Method of Locating License Plate of Moving Vehicle

• US 20110135156A1
• Filed: 06/04/2010
• Published: 06/09/2011
• Est. Priority Date: 12/04/2009
• Status: Active Grant

• Current Assignee: Huper Laboratories Co., Ltd.
• Original Assignee: Huper Laboratories Co., Ltd.
• Inventors: Huang, Chi-Ming, Chen, Chao-Ho, Huang, Hui-Pin
• Granted Patent: US 8,290,213 B2
• Time in Patent Office: Days
• US Class Current: 382/105
• CPC Class Codes: G06V 10/28 Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
• CPC Class Codes: G06V 20/62 Text, e.g. of license plates, overlay texts or captions on TV images
• CPC Class Codes: G06V 20/625 License plates

Abstract

While locating a license plate of a moving vehicle on consecutive images, motion detection is first performed on the consecutive images to detect a moving vehicle image, which is segmented using edge detection, and the segmented moving vehicle image is analyzed to retrieve characteristics for locating a license plate image and determining characters on the located license plate image. As a result, a precise location of the license plate is thus precisely located for further recognition no matter what weathers in which the consecutive images are recorded. The above-mentioned technique requires merely few calculations, is easily implemented, and may be applied on an intelligent digital video recording (DVR) system including many computer-vision functions.

Claims

1. A method for locating license plate of a moving vehicle, comprising:

transforming a color image, which films a moving vehicle, into a first gray level image;

performing edge detection on the first gray level image by using a Sobel operator, so as to generate a second gray level image, wherein the second gray level image comprises a plurality of gradients of the first gray level image;

determining a first intermediate gradient, which is from the plurality of gradients of the first gray level image, to be a threshold value, and processing the second gray level image according to the threshold value so as to generate a third gray level image, when a first pixel amount, which is corresponding to the first intermediate gradient and is of the second gray level image, is larger than a second pixel amount, which is corresponding to a second intermediate gradient and is of the second gray level image, by a multiple, wherein the second intermediate gradient is from the plurality of gradients of the first gray level image, and wherein the first intermediate gradient is higher than the second intermediate gradient;

performing a morphological operation on the third gray level image, so as to generate a fourth gray level image;

scanning the fourth gray level image by using a rectangular mask, so as to determine an edge density of the fourth gray level image;

comparing the edge density with a critical edge density, so as to confirm whether there is a license plate image on the fourth gray level image or not; and

locating the fourth gray level image so as to locate the license plate image and to display the located license plate image on a screen, when the license plate image is determined to be on the fourth gray level image.

2. The method of claim 1,

wherein transforming the color image into the first gray level image comprises:

transforming a plurality of red, green, and blue pixel values in the color image into corresponding gray values on the first gray level image, according to an equation;

Gray_i,j=0.299R_i,j+0.587G_i,j+0.114B_i,j;

wherein R_i,j, G_i,j, and B_i,j respectively indicate a red pixel value, a green pixel value, and a blue pixel value at a coordinate (i, j) on the color image, and Gray_i,j indicates a gray value at the coordinate (i, j) on the color image.

3. The method of claim 1,

wherein the performing edge detection on the first gray level image by using the Sobel operator so as to generate the second gray level image comprises:

performing the edge detection according to a vertical mask of the Sobel operator, so as to generate the second gray level image.

4. The method of claim 1 further comprising:

adjusting a plurality of gradients of pixels, each of which has a gradient higher than an upper bound gradient and is on the second gray level image, to the upper bound gradient.

5. The method of claim 1, further comprising:

calculating a plurality of gradients on the second gray level image for different values of gradients, so as to determine an amount of pixels for each value of gradients on the second gray level image.

6. The method of claim 1 further comprising:

confirming whether the first pixel amount is increased with respect to an amount of pixels corresponding to the first intermediate gradient on a previously-detected color image; and

confirming whether the first pixel amount is larger than the second pixel amount by the multiple or not, when the first pixel amount is confirmed to be increased with respect to the amount of pixels corresponding to the first intermediate gradient on the previously-detected color image.

7. The method of claim 1,

wherein determining the first intermediate gradient to be the threshold value comprises:

confirming whether a pixel amount corresponding to an intermediate gradient i is larger than a pixel amount corresponding to an intermediate gradient (i−

1) or not, according to an equation;

8. The method of claim 7, further comprising:

confirming whether the first intermediate gradient is higher than a lower bound gradient or not, when the first pixel amount is not larger than the second pixel amount by the multiple;

lowering the value of the first intermediate gradient so as to generate a third intermediate gradient when the first intermediate gradient is higher than the lower bound gradient, and confirming whether a pixel amount corresponding to the third intermediate gradient on the second gray level image is increased or not with respect to an amount of pixels corresponding to the third intermediate gradient on a previously-detected color image;

confirming whether a third pixel amount, which indicates a pixel amount corresponding to the third intermediate gradient on the second gray level image, is larger than a fourth pixel amount, which indicates a pixel amount corresponding to a fourth intermediate gradient on the second gray level image, by the multiple or not, wherein the third intermediate gradient is higher than the fourth intermediate gradient; and

confirming whether the third intermediate gradient is higher than the lower bound gradient, when the third pixel amount is confirmed to be not larger than the fourth pixel amount by the multiple.

9. The method of claim 1, wherein performing the morphological operation on the third gray level image so as to generate the fourth gray level image comprises:

performing a dilation operation of the morphological operation on the third gray level image, so as to generate the fourth gray level image.

10. The method of claim 1,

wherein scanning the fourth gray level image by using the rectangular mask, so as to determine the edge density of the fourth gray level image comprises:

calculating an edge density D, according to an equation

11. The method of claim 1,

wherein comparing the edge density with the critical edge density so as to confirm whether there is a license plate image on the fourth gray level image or not comprises:

confirming whether there is a license plate image on the fourth gray level image or not by using horizontal projection, when the edge density is higher than the critical edge density.

12. The method of claim 11,

wherein confirming whether there is a license plate image on the fourth gray level image or not by using the horizontal projection comprises:

confirming whether the fourth gray level image has been completely scanned or not, when there is no license plate image on the fourth gray level by using the horizontal projection.

13. The method of claim 12, further comprising:

reducing a size of the rectangular mask, and scanning the fourth gray level image by using the reduced rectangular mask, when the fourth gray level image has been confirmed to be completely scanned.

14. The method of claim 1,

wherein comparing the edge density with the critical edge density so as to confirm whether there is a license plate image on the fourth gray level image or not comprises:

confirming whether the fourth gray level image has been completely scanned or not, when the edge density is lower than the critical edge density.

15. The method of claim 14, further comprising:

reducing a size of the rectangular mask, and scanning the fourth gray level image by using the reduced rectangular mask, when the fourth gray level image has been confirmed to be completely scanned.