CN111666789A - One-car multi-license plate analysis method and system based on image searching - Google Patents
One-car multi-license plate analysis method and system based on image searching Download PDFInfo
- Publication number
- CN111666789A CN111666789A CN201910170832.0A CN201910170832A CN111666789A CN 111666789 A CN111666789 A CN 111666789A CN 201910170832 A CN201910170832 A CN 201910170832A CN 111666789 A CN111666789 A CN 111666789A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- picture
- application server
- analysis method
- characteristic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
- G06V20/584—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of vehicle lights or traffic lights
Abstract
The embodiment of the application provides a one-car multi-brand analysis method based on searching a map by using the map, which comprises the following steps of: vehicle picture acquisition: the method comprises the steps that vehicle pictures are collected through snapshot machines arranged at all gates and uploaded to a vehicle gateway service through a network; a characteristic extraction step: carrying out feature identification on the vehicle picture in the gateway service, and then inputting feature data formed by features into a database; client condition selection: the user sends a retrieval request to the application server through the retrieval option; and an application server retrieval step: the application server calls the characteristic data in the database, and the characteristic data is searched and then is compared in groups to obtain a one-car multi-license analysis result, so that the technical problems that the efficiency is not high and the accuracy is not high due to the fact that a large number of vehicle characteristic pictures are compared and then grouped in the existing one-car multi-license analysis method are solved.
Description
Technical Field
The embodiment of the application relates to the field of image processing technology and computer vision, in particular to a one-car multi-card analysis method and system based on image searching.
Background
In the existing one-car multi-license analysis, vehicles are grouped in a mass of bayonet pictures through information such as vehicle characteristics, each group is considered as the same vehicle, then whether the license plate numbers in the current group are the same or not is judged in each group, and if the license plate numbers are different, the current vehicle is considered to have a plurality of license plates.
At present, no one-car multi-brand analysis implementation scheme based on image searching is available in an efficient and accurate mode.
Above based on current a car many boards analysis method, because need compare earlier then grouping to magnanimity vehicle characteristic picture, so efficiency is not high, and this kind of mode of vehicle characteristic grouping itself has the limitation, leads to the not high technical problem of degree of accuracy.
Disclosure of Invention
The embodiment of the application provides a one-car multi-license plate analysis method and system based on a map search, and solves the technical problems that the existing one-car multi-license plate analysis method is low in efficiency because massive vehicle characteristic pictures are compared and then grouped, and the accuracy is low due to the limitation of the vehicle characteristic grouping mode.
The embodiment of the application provides a one-car multi-brand analysis method based on searching a map, which is characterized by comprising the following steps of:
vehicle picture acquisition: the method comprises the steps that vehicle pictures are collected through snapshot machines arranged at all gates and uploaded to a vehicle gateway service through a network;
a characteristic extraction step: carrying out feature identification on the vehicle picture in the gateway service, and then inputting feature data formed by the features into a database;
client condition selection: the user sends a retrieval request to the application server through the retrieval option;
and an application server retrieval step: and the application server calls the characteristic data in the database, and performs group comparison on the characteristic data after retrieval to obtain a one-car multi-license analysis result.
Specifically, the feature recognition is performed on the vehicle picture in the gateway service, specifically:
and carrying out vehicle attribute identification, license plate characteristic identification and picture characteristic identification on the vehicle picture to acquire vehicle attribute information, license plate characteristic information and picture characteristic information.
Preferably, the vehicle characteristic information comprises hanging decorations, a sun shield, a tissue box and an annual inspection mark.
Preferably, the retrieval options include any one or more of time period, passing card point, license plate brand, vehicle type, vehicle body color and vehicle sub-brand.
Preferably, the application server is configured to process and display the retrieval result.
Preferably, the grouping is specifically performed according to similarity thresholds of the retrieved hanging decorations, the sun shield, the tissue box, the annual inspection mark and the picture characteristics of the feature data.
Preferably, each group in the group is regarded as the same vehicle, the comparison specifically is to compare data of each group, and if a plurality of license plate numbers exist in a group of data, it is regarded that the record of the current group meets the condition of one vehicle with multiple license plates.
Preferably, the retrieval is a big data retrieval cluster.
A one-car multi-brand analysis system based on picture searching is characterized by comprising a snapshot machine, a gateway server, a database, a client and an application server;
the snapshot machine is arranged at a gate to collect a vehicle picture and transmit the vehicle picture to the gateway server through a network;
the gateway server is used for calling an algorithm to identify the characteristics in the picture and storing the identified characteristic information into a database;
the database is used for storing the characteristic information;
the client is used for sending a retrieval request to the application server;
and the application server is used for processing the retrieved result and displaying the analysis result.
Preferably, the algorithm is an SDK algorithm.
According to the technical scheme, the embodiment of the application has the following advantages:
the embodiment of the application provides a one-car multi-brand analysis method based on searching a map, which is characterized by comprising the following steps of: vehicle picture acquisition: the method comprises the steps that vehicle pictures are collected through snapshot machines arranged at all gates and uploaded to a vehicle gateway service through a network; a characteristic extraction step: carrying out feature identification on the vehicle picture in the gateway service, and then inputting feature data formed by the features into a database; client condition selection: the user sends a retrieval request to the application server through the retrieval option; and an application server retrieval step: and the application server calls the characteristic data in the database, and performs group comparison on the characteristic data after retrieval to obtain a one-car multi-license analysis result. The method solves the technical problems that the efficiency is not high because massive vehicle characteristic pictures are compared and then grouped based on the existing one-vehicle multi-plate analysis method, and the accuracy is not high due to the limitation of the vehicle characteristic grouping mode.
Drawings
FIG. 1 is a schematic flow chart illustrating an embodiment of a one-car multi-brand analysis method based on a chart search in an embodiment of the present application;
fig. 2 is a schematic diagram of a one-car multi-brand analysis system based on a chart search in an embodiment of the present application.
Reference numerals:
a one-car multi-brand analysis system 1 based on searching pictures with pictures;
a snapshot machine 11;
a gateway server 12;
a database 13;
an application server 14;
a client terminal 15.
Detailed Description
The embodiment of the application provides a one-car multi-license plate analysis method and system based on a map search, and solves the technical problems that the existing one-car multi-license plate analysis method is low in efficiency because massive vehicle characteristic pictures are compared and then grouped, and the accuracy is low due to the limitation of the vehicle characteristic grouping mode.
In order to make the objects, features and advantages of the embodiments of the present application more obvious and understandable, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the embodiments described below are only a part of the embodiments of the present application, but not all the embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments in the present application without any creative effort belong to the protection scope of the embodiments in the present application.
Referring to fig. 1, an embodiment of a method for analyzing a plurality of cards based on a map search according to an embodiment of the present application includes the following steps:
vehicle picture acquisition: the vehicle pictures are collected through the snapshot machines 11 arranged at the bayonets and uploaded to a vehicle gateway service through a network;
a characteristic extraction step: carrying out feature identification on the vehicle picture in the gateway service, and then inputting feature data formed by the features into a database;
client condition selection: the user sends a retrieval request to the application server 14 via the retrieval option;
and an application server retrieval step: the application server 14 calls the feature data in the database 13, and performs group comparison on the feature data after retrieval to obtain a one-car multi-brand analysis result.
Specifically, the feature recognition is performed on the vehicle picture in the gateway service, specifically:
and carrying out vehicle attribute identification, license plate characteristic identification and picture characteristic identification on the vehicle picture to acquire vehicle attribute information, license plate characteristic information and picture characteristic information.
Preferably, the vehicle characteristic information comprises hanging decorations, a sun shield, a tissue box and an annual inspection mark.
Preferably, the retrieval options include any one or more of time period, passing card point, license plate brand, vehicle type, vehicle body color and vehicle sub-brand.
Preferably, the application server 14 is configured to process and display the retrieval result.
Preferably, the grouping is specifically performed according to similarity thresholds of the retrieved hanging decorations, the sun shield, the tissue box, the annual inspection mark and the picture characteristics of the feature data.
Preferably, each group in the group is regarded as the same vehicle, the comparison specifically is to compare data of each group, and if a plurality of license plate numbers exist in a group of data, it is regarded that the record of the current group meets the condition of one vehicle with multiple license plates.
Preferably, the retrieval is a big data retrieval cluster.
The method comprises the steps of inquiring vehicle passing records meeting requirements from massive vehicle passing records according to conditions of vehicle bayonet equipment to be inquired, starting and ending time, vehicle brands, types, vehicle body colors, sub-brands, hanging decorations, sun shields, tissue boxes, annual inspection marks and the like, then grouping a small number of inquiry results, considering that each group is the same vehicle, analyzing each group, and considering that the current vehicle is a vehicle with multiple plates if multiple plates exist in the group. The method has the advantages that massive data are filtered through the vehicle attribute conditions, then a small amount of data are analyzed, so that the analysis efficiency is greatly improved, and the accuracy can also be greatly improved through analyzing conditions such as vehicle brands, types, body colors, sub-brands, hanging decorations, sun shields, tissue boxes, annual inspection marks and the like of each vehicle.
One embodiment of the present application is specifically:
the system comprises an algorithm service for providing vehicle attributes, license plate characteristics and picture characteristics from a vehicle picture.
The system comprises a big data retrieval cluster which is used for retrieving useful information from mass vehicle attributes, license plate characteristics and picture characteristics.
And the gateway service is used for calling an algorithm SDK to identify the characteristics in the picture and storing the identified information into the big data and the database.
An application server 14 is included for processing and displaying the retrieved results.
The client 15 interacts with the application server 14, the application server 14 interacts with the big data retrieval service, and the gateway service interacts with the big data service to implement business functions.
In one embodiment, the one-car multi-brand analysis method based on searching the map includes the following steps:
vehicle picture acquisition: the vehicle pictures are collected by the snapshoters 11 of the bayonets and then uploaded to the vehicle gateway service through the network.
A characteristic extraction step: the gateway service identifies features in the picture by invoking the algorithm SDK and then feeds the data into the big data cluster and database 13.
Client condition selection: the user sends a request to the application server 14 by selecting a multi-brand time slot for a car to be retrieved, by passing conditions such as a stuck point, a brand of the car plate, a type of the car, a color of the car body, a sub-brand of the car, and the like, and then clicking on the retrieval.
And an application server retrieval step: after receiving the request, the application server 14 calls an interface of big data retrieval to put the conditions selected by the user into the interface, the big data retrieval cluster firstly retrieves a small amount of data meeting requirements from mass data through time periods, character conditions such as card points, license plate brands, vehicle types, vehicle body colors, vehicle sub-brands and the like, then compares and groups hanging decorations, sun shields, tissue boxes, annual inspection marks, picture feature similarity threshold values and the like of the retrieved data, each group is considered as the same vehicle, then compares the data of each group, and if a plurality of license plate numbers exist in one group of data, the record of the current group is considered to meet the condition of one vehicle with a plurality of license plates. And finally returns the grouped data corresponding to the license plate of the vehicle to the application server 14.
The embodiment of the application has the following advantages: because before carrying out the picture comparison based on searching for the picture with the picture, at first filtered most data through the characters condition, so retrieval efficiency has obtained very big promotion, when carrying out a car many boards analysis to the data that satisfies the requirement, through hanging decorations, sunshading board, paper handkerchief box, annual inspection sign these vehicle marks carry out the comparison, then compare picture characteristic similarity threshold, after having carried out double comparison like this, can be more accurate group the same vehicle for the record degree of accuracy of retrieval is higher.
A one-car multi-brand analysis system 1 based on searching pictures is characterized by comprising a snapshot machine 11, a gateway server 12, a database 13, a client 15 and an application server 14;
the snapshot machine 11 is configured to be arranged at a gate to collect a vehicle picture, and transmit the vehicle picture to the gateway server 12 through a network;
the gateway server 12 is used for calling an algorithm to identify the features in the picture and storing the identified feature information into the database 13;
the database 13 is used for storing the characteristic information;
the client 15 is configured to send a retrieval request to the application server 14;
and the application server 14 is used for processing the retrieved result and displaying the analysis result.
In a more preferred embodiment, the algorithm is an SDK algorithm.
In the embodiment of the application, the text query and the filtration are performed on the mass data according to the passing of the bayonet, the time period, the vehicle brand, the vehicle type, the vehicle body color and the sub-brand, the range of searching the images with the images is reduced to the maximum extent, and then the image searching with the images is performed for comparison, so that the comparison efficiency is higher than that of the conventional method in which all images are searched with the images. And data are grouped according to the brand of the vehicle, the type of the vehicle, the color of the vehicle body, the sub-brand, the hanging decoration, the sun shield, the tissue box and the annual inspection mark, then the same vehicle is grouped according to whether the similarity threshold value reaches a certain value, and the grouping is more accurate compared with the traditional method that the threshold value is completely compared by searching the image through the image.
The above embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions in the embodiments of the present application.
Claims (10)
1. A one-car multi-brand analysis method based on searching pictures is characterized by comprising the following steps:
vehicle picture acquisition: the method comprises the steps that vehicle pictures are collected through snapshot machines arranged at all gates and uploaded to a vehicle gateway service through a network;
a characteristic extraction step: carrying out feature identification on the vehicle picture in the gateway service, and then inputting feature data formed by the features into a database;
client condition selection: the user sends a retrieval request to the application server through the retrieval option;
and an application server retrieval step: and the application server calls the characteristic data in the database, and performs group comparison on the characteristic data after retrieval to obtain a one-car multi-license analysis result.
2. The vehicle multi-plate analysis method based on map search as claimed in claim 1, wherein the vehicle image in the gateway service is subjected to feature recognition, specifically:
and carrying out vehicle attribute identification, license plate characteristic identification and picture characteristic identification on the vehicle picture to acquire vehicle attribute information, license plate characteristic information and picture characteristic information.
3. The one-car multi-brand analysis method based on searching charts in the claim 2, wherein the car characteristic information comprises hanging decorations, sun visors, tissue boxes and annual inspection marks.
4. The vehicle multi-plate analysis method based on map search according to claim 1, wherein the search options comprise any one or more of time period, passing card point, brand name of license plate, type of vehicle, color of vehicle body, and sub-brand of vehicle.
5. The vehicle multi-plate analysis method based on graph search as claimed in claim 1, wherein the application server is configured to process and display the search result.
6. The vehicle multi-plate analysis method based on image searching as claimed in claim 1, wherein the grouping is performed according to similarity thresholds of the retrieved feature data, such as hanging decorations, sun visors, tissue boxes, annual inspection marks and image features.
7. The vehicle multi-plate analysis method based on map search as claimed in claim 1, wherein each group in the group is considered as the same vehicle, the comparison is specifically performed on the data of each group, and if there are multiple plate numbers in a group of data, the record of the current group is considered to satisfy the vehicle multi-plate condition.
8. The vehicle multi-board analysis method based on graph search as claimed in claim 1, wherein the search is a big data search cluster.
9. A one-car multi-brand analysis system based on picture searching is characterized by comprising a snapshot machine, a gateway server, a database, a client and an application server;
the snapshot machine is arranged at a gate to collect a vehicle picture and transmit the vehicle picture to the gateway server through a network;
the gateway server is used for calling an algorithm to identify the characteristics in the picture and storing the identified characteristic information into a database;
the database is used for storing the characteristic information;
the client is used for sending a retrieval request to the application server;
and the application server is used for processing the retrieved result and displaying the analysis result.
10. The vehicle multi-board analysis system based on searching map according to claim 9, wherein the algorithm is SDK algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910170832.0A CN111666789A (en) | 2019-03-07 | 2019-03-07 | One-car multi-license plate analysis method and system based on image searching |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910170832.0A CN111666789A (en) | 2019-03-07 | 2019-03-07 | One-car multi-license plate analysis method and system based on image searching |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111666789A true CN111666789A (en) | 2020-09-15 |
Family
ID=72381926
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910170832.0A Pending CN111666789A (en) | 2019-03-07 | 2019-03-07 | One-car multi-license plate analysis method and system based on image searching |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111666789A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105808691A (en) * | 2016-03-03 | 2016-07-27 | 博康智能网络科技股份有限公司 | Gate vehicle retrieval method and system |
| CN108595576A (en) * | 2018-04-17 | 2018-09-28 | 高新兴科技集团股份有限公司 | It is a kind of based on the vehicle of database to scheme to search drawing method |
| CN108984644A (en) * | 2018-06-25 | 2018-12-11 | 高新兴科技集团股份有限公司 | A kind of fake-licensed car search method and system |
| CN109325420A (en) * | 2018-08-27 | 2019-02-12 | 广州烽火众智数字技术有限公司 | A kind of vehicle characteristics search method and system based on big data |
-
2019
- 2019-03-07 CN CN201910170832.0A patent/CN111666789A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105808691A (en) * | 2016-03-03 | 2016-07-27 | 博康智能网络科技股份有限公司 | Gate vehicle retrieval method and system |
| CN108595576A (en) * | 2018-04-17 | 2018-09-28 | 高新兴科技集团股份有限公司 | It is a kind of based on the vehicle of database to scheme to search drawing method |
| CN108984644A (en) * | 2018-06-25 | 2018-12-11 | 高新兴科技集团股份有限公司 | A kind of fake-licensed car search method and system |
| CN109325420A (en) * | 2018-08-27 | 2019-02-12 | 广州烽火众智数字技术有限公司 | A kind of vehicle characteristics search method and system based on big data |
Non-Patent Citations (1)
| Title |
|---|
| BOOSTIV: ""车辆识别"，不仅仅是车牌识别！", 《WWW.BOOSTIV.COM》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105677728A (en) | Object image recognition and classification managing method | |
| US8170343B2 (en) | Method and system for searching images with figures and recording medium storing metadata of image | |
| CN103679147A (en) | Method and device for identifying model of mobile phone | |
| WO2013119804A1 (en) | Color-based identification, searching and matching enhancement of supply chain and inventory management systems | |
| CN111625687B (en) | Method and system for quickly searching people in media asset video library through human faces | |
| CN108446717A (en) | A kind of board state collection method and system based on image recognition | |
| CN108427959A (en) | Board state collection method based on image recognition and system | |
| CN107391506A (en) | Method and apparatus for inquiring about data | |
| CN109711298A (en) | The method and system of efficient face characteristic value retrieval based on faiss | |
| CN110727756A (en) | Management method and device of space-time trajectory data | |
| CN105930313A (en) | Method and device for processing notification message | |
| CN114201640A (en) | Garden management method and system based on video acquisition and related equipment | |
| CN114036318A (en) | Public safety knowledge graph generation method facing to pan-aware data and based on time sequence diagram | |
| CN116484109B (en) | Customer portrait analysis system and method based on artificial intelligence | |
| CN106649829A (en) | Method and device for processing business based on palmprint data | |
| CN112417195A (en) | Trademark inquiry system and method based on mobile terminal and storage medium | |
| CN111401438A (en) | Image sorting method, device and system | |
| CN111666789A (en) | One-car multi-license plate analysis method and system based on image searching | |
| CN110796014A (en) | Garbage throwing habit analysis method, system and device and storage medium | |
| CN112579539B (en) | Management method and system for enterprise cluster big data | |
| CN114971432A (en) | Enterprise financial risk early warning method and system | |
| CN114491049A (en) | Office system asset allocation method based on information management | |
| CN111475657B (en) | Display equipment, display system and entity alignment method | |
| CN112446673A (en) | Trademark change judgment method, system, equipment and readable storage medium | |
| CN111611322A (en) | User information correlation method and system |
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 |