Underground parking garage parking space detection method based on image processing

An underground garage and image processing technology, applied in the field of intelligent traffic management, can solve the problems of high cost, complicated installation, heavy maintenance workload, etc., and achieve the effect of low cost and simple maintenance

Active Publication Date: 2019-09-13

ZHEJIANG UNIV OF TECH

13 Cites 11 Cited by

AI-Extracted Technical Summary

Problems solved by technology

In addition to computer vision, other methods are complicate...

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

Abstract

The invention discloses an underground parking garage parking space detection method based on image processing. The method comprises the following steps that: S1: through a YOLOv3 deep learning model,detecting parking space angular point regions, and marking m pieces of detected angular point rectangular region sets as A={Rt|t=1,2,...,m}, wherein Rt shows the tth angular point rectangular region;S2: adopting an adaptive mean value method to carry out binaryzation on the subimage of each region Rt; S3: carrying out angular point detection on the subimage of each region Rt; and S4: according to S3, obtaining the angular point coordinates of all angular point regions in A, and sorting the angular point coordinates according to an abscissa. The detection method has the advantages that the interference of vehicles, rays and obstacles can be effectively eliminated, image processing is used for accurately detecting stop lines and parking space angular points so as to finish parking space detection, and the method is simple in maintenance and low in cost.

Application Domain

Image enhancementImage analysis +2

Technology Topic

Self adaptiveParking space +4

Image

Examples

Experimental program(1)

Example Embodiment

[0042] The specific implementation of the underground garage parking space detection method based on image processing of the present invention will be described in detail below with reference to embodiments.

[0043] An underground garage parking space detection method based on image processing includes the following steps:

[0044] Step 1: Detect the corner area of the parking space through the YOLOv3 deep learning model, and record the set of m corner rectangular areas detected as A={R t |t=1,2,...,m}, where R t Represents the t-th corner rectangular area; if the corner rectangular area is not uniformly illuminated, the illumination is compensated, and the image is corrected through histogram equalization;

[0045] Step 2: Use the adaptive mean method for each region R t Binarize the sub-images of;

[0046] Step 3: For each area R t The corner detection is performed on the sub-images, as follows:

[0047] Step 3.1: Use Gaussian filter to smooth the image to be processed;

[0048] Step 3.2: Calculate the gradient amplitude and direction of each pixel;

[0049] Step 3.3: Using the non-maximum value suppression method, find all local gradient maximum values and find the local maximum position of the pixel through the gradient direction, and set the gray value of the pixel not at the maximum value to zero;

[0050] Step 3.4: Use the dual-threshold Canny method to detect edge contours to obtain the contours of parking spaces;

[0051] Step 3.5: Perform Hough line transformation based on the parking space line contour to obtain the line segment set B={L i =(a i ,b i ,c i ,d i )|i=1,2,...,n}, where L i Represents the i-th edge line segment, (a i ,b i ) And (c i ,d i ) Respectively represent L i Two endpoints; satisfy those from B Select the line segments and form a set of horizontal line segments H = {L j =(a j ,b j ,c j ,d j ,r j ,l j )|j=1,2,…,n H }, where r j For L j Category coefficient and l j For L j The length of n H Indicates the number of line segments in H; at the same time, satisfy those from B Selected line segments to form a vertical line segment set V = {L k =(a k ,b k ,c k ,d k ,r k ,l k )|k=1,2,…,n V }, where (a k ,b k ) And (c k ,d k ) Respectively represent L k Two endpoints, r k For L k Category coefficient and l k For L k The length of n V Indicates the number of line segments in V;

[0052] Step 3.6: Delete the line segments satisfying formulas (1) and (2) from H, and delete the line segments satisfying formulas (3) and (4) from V;

[0053]

[0054]

[0055]

[0056]

[0057] Where R H And R V Indicates the confidence of the line segment category;

[0058] Step 3.7: Fit a straight line L according to the line segment in V obtained in step 3.6 v , The straight line satisfies formula (5); similarly, a straight line L is fitted according to the line segment in H obtained in step 3.6 h , The straight line satisfies formula (6);

[0059]

[0060]

[0061] Where x k1 Represents the straight line y=b k With L v The abscissa of the intersection point, x k2 Represents the straight line y=d k With L v The abscissa of the intersection point, (a k ,b k ,c k ,d k )∈V; y j1 Represents the straight line x=a j With L h Ordinate of the intersection point, y j2 Represents the straight line x=c j With L h The ordinate of the intersection point, (a j ,b j ,c j ,d j )∈H;

[0062] Step 3.8: Calculate the straight line L v And L h The intersection point is the area R t The corner coordinates of, denoted as (x t ,y t );

[0063] Step 4: According to step 4, the corner coordinates of all corner areas in A can be obtained, and they are sorted according to the abscissa to form an ordered set of corner coordinates Q={(x t ,y t )|t=1,2,...,m};

[0064] Step 5: If m=1, there is no parking space; otherwise, the parking space set Y={(x iL ,y iL ,x iR ,y iR )|i=1,2,…,n Y };

[0065]

[0066] Where (x iL ,y iL ) Represents the coordinates of the left corner point of the i-th parking space, (x iR ,y iR ) Represents the coordinates of the right corner point of the i-th parking space, δ 0 And δ 1 Respectively represent the minimum width and maximum width of the pre-calibrated parking space, and the unit is pixel. In this embodiment, δ 0 And δ 1 Take 825 pixels and 975 pixels respectively.

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

Description & Claims & Application Information

We can also present the details of the Description, Claims and Application information to help users get a comprehensive understanding of the technical details of the patent, such as background art, summary of invention, brief description of drawings, description of embodiments, and other original content. On the other hand, users can also determine the specific scope of protection of the technology through the list of claims; as well as understand the changes in the life cycle of the technology with the presentation of the patent timeline. Login to view more.

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

Classification and recommendation of technical efficacy words

Easy maintenance
Low cost

Underground parking garage parking space detection method based on image processing

An underground garage and image processing technology, applied in the field of intelligent traffic management, can solve the problems of high cost, complicated installation, heavy maintenance workload, etc., and achieve the effect of low cost and simple maintenance

AI-Extracted Technical Summary

Problems solved by technology

Abstract

Image

Examples

Example Embodiment

PUM

Description & Claims & Application Information

Similar technology patents

Bionic machinery chelonian with two-stage freedom degree flipper mechanism

Fast-disassembly method of aeroplane access cover and bearing lock

Apparatus for automatically frying foodstuff

Semiconductor device mounted structure and its manufacturing method

Disk package for a centrifuge rotor

Classification and recommendation of technical efficacy words

Networked Device Control Architecture

High clearance vehicle

Three dimensional printer

Remote medical rehabilitation system in smart city

Method and system of knowledge based search engine using text mining

End of life transmission system for surgical instruments

Method, apparatus and article for microfluidic control via electrowetting, for chemical, biochemical and biological assays and the like

Quantum dot light emitting layer

System and method for transmitting wireless digital service signals via power transmission lines

Nanowire structures comprising carbon