WO2012000296A1

WO2012000296A1 - Image tilt correction method and apparatus

Info

Publication number: WO2012000296A1
Application number: PCT/CN2010/080304
Authority: WO
Inventors: 李挺; 陈维强; 李月高; 裴雷; 刘微
Original assignee: 青岛海信网络科技股份有限公司
Priority date: 2010-06-30
Filing date: 2010-12-27
Publication date: 2012-01-05
Also published as: CN101923710A

Abstract

An image tilt correction method and apparatus are provided, and the method comprises: performing greyscale processing on the acquired image to obtain a greyscale image, performing projection processing on the greyscale image to judge whether the greyscale image is tilted, and performing tilt correction processing on the greyscale image if the greyscale image is tilted. The tilt correction processing is not performed on the greyscale image which is not tilted, thus the efficiency of the image tilt correction method is improved.

Description

The present invention claims the priority of the Chinese Patent Application entitled "Image Tilt Correction Method and Apparatus", which is filed on June 30, 2010, the Chinese Patent Application No. 201010221775.3, the entire contents of which are incorporated by reference. In this application. Technical field

The present invention relates to the field of image processing technologies, and in particular, to an image tilt correction method and apparatus. Background technique

At present, with the development of intelligent transportation systems, image processing technology is widely used in intelligent transportation systems. In the intelligent transportation system, the image recognition technology is usually used to image the license plate of the vehicle, thereby realizing the automatic acquisition of the license plate number of the vehicle.

Due to the influence of different road environments, vehicles traveling on the road may be inclined. In order to accurately obtain the license plate information of the vehicle, the prior art license plate image recognition system needs to perform image correction processing on the license plates of all the detected vehicles, so that the image of the license plate can facilitate the image recognition system to recognize the license plate number.

It can be seen from the above that the image correction processing method in the prior art performs correction processing on all detected images, and performs image correction processing on vehicles that have not tilted, which greatly increases the burden of image correction processing and affects the license plate. The speed at which the tilted vehicle performs image correction processing reduces the correction efficiency of the image correction processing. Therefore, the processing method of the prior art method is low in efficiency. Summary of the invention

The present invention provides an image tilt correction method and apparatus for solving the defect of low efficiency of the image tilt correction method in the prior art, and improving the correction efficiency of the image tilt correction method.

An image tilt correction method provided by an embodiment of the present invention includes:

Performing grayscale processing on the acquired image to obtain a grayscale image; Performing a projection process on the grayscale image to determine whether the grayscale image is tilted; if the grayscale image is tilted, performing a tilt correction process on the grayscale image;

The step of performing a projection process on the grayscale image to determine whether the grayscale image is tilted, specifically:

Performing binarization processing on the grayscale image to obtain a binarized image;

Vertically projecting the binarized image to obtain a projection length of the binarized image; comparing the projection length with a preset projection length threshold;

Whether the grayscale image is tilted is determined based on the comparison result.

An embodiment of the present invention further provides an image tilt correction apparatus, including:

a grayscale processing module, configured to perform grayscale processing on the acquired image to obtain a grayscale image; and a projection processing module, configured to perform a projection process on the grayscale image to determine whether the grayscale image is tilted; : performing binarization processing on the grayscale image to obtain a binarized image; performing vertical projection on the binarized image to obtain a projection length of the binarized image; and using the projection length and the pre-projection length Setting a projection length threshold for comparison; determining whether the grayscale image is tilted according to a comparison result;

And a correction processing module, configured to perform a tilt correction process on the grayscale image if the grayscale image is tilted.

The image tilt correction method and apparatus provided by the present invention determines whether the grayscale image is tilted by performing projection processing on the grayscale image, and performs tilt correction processing on the grayscale image if the grayscale image is in an oblique state, and The oblique grayscale image is not subjected to the tilt correction processing, and the correction efficiency of the image tilt correction method is improved. DRAWINGS

1 is a flow chart of an embodiment of an image tilt correction method according to the present invention;

2 is a specific flowchart of step 102 in the image tilt correction method embodiment of the present invention; FIG. 3 is a specific flowchart of step 103 in the image tilt correction method embodiment of the present invention; 4 is a schematic structural view of an embodiment of an image tilt correction device according to the present invention;

FIG. 5 is a schematic structural diagram of a projection processing module in an embodiment of an image tilt correction apparatus according to the present invention; FIG.

Fig. 6 is a schematic view showing the detailed structure of a correction processing module in the embodiment of the image tilt correction device of the present invention. detailed description

The technical solutions in the embodiments of the present invention are clearly and completely described in the following with reference to the accompanying drawings in the embodiments of the present invention. It is a partial embodiment of the invention, and not all of the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.

1 is a flow chart of an embodiment of an image tilt correction method of the present invention. As shown in FIG. 1, the image tilt correction method of this embodiment includes:

Step 101: Perform grayscale processing on the acquired image to obtain a grayscale image.

Specifically, step 101 in the present embodiment processes the acquired image information so that the processed image becomes a grayscale image. In the embodiment, the vehicle license plate is detected in the intelligent transportation system as an example. The image information of the vehicle license plate on the road is obtained by the image acquisition device such as the camera provided on the road, and then the image information of the license plate is obtained through step 101. Grayscale processing is performed to obtain a grayscale image of the license plate image.

Step 102: Perform a projection process on the grayscale image to determine whether the grayscale image is tilted.

Specifically, after obtaining the grayscale image of the license plate by step 101, the obtained grayscale image is subjected to projection processing by step 102, and it is determined whether the grayscale image is tilted based on the image information obtained by the projection.

Step 103: If the grayscale image is tilted, the grayscale image is subjected to tilt correction processing.

Specifically, when the grayscale image is processed by the step 102, the grayscale image is obtained as a tilted image. After the image, the grayscale image is subjected to tilt correction processing in step 103. For example, after it is known in step 102 that the obtained grayscale image of the license plate is in an inclined state, it can be determined that the detected license plate of the vehicle is inclined, and it is necessary to perform tilt correction processing on the grayscale image of the inclined license plate, and then step through 103 corrects the grayscale image of the license plate to obtain a grayscale image of the license plate without the oblique angle, so that the subsequent program obtains the license plate number according to the grayscale image without the oblique angle.

It should be particularly noted that the image tilt correction method of the present embodiment can be applied to the license plate recognition system of the intelligent transportation system, and can also be used for other occasions where the image is subjected to the tilt correction processing.

In the image tilt correction method of the embodiment, the grayscale image is subjected to projection processing to determine whether the grayscale image is tilted, and if the grayscale image is in an oblique state, the grayscale image is subjected to tilt correction processing, and for the grayscale without tilting The degree image is not subjected to the tilt correction processing, and the correction efficiency of the image tilt correction method is improved. In addition, the image tilt correction method of the embodiment performs the tilt correction processing only on the image in the tilt state, and the tilt-free image does not need to perform the tilt correction processing, thereby effectively avoiding image information loss in the non-tilted image, which is beneficial to improving the entire image. The efficiency of the process.

Based on the foregoing technical solution, a specific implementation manner of step 102 in this embodiment is shown in FIG. 2, and specifically includes the following steps:

Step 1021: Perform a binarization process on the grayscale image to obtain a binarized image. Specifically, in step 1021, the grayscale image obtained in step 101 is binarized to convert the grayscale image into a binarized image. For example: When using a black and white binarized image, after the grayscale image of the license plate is processed by binarization, the license plate number will turn white and the background will turn black, thus forming a black and white binarized image.

Step 1022: Perform vertical projection on the binarized image to obtain a projection length of the binarized image. Specifically, a binarized image of the grayscale image is obtained by step 1021, and a binarized image is vertically projected by step 1022, so that the projection length of the binarized image can be obtained. For example: After the black and white binarized image of the license plate is vertically projected, a black and white spaced projection is formed in the X-axis direction, and the projection length can be the total length of the white projection of the white license plate number on the X-axis, or X. The total length of the black projection between the white projections on the axis.

Step 1023: Compare the projection length with a preset projection length threshold. Specifically, by step After the vertical projection of the binarized image, the length of the white projection area of the binarized image and the length of the black projection area are obtained (in the case of the license plate, for example, the white license plate number is white on the X-axis). The total length of the projection and the total length of the black projection between the white projections on the X-axis). Since the projection of the binarized image of black and white includes two parts of black and white, the projection length threshold also includes a black length threshold and a white length threshold. Wherein, the black length threshold is the length value of the black projection length obtained on the X-axis after the vertical projection processing of the black-and-white binary image in the non-tilted state, and the white length value is the black and white in the non-tilted state. The value of the length of the white projection length obtained on the X-axis after the vertical projection of the image. The white projection length on the X-axis of the currently processed binarized image is compared to the white length threshold; the black projection length on the X-axis of the currently processed binarized image is compared to the black length threshold.

Step 1024: If the black projection length of the currently processed binarized image on the X axis is less than the black length threshold, determine that the grayscale image is in a tilt state; or, if the white projection length of the binarized image is greater than the white length threshold, Then, it is determined that the grayscale image is in an inclined state. Specifically, the projection length obtained on the X-axis after the vertical projection of the black-and-white binary image of the license plate is described. In the binarized image of the license plate, the image of the license plate number is white, and the image of the background is black. Since the interval between the license plate numbers is fixed, when the license plate is tilted, in the corresponding black and white binarized image, the white projection length of the license plate number on the X axis will increase, and the background area between the license plate numbers is on the X axis. The black projection length on the upper side is correspondingly shortened, so that when the black projection length of the black and white binarized image of the license plate on the X axis is smaller than the black length threshold, it is determined that the grayscale image is in an inclined state, thereby determining that the license plate is inclined. Alternatively, when the white projection length of the black and white binarized image of the license plate on the X-axis is greater than the white length threshold, it is determined that the grayscale image is tilted, thereby determining that the license plate is tilted.

In practice, the white length threshold and the black length threshold may be set according to specific application scenarios. Taking the above image correction processing for the license plate of the vehicle as an example, the license plate number is a white character, and the background area between the license plate numbers is black. Generally, the number of characters of the license plate number and the interval between the characters are relatively fixed. Therefore, the black length threshold described above is an empirical value that can be predetermined based on the number of license plate characters and the interval between characters. For white lengths, in black and white The valued image is a small value without any inclination. Therefore, a reasonable empirical value can be set as the white length threshold according to the allowable tilt, when the black and white binarized image of the license plate is at

When the white projection length on the X-axis is greater than the set white length threshold, it is determined that the corresponding license plate is tilted.

It should be noted that the image tilt correction method disclosed in the above embodiment is specifically described as an example of a license plate applied to a smart transportation system. In practice, it can also be used in other situations where it is necessary to perform tilt correction processing on an image.

In summary, the image tilt correction method of the present embodiment knows the projection length of the grayscale image by performing projection processing on the grayscale image, and then compares the projection length with the preset length threshold, so that it can be conveniently obtained according to the projection length. Whether or not the grayscale image is oblique is more advantageous for improving the correction efficiency of the image tilt correction method of the present embodiment.

Based on the foregoing technical solution, optionally, as shown in FIG. 3, step 103 in this embodiment specifically includes the following steps:

Step 1031: Acquire edge information of the grayscale image. Specifically, after the gray image is tilted by step 102, the tilt correction process needs to be performed in step 103. Step 1031 processes the grayscale image to obtain edge information of the grayscale image. In order to obtain the edge information of the grayscale image more reliably, the step 1031 in this embodiment can obtain the edge information of the grayscale image by using the Canny operator, and the Canny operator can increase the applicable range of the edge detection in a larger range, thereby further Conducive to accurate and reliable edge information of grayscale images.

Step 1032: Calculate a tilt angle of the grayscale image according to the edge information. Specifically, after obtaining the edge information of the gray image by step 1031, the tilt angle of the gray image is calculated according to the edge information by step 1032. In order to obtain the tilt angle of the grayscale image more accurately and efficiently, step 1032 in this embodiment may perform the contrast processing on the edge information of both sides of the grayscale image by Hough transform to calculate the level of the grayscale image. slope. For example: For the edge information corresponding to the inclined license plate, the effective edge information of the upper and lower parts of the license plate is extracted by Hough transform, and the comparison processing is performed, which can quickly and accurately calculate the horizontal inclination of the license plate, effectively avoiding the middle of the license plate, etc. Image The influence of the information interference line improves the calculated correct rate of the license plate inclination.

Step 1033: Rotate the grayscale image according to the tilt angle to output the grayscale image without the tilt angle. Specifically, after the tilt angle corresponding to the gray image is calculated in step 1032, the gray image may be rotated according to the tilt angle in step 1033 to output the gray image without the tilt angle. In order to rotate the grayscale image quickly and reliably, and reduce the information loss of the grayscale image during the rotation process, step 1033 in this embodiment may perform rotation correction on the grayscale image by the bilinear interpolation algorithm according to the tilt angle. After the gray image is rotated by the bilinear interpolation algorithm, the gray image without tilt angle can be obtained, which can facilitate the subsequent image processing.

In the image tilt correction method of the embodiment, the edge information of the gray image is obtained, and the tilt angle is calculated according to the edge information. Finally, the gray image is rotated according to the tilt angle to obtain the gray image without the tilt angle, which can be quickly and effectively The grayscale image that requires the tilt correction processing is processed, which is more advantageous for improving the correction efficiency of the image tilt correction method of the present embodiment.

4 is a schematic structural view of an embodiment of an image tilt correction device of the present invention. As shown in FIG. 4, the image tilt correction apparatus of the present embodiment includes: a gradation processing module 1, a projection processing module 2, and a correction processing module 3.

The gradation processing module 1 is configured to perform gradation processing on the acquired image to obtain a grayscale image; the projection processing module 2 is configured to perform projection processing on the grayscale image to determine whether the gradation image is tilted; and the correction processing module 3 is configured to If the grayscale image is tilted, the grayscale image is subjected to tilt correction processing. Specifically, the grayscale processing module 1 in this embodiment performs grayscale processing on the acquired image; then, the projection processing module 2 performs projection processing on the grayscale image generated by the grayscale processing module 1 to determine gray. Whether the degree image is tilted; Finally, the correction processing module 3 performs a correction process on the tilted grayscale image to obtain a normal state non-tilted grayscale image. The specific processing procedure of the image tilting correction apparatus of this embodiment can be referred to the description of the embodiment of the image tilt correction method of the present invention, and details are not described herein again.

In the image tilt correction device of the embodiment, the grayscale image is subjected to projection processing to determine whether the grayscale image is tilted, and if the grayscale image is in an oblique state, the grayscale image is tilt corrected. However, the tilt correction processing is not performed on the grayscale image without tilt, and the correction efficiency of the image tilt correction method is improved. In addition, the image tilt correction method of the embodiment performs the tilt correction processing only on the image in the tilt state, and the tilt-free image does not need to perform the tilt correction processing, thereby effectively avoiding image information loss in the non-tilted image, which is beneficial to improving the entire image. The efficiency of the process.

Based on the foregoing technical solution, optionally, as shown in FIG. 5, the projection processing module 2 in this embodiment includes: a projection sub-module 21, a judging sub-module 22, a first determining sub-module 23, and a second determining sub-module.

twenty four.

The projection sub-module 21 is configured to project a grayscale image to obtain a projection length of the grayscale image; the determining sub-module 22 is configured to compare the projection length with a preset length threshold; the first determining sub-module 23 is configured to compare As a result, the grayscale image is determined to be in a tilted state; the second determining sub-module 24 is configured to determine that the grayscale image is in a normal state based on the comparison result.

In combination with the foregoing method embodiment, the projection sub-module 21 is specifically configured to: perform binarization processing on the grayscale image to obtain a binary image of black and white, wherein black represents the background color of the image; and vertically project the binary image of black and white, Obtain a black projection length or a white projection length on the X-axis;

The determining sub-module 22 is specifically configured to: compare the black projection length with a preset black length threshold, or compare the white projection length with a preset white length threshold;

The first determining sub-module 23 is specifically configured to: if the comparison result is that the black projection length is less than the black length threshold, determine that the grayscale image is in a tilt state; or, if the white projection length is greater than the white length threshold, determine that the grayscale image is tilted State

The second determining sub-module 24 is specifically configured to: if the black projection length is not less than the black length threshold, determine that the grayscale image is in a normal state; or, if the white projection length is not greater than the white length threshold, determine that the grayscale image is in a normal state.

In the image tilt correction device of the embodiment, the projection length of the grayscale image is obtained by performing projection processing on the grayscale image, and then the projection length is compared with a preset length threshold, so that the gray can be conveniently obtained according to the projection length. Whether the degree image is oblique or not is more advantageous for improving the correction efficiency of the image tilt correction method of the present embodiment. Based on the foregoing technical solution, optionally, as shown in FIG. 6, the correction processing module 3 in this embodiment includes: an acquisition sub-module 31, a calculation sub-module 32, and a rotation sub-module 33.

The obtaining sub-module 31 is configured to acquire edge information of the grayscale image if the grayscale image is tilted; the calculation sub-module 32 is configured to calculate a tilt angle of the grayscale image according to the edge information; and the rotation sub-module 33 And for rotating the grayscale image according to the tilt angle to output the grayscale image without an oblique angle.

In order to obtain the edge information of the grayscale image more reliably, the obtaining sub-module 31 in this embodiment is also used to obtain the edge information of the grayscale image by the Canny operator; in order to obtain the tilt of the grayscale image more accurately and effectively The calculation sub-module 32 is further configured to perform edge processing on the edge information of the gray image by Hough transform to calculate the horizontal tilt angle of the gray image; and to rotate the gray image quickly and reliably, and reduce The information loss of the grayscale image during the small rotation, the rotation sub-module 33 is also used to perform rotation correction on the grayscale image by the bilinear interpolation algorithm according to the tilt angle.

The image tilt correction device of the embodiment obtains the edge information of the gray image and calculates the tilt angle according to the edge information. Finally, the gray image is rotated according to the tilt angle to obtain the gray image without the tilt angle, which can be quickly and effectively The grayscale image that requires the tilt correction processing is processed, which is more advantageous for improving the correction efficiency of the image tilt correction method of the present embodiment.

Those skilled in the art will appreciate that embodiments of the present invention can be provided as a method, system, or computer program product. Thus, the present invention can take the form of an entirely hardware embodiment, an entirely software embodiment, or a combination of software and hardware. Moreover, the present invention may employ a computer usable storage medium (including but not limited to disk storage, in one or more of which contains computer usable program code.

The form of a computer program product implemented on a CD-ROM, optical storage, etc.).

The present invention has been described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (system), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowcharts and/or block diagrams, and combinations of flows and/or blocks in the flowcharts and/or <RTIgt; These computer program instructions can be provided to a processor of a general purpose computer, a special purpose computer, an embedded processor or other programmable data processing device to produce a machine such that The instructions executed by the processor of the machine or other programmable data processing device generate means for implementing the functions specified in one or more blocks of the flow or in a block or blocks of the flowchart.

The computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device. The apparatus implements the functions specified in one or more blocks of a flow or a flow and/or block diagram of the flowchart.

These computer program instructions can also be loaded onto a computer or other programmable data processing device such that a series of operational steps are performed on a computer or other programmable device to produce computer-implemented processing for execution on a computer or other programmable device. The instructions provide steps for implementing the functions specified in one or more of the flow or in a block or blocks of a flow diagram.

Although the preferred embodiment of the invention has been described, it will be apparent to those skilled in the art that, Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and the modifications and modifications The spirit and scope of the Ming. Thus, it is intended that the present invention cover the modifications and variations of the inventions

Claims

Rights request

An image tilt correction method, comprising:

Performing grayscale processing on the acquired image to obtain a grayscale image;

Performing a projection process on the grayscale image to determine whether the grayscale image is tilted; if the grayscale image is tilted, performing a tilt correction process on the grayscale image;

The image tilt correction method according to claim 1, wherein the grayscale image is binarized to obtain a binarized image, specifically: binarizing the grayscale image Processing to obtain a black and white binarized image, where black represents the background color of the image;

Performing vertical projection on the binarized image to obtain a projection length of the binarized image, specifically: vertically projecting the black and white binarized image to obtain a black projection length or a white projection on the X axis Length

Comparing the projection length with a preset projection length threshold, specifically: comparing the black projection length with a preset black length threshold, or comparing the white projection length with a preset white length threshold;

Determining whether the grayscale image is tilted according to the comparison result, specifically: if the black projection length is less than the black length threshold, determining that the grayscale image is in an inclined state; or, if the white projection length is greater than The white length alarm value determines that the grayscale image is in an inclined state.

The image tilt correction method according to claim 1 or 2, wherein the tilt correction processing is performed on the grayscale image, specifically:

Obtaining edge information of the grayscale image; Calculating a tilt angle of the grayscale image according to the edge information;

The grayscale image is rotated according to the tilt angle to output the grayscale image without an oblique angle.

The image tilt correction method according to claim 3, wherein the acquiring edge information of the grayscale image is specifically: obtaining edge information of the grayscale image by using a Canny operator.

The image tilt correction method according to claim 3, wherein the calculating the tilt angle of the gray image according to the edge information is specifically: performing two sides of the gray image by Hough transform The edge information is subjected to a collation process to calculate a horizontal tilt angle of the grayscale image.

The image tilt correction method according to claim 3, wherein the rotating the grayscale image according to the tilt angle is specifically:: according to the tilt angle, by using a bilinear interpolation algorithm The grayscale image is rotated for correction.

7. An image tilt correction device, comprising:

a grayscale processing module, configured to perform grayscale processing on the captured image to obtain a grayscale image; and a projection processing module, configured to perform projection processing on the grayscale image to determine whether the grayscale image is tilted; The method includes: performing binarization processing on the grayscale image to obtain a binarized image; performing vertical projection on the binarized image to obtain a projection length of the binarized image; Comparing the preset projection length thresholds; determining whether the grayscale image is tilted according to the comparison result;

8. The image tilt correction apparatus according to claim 7, wherein the projection processing module comprises:

a projection submodule, configured to perform binarization processing on the grayscale image to obtain a binarized image, and vertically project the binarized image to obtain a projection length of the binarized image;

a determining submodule, configured to compare the projection length with a preset length threshold; a first determining submodule, configured to determine, according to the comparison result, that the grayscale image is in a tilt state; and a second determining submodule, configured to determine, according to the comparison result, that the grayscale image is in a normal state.

The image tilt correction device according to claim 8, wherein the projection sub-module is specifically configured to: perform binarization processing on the grayscale image to obtain a black and white binarized image, wherein black represents Image background color; performing vertical projection on the black and white binarized image to obtain a black projection length or a white projection length on the X axis;

The determining sub-module is specifically configured to: compare the black projection length with a preset black length threshold, or compare the white projection length with a preset white length threshold;

The first determining sub-module is specifically configured to: if the black projection length is less than the black length alarm value, determine that the grayscale image is in a tilt state; or if the white projection length is greater than the white length threshold , determining that the grayscale image is in an inclined state;

The second determining sub-module is specifically configured to: if the black projection length is not less than the black length threshold, determine that the grayscale image is in a normal state; or, if the white projection length is not greater than the white length The threshold determines that the grayscale image is in a normal state.

The image tilt correction device according to claim 7, 8 or 9, wherein the correction processing module comprises:

Obtaining a submodule, configured to acquire edge information of the grayscale image if the grayscale image is tilted; and a calculation submodule, configured to calculate a tilt angle of the grayscale image according to the edge information; and a rotation submodule, And for rotating the grayscale image according to the tilt angle to output the grayscale image without an oblique angle.

The image tilt correction device according to claim 10, wherein the acquisition sub-module is further configured to obtain edge information of the grayscale image by using a Canny operator;

The calculation sub-module is further configured to compare the edge information of the two sides of the grayscale image by a Hough transform to calculate a horizontal tilt angle of the grayscale image;

The rotation sub-module is further configured to perform rotation correction on the grayscale image by a bilinear interpolation algorithm according to the tilt angle.