identity information of the inspector is submitted to the inspection mechanism. The detection mechanism transmits the received identity information of the inspector to the cloud (namely a background server of the automobile remote detection method). And after the identity information of the cloud auditing inspector passes, returning an identity auditing passing message of the inspector to the inspection mechanism. Then, the inspector registers successfully in the inspection mechanism, the inspection mechanism stores the identity information of the inspector, manages the inspector, and can conduct automobile transaction.
Step 22, storing all detection targets of the vehicle, all possible detection results corresponding to each detection target and semantic matching rules corresponding to each detection target in a vehicle defect semantic recognition rule base.
A vehicle defect semantic recognition rule base is predefined, and all detection targets of the vehicle and all possible detection results corresponding to each detection target are stored in the vehicle defect semantic recognition rule base, and semantic matching rules corresponding to each detection target are stored, wherein the semantic matching rules can comprise vehicle condition semantic codes represented by voice, pictures, mathematical symbols, marks and/or texts.
All detection targets may include: left front tire, left front hub, left front fender, left A pillar outer, left front door, left outside rear view mirror, left front door edge seal, left bottom large edge, left front door sill, left B pillar outer, left C pillar outer, left rear door edge seal, left rear door sill, left rear tire, left rear hub, fuel tank cap, left rear fender, front bumper, grille, right headlamp assembly, left front fog lamp, a right front fog light, a front windshield wiper, a roof, a rear windshield, a trunk lid edge, a rear windshield wiper, a right front hub, a right front tire, a right front fender, a right A pillar outer, a right outside rear view mirror, a right front door seal, a right bottom edge, a right front door sill, a right B pillar outer, a right rear door, a right C pillar outer, a right rear door seal, a right rear door sill, a right rear fender, a right rear hub, a right rear tire, a roof, a trunk, a left front door trim, a left A pillar trim, a left rear door trim, a left B pillar trim, a left C pillar trim, a steering wheel, a workbench, a front drain trim outlet, a glove box, an instrument, a center hub, a handle bar dust cover, a armrest box, a main drive seat, a secondary drive seat, a left rear seat, a right rear seat, a left front door trim, a right A pillar trim, a right rear door trim, a right B pillar trim, a right C pillar trim, a front bumper beam, a left front shock absorber mount, a removable tank frame, a left front lamp frame, a tank upper bracket, a tank lower bracket, a left front fender liner, a left front rail, a left front A pillar inner side and edge seal, a left B pillar inner and edge seal, a left C pillar inner and edge seal, a front perimeter barrier (firewall), a rear bumper beam, rear panel, left rear shock absorber seat, left tail lamp frame, vehicle body floor, luggage case banding, luggage case bulkhead, left rear longitudinal beam, right front shock absorber seat, right headlamp frame, right front fender lining, right front longitudinal beam, right B post inboard and banding, right C post inboard and banding, right A post inboard and banding, right rear shock absorber seat, right tail lamp frame, right rear longitudinal beam, left front door handle, left rear door handle, right front door handle, right rear door handle, front fog lamp, rear fog lamp, left front wheel trim, left rear wheel trim, right front wheel trim, right rear wheel trim, under front bumper fascia, under rear bumper fascia, front fog lamp grille and rear fog lamp grille.
For example, for a detection target of a paint film of a vehicle, the detection result includes: and (3) repairing paint and putty.
Step 23, configuring each detection item on the APP in a detection configuration database of the cloud, wherein each detection item comprises a plurality of detection targets.
A set of detection configuration database is also established in the cloud, and each detection item on the APP (Application) is configured in the detection configuration database according to the service requirement, and each detection item comprises a plurality of detection targets. The detection targets included in each detection item can be flexibly configured according to actual needs, and are increased or decreased appropriately.
And step 24, detecting a detection target of the vehicle by using an APP through various detection tools, and submitting a preliminary detection report of the defects of the vehicle to the cloud.
Firstly, a inspector determines a detection item according to actual detection requirements, and queries a detection configuration database according to the detection item to obtain a detection target corresponding to the detection item. Then, an inspector detects a detection target of the vehicle through various detection tools by using the APP, and obtains a preliminary detection report of the defect of the vehicle through describing the detected defect problem through voice, pictures, characters and the like, and submits the preliminary detection report of the defect of the vehicle to a cloud after the detection is finished, and waits for an automobile evaluation result returned by the cloud.
And step 25, carrying out semantic analysis on text content in the defect preliminary detection report by utilizing a semantic matching rule to obtain vehicle defect description key information, and obtaining a vehicle evaluation report by combining a license plate blurring process.
After the cloud receives the defect preliminary detection report of the vehicle, extracting text content in the defect preliminary detection report, carrying out semantic analysis on the text content, and carrying out semantic analysis on the text content by utilizing a semantic matching rule to obtain vehicle defect description key information. The processing flow of the semantic analysis is shown in fig. 3, and comprises the following processing steps:
step_1, performing Chinese special character processing on the text content, and converting full-angle characters in the text content into half-angle characters;
step_2, dividing sentences of the text content by English semicolon(s) to obtain a sentence set to be processed;
step 3, performing word segmentation processing on each sentence to obtain a vocabulary sequence of each sentence;
step 4, traversing the vocabulary sequence of the sentence, and replacing the mapping words with corresponding mapping content sets to obtain a set of mapped vocabulary sequences;
step_5, recombining the mapped vocabulary sequences to obtain a recombined sentence set, and segmenting the recombined sentences to obtain a new vocabulary sequence set;
step_6, mapping each vocabulary in the vocabulary sequence into a corresponding detection target stored in a vehicle defect semantic recognition rule base, and carrying out rule matching on each vocabulary sequence according to a semantic matching rule corresponding to the detection target stored in the vehicle defect semantic recognition rule base to obtain candidate vehicle defect detection information;
and Step 7, extracting the vehicle defect description key information from the candidate vehicle defect detection information set by using a key information extraction flow.
The key information extraction process comprises the following steps:
1. traversing a vocabulary sequence in the candidate vehicle defect detection information set;
2. finding a vocabulary C marked as a component;
3. traversing the online vocabulary forwards and backwards respectively based on the vocabulary C, traversing the window size of the context to be 5, and searching the vocabulary marked as position and description so as to form a key information combination;
4. when traversing the context, the maximum number of position and depth vocabularies found forward and backward is 2;
5. and obtaining the vehicle defect description key information.
Because different types of license plates have different purposes, the formats of the license plates can be different, and the license plate types are determined by the license plate colors, the color recognition is an important step of license plate recognition, and is also one of the key factors for determining whether the license plate recognition is correct, and the improvement of the recognition accuracy rate of the license plate recognition is beneficial to the improvement of the performance of the final license plate recognition algorithm. And identifying the license plate of the vehicle at the specified angle, and automatically blurring the identified license plate, thereby protecting the privacy of the user.
Picture recognition refers to a technique of processing, analyzing, and understanding an image with a computer to recognize objects and objects of various modes. At present, the scheme of image recognition in the industry mainly comprises license plate number recognition smearing or manual shielding of license plates, so that the privacy of users is protected.
The license plate blurring process provided by the embodiment of the invention comprises the following steps: and extracting the picture content in the defect preliminary detection report, positioning an image area where the license plate is located in the picture content, and blurring the image area where the license plate is located. The image area positioning of the license plate mainly uses a lightweight deep convolutional neural network to extract high-level features of the license plate in the image, coordinate points and rotation angles of rectangles at the position of the license plate are obtained through forward calculation, the image area of the license plate is accurately framed, and the model positioning license plate accuracy is improved by using a mode of fusion of the shallow features and the deep features. After the image area where the license plate is located is obtained, a mean value filter is used for realizing license plate blurring.
The cloud can quickly identify style information of the vehicle by utilizing the vehicle type database according to the vehicle defect description key information, then confirm vehicle configuration information, and also can inquire a vehicle history maintenance record and an insurance record through a background maintenance inquiry interface so as to avoid meter adjustment and accident vehicles. And the evaluator identifies the basic photos and the procedure photos of the vehicle, if the photographed photos are not clear, the photos are refused to an inspector, re-inspection is required, and the authenticity and the effectiveness of the vehicle inspection data are ensured. The cloud comprehensively examines the vehicle conditions, synthesizes various detection data, comprehensively ranks and scores the vehicles, and provides a vehicle assessment report in combination with the license plate blurring process. Subsequently, an online transaction may be conducted on the vehicle based on the vehicle assessment report.
A large number of semantic rule codes of the second-hand vehicle condition are made in the cloud background, so that the accuracy of detecting the picture mark detection problem is up to more than 95% in the vehicle condition detection process
In summary, the embodiment of the invention has the data learning function by presetting the vehicle defect semantic recognition rule base, so that the more used inspectors, the more accurate the data, the greatly simplified the threshold of the existing second-hand vehicle industry detection, and the detection capability of the inspectors is improved. Through designing and utilizing the semantic matching rule corresponding to the detection target, the non-standardized voice and text input of the inspector is realized, the voice and text input is automatically converted into standard vehicle condition information, the method has extremely high flexibility and convenience, the inspector is not required to shoot all vehicle condition photos according to the detection requirement, and only the vehicle condition found out to be shot for remarks is required.
The invention has higher recognition capability for the colors, angles and scenes of the license plates by utilizing the license plate blurring technology, supports automatic smearing, reduces the detection cost of detection personnel, and improves the detection efficiency, which is why we want to protect.
The remote automobile detection method provided by the embodiment of the invention supports different detection personnel, detects the second-hand automobile in different regions and scenes and generates a unified detection report, thereby greatly reducing the training cost of the detection personnel of enterprises, expanding the service range of the detection per se and improving the detection efficiency and quality. The detection content can be customized in a set of tools according to the requirements of clients and business requirements, and a universal detection report is generated.
The remote automobile detection method provided by the embodiment of the invention can provide a detection evaluation model for automobile detection based on the automobile model database data, paint film data and historical evaluation data of more than 4 ten thousand models of automobile easily shot, and can quickly and accurately generate a detection report.
Those of ordinary skill in the art will appreciate that: the drawing is a schematic diagram of one embodiment and the modules or flows in the drawing are not necessarily required to practice the invention.
From the above description of embodiments, it will be apparent to those skilled in the art that the present invention may be implemented in software plus a necessary general hardware platform. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a storage medium, such as a ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the embodiments or some parts of the embodiments of the present invention.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for apparatus or system embodiments, since they are substantially similar to method embodiments, the description is relatively simple, with reference to the description of method embodiments in part. The apparatus and system embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
The present invention is not limited to the above-mentioned embodiments, and any changes or substitutions that can be easily understood by those skilled in the art within the technical scope of the present invention are intended to be included in the scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.
Claims (6)
1. The automobile remote detection method is characterized in that a detection target of a vehicle and a semantic matching rule corresponding to the detection target are stored in a vehicle defect semantic recognition rule base, and the method comprises the following steps:
determining a detection item according to detection requirements, acquiring a detection target corresponding to the detection item, detecting the detection target of the vehicle through a detection tool, and submitting a preliminary defect detection report of the detected vehicle to a background server;
the background server performs semantic analysis on text content in the defect preliminary detection report by utilizing semantic matching rules corresponding to detection targets stored in the vehicle defect semantic recognition rule base to obtain vehicle defect description key information, and obtains a vehicle evaluation report based on the vehicle defect description key information, and the method specifically comprises the following steps:
after receiving the defect preliminary detection report of the vehicle, the background server extracts text content in the defect preliminary detection report, and performs semantic analysis on the text content by utilizing semantic matching rules to obtain vehicle defect description key information;
the background server extracts the picture content in the defect preliminary detection report, locates an image area where the license plate is located in the picture content, and performs blurring on the image area where the license plate is located;
the background server utilizes a vehicle model database to identify the style information of the vehicle according to the vehicle defect description key information, confirms the vehicle configuration information, inquires the vehicle history maintenance record and the insurance record through a maintenance inquiry interface of the background, synthesizes various detection data, and combines the vehicle defect description key information and a license plate blurring result to obtain a vehicle evaluation report;
the semantic analysis of the text content by using the semantic matching rule comprises the following processing steps:
step_1, performing Chinese special character processing on the text content, and converting full-angle characters in the text content into half-angle characters;
step_2, dividing the English number by the English number; sentence division is carried out on the text content, and a sentence set to be processed is obtained;
step 3, performing word segmentation processing on each sentence to obtain a vocabulary sequence of each sentence;
step 4, traversing the vocabulary sequence of the sentence, and replacing the mapping words with corresponding mapping content sets to obtain a set of mapped vocabulary sequences;
step_5, recombining the mapped vocabulary sequences to obtain a recombined sentence set, and segmenting the recombined sentences to obtain a new vocabulary sequence set;
step_6, mapping each vocabulary in the vocabulary sequence into a corresponding detection target stored in a vehicle defect semantic recognition rule base, and carrying out rule matching on each vocabulary sequence according to a semantic matching rule corresponding to the detection target stored in the vehicle defect semantic recognition rule base to obtain candidate vehicle defect detection information;
step 7, extracting vehicle defect description key information from the candidate vehicle defect detection information set by using a key information extraction flow;
the key information extraction process comprises the following steps:
1. traversing a vocabulary sequence in the candidate vehicle defect detection information set;
2. finding a vocabulary C marked as a component;
3. traversing the online vocabulary forwards and backwards respectively based on the vocabulary C, traversing the window size of the context to be 5, and searching the vocabulary marked as position and description so as to form a key information combination;
4. when traversing the context, the maximum number of position and depth vocabularies found forward and backward is 2;
5. and obtaining the vehicle defect description key information.
2. The method of claim 1, wherein the method further comprises:
the method comprises the steps that firstly, a detector registers in a detection mechanism, identity information of the detector is submitted to the detection mechanism, the detection mechanism transmits the received identity information of the detector to a background server, after the background server checks that the identity information of the detector passes, an identity checking passing message of the detector is returned to the detection mechanism, the detector registers successfully in the detection mechanism, and the detection mechanism stores and manages the identity information of the detector.
3. The method according to claim 1, wherein storing the detection target of the vehicle and the semantic matching rule corresponding to the detection target in the vehicle defect semantic recognition rule base includes:
a vehicle defect semantic recognition rule base is managed in a background server, and all detection targets of the vehicle, all possible detection results corresponding to each detection target and semantic matching rules corresponding to each detection target are stored in the vehicle defect semantic recognition rule base, wherein the semantic matching rules comprise vehicle condition semantic codes represented by voice, pictures, mathematical symbols, marks and/or texts.
4. A method according to claim 3, wherein said all detection targets comprise: left front tire, left front hub, left front fender, left A pillar outer, left front door, left outside rear view mirror, left front door edge seal, left bottom large edge, left front door sill, left B pillar outer, left C pillar outer, left rear door edge seal, left rear door sill, left rear tire, left rear hub, fuel tank cap, left rear fender, front bumper, grille, right headlamp assembly, left front fog lamp, a right front fog light, a front windshield wiper, a roof, a rear windshield, a trunk lid edge, a rear windshield wiper, a right front hub, a right front tire, a right front fender, a right A pillar outer, a right outside rear view mirror, a right front door seal, a right bottom edge, a right front door sill, a right B pillar outer, a right rear door, a right C pillar outer, a right rear door seal, a right rear door sill, a right rear fender, a right rear hub, a right rear tire, a trunk, a left front door trim, a left A pillar trim, a left rear door trim, a left B pillar trim, a left C pillar trim, a steering wheel, a workbench, a front air outlet, a glove box, an instrument, a center console, a handle bar, a boot, a armrest box, a main drive seat, a secondary drive seat, a left rear seat, a right A pillar trim, a right rear door trim, a right B pillar trim, a right C pillar trim, a front bumper beam, a left front shock absorber mount, a removable water tank frame, a left front headlamp frame, a water tank upper bracket, a water tank lower bracket, a left front fender liner, a left front side rail, a left front A pillar inner side and edge seal, a left B pillar inner side and edge seal, a left C pillar inner side and front perimeter baffle, a rear bumper beam, a rear dash panel, a left rear bumper mount, a left tail lamp frame, a vehicle body floor, the front bumper comprises a luggage case sealing edge, a luggage case partition board, a left rear longitudinal beam, a right front shock absorber seat, a right headlamp frame, a right front fender liner, a right front longitudinal beam, a right B column inner side and sealing edge, a right C column inner side and sealing edge, a right A column inner side and sealing edge, a right rear shock absorber seat, a right tail lamp frame, a right rear longitudinal beam, a left front door handle, a left rear door handle, a right front door handle, a right rear door handle, a front fog lamp, a rear fog lamp, a left front wheel arch, a left rear wheel arch, a right front wheel arch, a right rear wheel arch, a front bumper lower decorative plate, a rear bumper lower decorative plate, a front fog lamp grille and a rear fog lamp grille.
5. The method according to claim 3 or 4, wherein the determining the detection item according to the detection requirement, obtaining the detection target corresponding to the detection item, detecting the detection target of the vehicle by the detection tool, and submitting the detected defect preliminary detection report of the vehicle to the background server includes:
a detection configuration database is managed in a background server, each detection item on the computer application program APP is configured in the detection configuration database according to the service requirement, and each detection item comprises a plurality of detection targets;
firstly, a inspector determines an inspection item according to actual inspection requirements, inquires an inspection configuration database according to the inspection item to obtain an inspection target corresponding to the inspection item, the inspector utilizes an APP to inspect the inspection target of the vehicle through various inspection tools, and the inspection target is inspected through voice, pictures and/or text description to obtain an inspection preliminary inspection report of the vehicle, and the inspection preliminary inspection report of the vehicle is submitted to a background server.
6. The method of claim 1, wherein the background server extracts the picture content in the defect preliminary detection report, locates an image area where the license plate is located in the picture content, and performs blurring on the image area where the license plate is located, including:
the background server extracts picture content in a defect preliminary detection report, a deep convolutional neural network is used for extracting high-level features of license plates in images, a rectangular coordinate point and a rotation angle of the position of the license plates are obtained through forward calculation in a mode of fusion of shallow features and deep features, an image area of the license plates is framed according to the rectangular coordinate point and the rotation angle, and license plate blurring processing is carried out on the image area of the license plates by means of a mean value filtering process.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811425532.4A CN109597991B (en) | 2018-11-27 | 2018-11-27 | Automobile remote detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811425532.4A CN109597991B (en) | 2018-11-27 | 2018-11-27 | Automobile remote detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109597991A CN109597991A (en) | 2019-04-09 |
| CN109597991B true CN109597991B (en) | 2023-04-28 |
Family
ID=65960473
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811425532.4A Active CN109597991B (en) | 2018-11-27 | 2018-11-27 | Automobile remote detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109597991B (en) |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110610007A (en) * | 2019-09-20 | 2019-12-24 | 广州穗圣信息科技有限公司 | Maintenance vehicle condition intelligent identification method and device based on NLP |
| CN110889593A (en) * | 2019-11-01 | 2020-03-17 | 浙江大搜车软件技术有限公司 | Vehicle evaluation method, device, computer-readable storage medium and computer equipment |
| CN112102838A (en) * | 2020-03-04 | 2020-12-18 | 浙江大搜车软件技术有限公司 | Vehicle detection report generation method and system and electronic equipment |
| CN111626690A (en) * | 2020-04-23 | 2020-09-04 | 成都检车家汽车服务有限公司 | Vehicle detection system and method |
| CN111553285B (en) * | 2020-04-29 | 2022-10-04 | 上海智勘科技有限公司 | Method and system for protecting user privacy information in vehicle remote damage assessment process |
| CN111626815B (en) * | 2020-05-07 | 2023-09-12 | 成都检车家汽车服务有限公司 | Processing system and method for vehicle detection order |
| CN112785561A (en) * | 2021-01-07 | 2021-05-11 | 天津狮拓信息技术有限公司 | Second-hand commercial vehicle condition detection method based on improved Faster RCNN prediction model |
| CN113254113A (en) * | 2021-05-24 | 2021-08-13 | 国家计算机网络与信息安全管理中心 | Vehicle safety defect verification method and system |
| CN113487047B (en) * | 2021-07-01 | 2022-11-08 | 石家庄华燕交通科技有限公司 | Intelligent guiding system for motor vehicle inspection business |
| CN114279718B (en) * | 2021-11-30 | 2023-02-28 | 重庆长安汽车股份有限公司 | Configurable fault diagnosis method and system based on real-time vehicle condition |
| CN114584584A (en) * | 2022-02-28 | 2022-06-03 | 福思(杭州)智能科技有限公司 | System and method for processing vehicle driving data and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101853478A (en) * | 2009-04-01 | 2010-10-06 | 杭州信雅达科技有限公司 | Automobile fault diagnosis system based on ontology and mobile network |
| CN105610850A (en) * | 2016-01-13 | 2016-05-25 | 天津大学 | Vehicle service and management system based on internet of vehicles |
| CN106169098A (en) * | 2016-06-28 | 2016-11-30 | 江苏车置宝信息科技股份有限公司 | Used car vehicle condition intelligent checking system and method |
| CN106708017A (en) * | 2016-12-26 | 2017-05-24 | 华东理工大学 | Vehicle fault real-time monitoring and service system and application thereof |
| CN106971322A (en) * | 2017-05-17 | 2017-07-21 | 因车（上海）企业管理有限公司 | A kind of mirror valency system and its mirror valency method for used car remote evaluation |
| JP2018097540A (en) * | 2016-12-12 | 2018-06-21 | 三菱自動車工業株式会社 | Positional information reporting device for defective vehicle |
| CN108364653A (en) * | 2018-02-12 | 2018-08-03 | 王磊 | Voice data processing method and processing unit |
-
2018
- 2018-11-27 CN CN201811425532.4A patent/CN109597991B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101853478A (en) * | 2009-04-01 | 2010-10-06 | 杭州信雅达科技有限公司 | Automobile fault diagnosis system based on ontology and mobile network |
| CN105610850A (en) * | 2016-01-13 | 2016-05-25 | 天津大学 | Vehicle service and management system based on internet of vehicles |
| CN106169098A (en) * | 2016-06-28 | 2016-11-30 | 江苏车置宝信息科技股份有限公司 | Used car vehicle condition intelligent checking system and method |
| JP2018097540A (en) * | 2016-12-12 | 2018-06-21 | 三菱自動車工業株式会社 | Positional information reporting device for defective vehicle |
| CN106708017A (en) * | 2016-12-26 | 2017-05-24 | 华东理工大学 | Vehicle fault real-time monitoring and service system and application thereof |
| CN106971322A (en) * | 2017-05-17 | 2017-07-21 | 因车（上海）企业管理有限公司 | A kind of mirror valency system and its mirror valency method for used car remote evaluation |
| CN108364653A (en) * | 2018-02-12 | 2018-08-03 | 王磊 | Voice data processing method and processing unit |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109597991A (en) | 2019-04-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109597991B (en) | Automobile remote detection method | |
| CN110378236B (en) | Vehicle identity recognition model construction and recognition method and system based on deep learning | |
| CN111488875B (en) | Vehicle insurance claim settlement loss checking method and device based on image recognition and electronic equipment | |
| CN110569701B (en) | Computer-implemented vehicle damage assessment method and device | |
| US9082038B2 (en) | Dram c adjustment of automatic license plate recognition processing based on vehicle class information | |
| US9721302B2 (en) | Server for real-time accident documentation and claim submission | |
| US9082037B2 (en) | Method and system for automatically determining the issuing state of a license plate | |
| US9400936B2 (en) | Methods and systems for vehicle tag number recognition | |
| DE102011117585B4 (en) | Systems and methods for tracking objects | |
| WO2021136941A1 (en) | Method of determining inconsistent damage to parts of a vehicle | |
| JP2016058079A (en) | Domain adaptation for image classification using class prior probability | |
| US9792301B2 (en) | Multi-query privacy-preserving parking management system and method | |
| US20160239922A1 (en) | System and method of real-time imaging and analysis of real-world objects | |
| CN110570316A (en) | method and device for training damage recognition model | |
| CN110298257B (en) | Driver behavior recognition method based on human body multi-part characteristics | |
| CN106651881B (en) | Vehicle inspection system, vehicle part recognition method and system | |
| KR102105954B1 (en) | System and method for accident risk detection | |
| US20220129860A1 (en) | Artificial Intelligence Based Determination Of Damage To Physical Structures Via Video | |
| CN115810134A (en) | Image acquisition quality inspection method, system and device for preventing car insurance from cheating | |
| CN113705351B (en) | Vehicle damage assessment method, device and equipment | |
| US20200104940A1 (en) | Artificial intelligence enabled assessment of damage to automobiles | |
| CN112580566A (en) | Vehicle type identification method and electrophoresis process vehicle type matching method and system thereof | |
| US11989787B2 (en) | Remote vehicle damage assessment | |
| US20220036132A1 (en) | Semantic image segmentation for cognitive analysis of physical structures | |
| CN112818865A (en) | Vehicle-mounted field image identification method, identification model establishing method, device, electronic equipment and readable storage medium |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |