Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Vehicle information extraction method and system thereof

A technology for vehicle information and extraction methods, applied in character and pattern recognition, image data processing, instruments, etc., can solve problems that are susceptible to interference, cannot handle vehicles without license plates or vehicles whose license plates are not placed in standard positions, and affect the scope of application. and practical effects

Active Publication Date: 2014-09-24
HANGZHOU KEDU TECH
View PDF5 Cites 31 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] There are some problems in extracting vehicle brand information through the method of vehicle logo: 1) The car logo is generally small, so there are high requirements for the size and clarity of the vehicle area; The detection method cannot handle the changed or damaged or missing pictures of the car logo in the real situation; 3) also because the amount of information on the car logo is too small, the car logo of many vehicles cannot be recognized with high quality, and the false positive rate is very high. high
The problem with this method is firstly destructive installation; secondly, it is easily disturbed and the recognition rate is low; thirdly, it is physically contacted, consumes a lot and is easily damaged
There are some serious problems in this method, which will affect the scope of application and actual effect of the method
This includes: 1) Its brand recognition performance is highly dependent on the precise positioning of vehicle license plates, which cannot deal with vehicles without license plates or vehicles whose license plates are not placed in the standard position, and these situations often appear in criminal cases involving vehicles ; 2) Since it uses multiple vehicle component modules to form a feature vector, and the positioning of multiple modules also depends on the relationship between the relative license plate, it also cannot deal with unlicensed vehicles or vehicles whose license plate is not placed in the standard position; 3 ) It uses SIFT feature matching when correcting the position of the license plate. This method is slow and will affect the overall recognition speed

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Vehicle information extraction method and system thereof
  • Vehicle information extraction method and system thereof
  • Vehicle information extraction method and system thereof

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0072] The first aspect of the present invention relates to a method for extracting vehicle information, such as figure 1 shown, including the following steps:

[0073] S1. Read in the bayonet picture to be processed;

[0074] S2. Perform image preprocessing on the bayonet image to be processed, including adjusting the contrast range and noise reduction processing;

[0075] S3. Analyzing the hierarchical structure of the preprocessed image;

[0076] S4. Merge the vehicle information into a record and store it in the database.

[0077] In step S1, the system reads in the bayonet pictures to be processed, and supports processing methods such as single image files, multi-image files, and folder recursive input.

[0078] In step S2, an optimization method is used to dynamically adjust the contrast range of the bayonet picture to be processed, and the optimization method includes the following steps:

[0079] S201. Calculate image dynamic adjustment parameters a, b;

[0080] S...

Embodiment 2

[0140] Similar to Example 1, the implementation method of the present invention will be described in detail with specific implementation cases.

[0141] It should be noted that the HoG feature is a feature description for target detection. This technology counts the number of directional gradients that appear locally in the image. The Lab value describes all the colors that people with normal vision can see. GMM is a Gaussian model, and SVM is Support Vector Machines.

[0142] First classify vehicle types including:

[0143] Car classifier: the window size is 240*224, the block size is 16*16, the cell size and block overlap step are both 8*8, and the histogram accuracy is 9. The length of the HoG vector is the number of blocks (29*27) 783* the number of cells contained in each block (4) * histogram precision (9) = 28188.

[0144] Cart classifier: the window size is 288*288, the block size is 16*16, the cell size, and the block overlapping step are all 8*8, and the histogram ...

Embodiment 3

[0153] Similar to Example 2, the difference is that when the bayonet picture is multiple vehicles,

[0154] Input image: 1600*1200, the a and b obtained by image preprocessing are a =71, b =250, use these two parameters to stretch the contrast of the original image, and then reduce the image size to half of the original size for model The detection (big car, small car, tricycle three classifiers each run once), the recognition result is two small car areas, and the obtained area information RB 1 The coordinates of the upper left point (364, 412), the size of the rectangle is 514 in width and 480 in height, the unit is pixel, and the coefficient of zooming to the standard size of the car (240*224) is 2.142; RB 2 The coordinates of the upper left point are (856, 630), the size of the rectangle is 550 in width and 512 in height, and the units are all pixels. The coefficient of zooming to the standard size of the car (240*224) is 2.292.

[0155] In this example, two targets are h...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention relates to a vehicle information extraction method and s system thereof. The input of the system is a picture shot by an existing bayonet system. Through the pretreatment methods of the adjustment of a contrast range and de-noising processing, a hierarchical analysis method is employed, a plurality of areas comprising vehicles in the picture and the corresponding vehicle type information and vehicle color information, for a passenger car area, the brand and specific model of the car are calculated through a unified vehicle type recognition model, and vehicle license plate information is extracted for passenger car and large vehicle areas. Through the system, the vehicle type, brand and specific model of a vehicle in the picture can be analyzed in a 100-millisecond level.

Description

technical field [0001] The invention belongs to the technical field of image recognition, in particular to a vehicle information extraction method and system. Background technique [0002] With the road traffic becoming more and more convenient, driving crimes has become the main method for criminals to commit trans-regional crimes, especially in cases of property infringement. Correspondingly, the construction of high-definition monitoring has gradually developed to the interior of the city, forming a dense structure surrounded by inside and outside, and greatly improving the control ability of vehicles. The existing bayonet system mainly uses the license plate information of the vehicle on the picture to complete the query and retrieval of the massive picture database. This retrieval method relies on the premise that the license plate information of the vehicle is correct regardless of the license plate itself or the recognition result. However, according to the analysis...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/62G06K9/46G06T7/00
Inventor 丁濛杨锋
Owner HANGZHOU KEDU TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com