JP2010103779A - Aberration correction device and method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an aberration correction device and an aberration correction method capable of surely correcting barrel aberration while relieving a processing burden. <P>SOLUTION: The aberration correction device 1 which corrects the barrel aberration generated in a wide angle lens is provided with; an image input part 11 in which an image from a camera having the wide angle lens is input; a coordinate making processing part 13 which associates each pixel of the input image with coordinates on the image; a conversion coefficient determination part 14 which determines a coordinate conversion coefficient for converting coordinates on an ellipse into coordinates after correction by approximating the barrel aberration by the ellipse; and a correction processing part 16 which corrects the pixel associated with coordinates on a coordinate ellipse to the coordinates after correction according to the determined coordinate conversion coefficient. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an aberration correction apparatus and an aberration correction method for correcting barrel aberration occurring in a wide-angle lens.

  A wide-angle lens is used as an imaging means for projecting the rear blind spot area of the vehicle on a display or the like of the navigation device. By using a wide-angle lens, it is possible to obtain a wide range of images with a single imaging means. However, since a barrel-shaped aberration in which the peripheral portion of the image is distorted into a barrel shape occurs, an image projected on a display or the like and an actual vehicle An error occurs with the surrounding environment.

  Therefore, conventionally, as an aberration correction apparatus for correcting such barrel aberration, the aberration is expressed as a function of the distance from the center of the optical axis and the tilt angle from the horizontal axis, and the aberration is corrected using the function equation (described below). Patent Literature 1), and aberration rates in the X direction and Y direction are calculated in advance from captured images of target marks arranged at equal intervals in the XY direction, and each has a correction curve (see Patent Literature 2 below). See).

However, in such a conventional aberration correction apparatus, the function formula and the correction curve are complicated high-dimensional polynomials, so that it takes time to perform arithmetic processing. For example, the image is sequentially changed like a camera mounted on a vehicle. In such a case, there arises a disadvantage that the correction process is not properly executed.
JP 2007-286809 A JP 2001-133223 A

  However, in such a conventional aberration correction apparatus, the function formula and the correction curve are complicated high-dimensional polynomials, so that it takes time to perform arithmetic processing. For example, the image is sequentially changed like a camera mounted on a vehicle. In such a case, there arises a disadvantage that the correction process is not properly executed.

  In view of the above circumstances, an object of the present invention is to provide an aberration correction apparatus and an aberration correction method capable of reliably correcting barrel aberration while reducing the processing load.

  A first invention is an aberration correction apparatus for correcting barrel-shaped aberration generated in a wide-angle lens, and an image input unit to which an image from an imaging unit having the wide-angle lens is input, and the image input unit A coordinate conversion processing unit that associates each pixel of the image with the coordinates on the image, and a coordinate conversion coefficient for converting the coordinates on the ellipse into corrected coordinates by approximating the barrel aberration with an ellipse. A conversion coefficient determination unit to be determined, and a correction processing unit that corrects pixels associated with the coordinates on the ellipse to coordinates after correction according to the coordinate conversion coefficient determined by the conversion coefficient determination unit. It is characterized by providing.

  As a result of repeated investigations on various images using an optical lens, the present inventors have found that in each of the horizontal and vertical directions, in the barrel aberration, the coordinate points on the captured image with respect to the coordinate points on the actual image. Was found to be shifted on the elliptic curve.

  Based on this knowledge, according to the aberration correction apparatus of the first aspect of the present invention, by converting the coordinates on the ellipse to the corrected coordinates by approximating the barrel aberration to an ellipse, a quadratic equation is used. It can be obtained as a solution. Therefore, the processing load can be reduced, and the coordinates on the ellipse shifted due to the aberration according to the coordinate conversion coefficient can be corrected to the coordinate points on the actual image, and the barrel shape is surely reduced while reducing the processing load. Aberration can be corrected.

  An aberration correction apparatus according to a second aspect of the invention is the aberration correction apparatus according to the first aspect of the invention, wherein the conversion coefficient determination unit is a simultaneous expression of the expression of the ellipse and the expression of a straight line connecting the coordinates on the ellipse and the corrected coordinates. Is determined on the condition that the equation of the straight line passes through fixed coordinates, thereby determining the coordinate conversion coefficient.

  According to the aberration correction apparatus of the second aspect of the invention, barrel aberration occurs radially around the fixed coordinate. (I) The straight line connecting the coordinates before and after correction is used as the fixed coordinate, for example, the origin of the image center. pass. Further, (II) the coordinates before correction are on the coordinates on the ellipse. By solving the quadratic expression with these as conditions, it is possible to easily obtain the coordinate conversion coefficient before and after correction, and it is possible to reliably correct the barrel aberration while reducing the processing load.

  An aberration correction device according to a third aspect of the present invention is the aberration correction device according to the first or second aspect, wherein the conversion coefficient determination unit includes a data table in which a part or all of the coordinate conversion coefficients are calculated for each coordinate interval. The coordinate conversion coefficient is determined by referring to a data table.

  According to the aberration correction apparatus of the third aspect of the invention, a data table is provided in which some or all of the coordinate conversion coefficients are calculated for each coordinate interval. By referring to such a data table, part or all of the arithmetic processing for determining the coordinate transformation coefficient can be omitted, the processing load can be greatly reduced, and barrel aberrations can be reliably corrected. You can also

  An aberration correction method according to a fourth aspect of the present invention is an aberration correction method for correcting barrel aberration generated in a wide-angle lens, an image input step for inputting an image from an image pickup means having the wide-angle lens, and an input image A coordinate conversion step for associating each pixel of the image with the coordinates on the image, and approximating the barrel aberration with an ellipse, thereby determining a coordinate conversion coefficient for converting the coordinates on the ellipse into corrected coordinates. A conversion coefficient determining step; and a correction processing step of correcting pixels associated with the coordinates on the ellipse to coordinates after correction according to the coordinate conversion coefficient determined in the conversion coefficient determination step. It is characterized by providing.

  According to the aberration correction method of the fourth invention, each pixel of the captured image is associated with the coordinates on the image. Then, by approximating the barrel aberration to an ellipse, a coordinate conversion coefficient for converting coordinates on the ellipse into corrected coordinates can be obtained as a solution of a quadratic expression. Therefore, the processing load can be reduced, and the coordinates on the ellipse shifted due to the aberration according to the coordinate conversion coefficient can be corrected to the coordinate points on the actual image, and the barrel shape is surely reduced while reducing the processing load. Aberration can be corrected.

  First, the configuration of an aberration correction apparatus 1 as an embodiment of the present invention will be described with reference to FIG.

  The aberration correction device 1 is a device that corrects barrel-shaped aberration generated in a wide-angle lens, and includes an image input unit 11, a frame memory 12, a coordinate processing unit 13, a conversion coefficient determination unit 14, and a conversion table database 15. And a correction processing unit 16. The aberration correction apparatus 1 is connected to a camera 2 as an imaging unit.

  The camera 2 is a CCD camera provided with a wide-angle lens, and is arranged, for example, so as to capture a rear dead angle or a side dead angle of a vehicle.

  The image processing unit 11 sequentially receives captured images (hereinafter referred to as images) output from the camera 2 and outputs the input images to the frame memory 12 and the coordinate processing unit 13 for each frame.

  The frame memory 12 stores and holds images sequentially output from the image input unit 11 by a certain number of frames from the latest frame.

  The coordinate processing unit 13 performs processing for associating each pixel or pixel portion with the coordinates on the image with respect to the image sequentially output from the image input unit 11.

  The conversion coefficient determination unit 14 approximates the barrel aberration generated in the image with an ellipse with respect to the image in which the pixel (pixel portion) and the coordinate point are associated with each other by the coordinate processing unit 13. A process of determining a coordinate conversion coefficient for converting the coordinates on the ellipse to the corrected coordinates is executed.

  The conversion table database 15 is a data table that is referred to when the conversion coefficient determination unit 14 determines the coordinate conversion coefficient, and stores values obtained by calculating the coordinate conversion coefficient for each coordinate interval.

  The correction processing unit 16 performs processing for correcting the pixel (pixel portion) associated with the coordinates on the ellipse to the corrected coordinates according to the coordinate conversion coefficient determined by the conversion coefficient determination unit 14. Execute. Then, the image corrected by the correction processing unit 16 is output to a display device such as a display (not shown).

  In the present embodiment, the processing units 11 to 16 configuring the aberration correction apparatus 1 are configured by hardware such as a CPU, a ROM, and a RAM, and these may be configured by common hardware. A part or all of the above may be configured by different hardware.

  Next, an aberration correction method executed by the aberration correction apparatus 1 will be described with reference to the flowchart shown in FIG.

  First, when the image input unit 11 of the aberration correction apparatus 1 acquires an image from the camera 2 (STEP 11), the image input unit 11 outputs the image to the frame memory 12 for each frame and stores the image data in the frame memory 12 (STEP 12). ).

  At the same time, with respect to the image data output from the image input unit 12 to the coordinate processing unit 13, the coordinate processing unit 13 applies each pixel (pixel portion) to the image sequentially output from the image input unit 11. The coordinates on the image are associated (STEP 13). Specifically, with the image center as the origin and the X axis in the horizontal direction and the Y axis in the vertical direction set, the coordinate point of each pixel or a pixel portion consisting of a certain number of pixels is specified.

  Next, the conversion coefficient determination unit 14 determines a coordinate conversion coefficient for converting the coordinates on the ellipse into corrected coordinates by approximating the barrel aberration with an ellipse (STEP 14).

  Specifically, as shown in FIG. 3A, the transform coefficient determination unit 14 has a plurality of conversion coefficients as shown by a solid line in FIG. 3B for an image in which barrel aberration occurs at the edge of the image. Set the ellipse. Note that the image shown in FIG. 3 is obtained by cutting out only the upper side of the captured image.

  Each of the plurality of ellipses is an ellipse having a major axis length a as shown in FIG. 4, and the major axis length a is known from image distortion when the target mark is captured in advance. It can be calculated as a coordinate value.

  Then, the coordinate coefficient determination unit 14 determines a coordinate conversion coefficient for converting the coordinates on each ellipse into corrected coordinates as follows.

  As shown in FIG. 4, when the coordinate after correction is Q (m ′, n ′) with respect to the coordinate P (m, n) on the ellipse before correction, the coordinate P first exists on the ellipse. Therefore, the following formula (1) is satisfied.

  On the other hand, the barrel aberration causes a radial aberration with respect to the center of the image (fixed coordinates), and therefore the coordinate Q passes through the origin, so the following equation (2) is satisfied.

  Furthermore, since the equation (2) also passes the coordinate P, the following equation (3) is satisfied.

  When equations (1) and (3) are used as simultaneous equations and solved for m and n, the following equations (4) and (5) are obtained.

  In this way, the conversion coefficient determination unit 14 sets the coordinate conversion coefficient between the coordinate P (m, n) on the ellipse before correction and the corrected coordinate Q (m ′, n ′) (formula ( 4) and (5) are determined. Since the derivation of such coefficients can be easily obtained by solving the quadratic equation, the processing load can be reduced.

  Further, since the conversion table data database 15 stores values obtained by calculating such coordinate conversion coefficients for each value of m ′ as a data table, the conversion coefficient determination unit 14 further solves the quadratic expression. Can be omitted.

  That is, in the conversion table database, since the value of the coordinate conversion coefficient when m ′ is changed for each minimum coordinate interval on the x axis is prepared in the form of a table in advance, the conversion coefficient determination unit 14 performs correction. The conversion coefficient can be acquired by referring to the table from the x coordinate (m ′) of the later coordinate point. As a result, the processing load can be greatly reduced.

  Next, the correction processing unit 16 corrects the pixel (pixel portion) associated with the coordinate before correction to the corrected coordinate according to the determined coordinate conversion coefficient (STEP 15).

  Specifically, the correction processing unit 16 reads out the pixel (pixel portion) associated with the coordinate point before correction from the frame memory 12, and associates the read pixel (pixel portion) with the coordinate point after correction. Display control processing to be displayed.

  The processing of STEPs 14 and 15 is executed for all of the plurality of ellipses that approximate the barrel aberration, and the entire image is corrected and output to a display, monitor, etc. (STEP 16). The image shown in a) is corrected as shown in FIG. 5, and the barrel aberration can be reliably corrected.

1 is an overall configuration diagram of an aberration correction apparatus in the present embodiment. The flowchart which shows the process of the aberration correction method in this Embodiment. The figure which shows the example of the image processed by STEP14 of FIG. The coordinate diagram which shows the processing content of STEP14 of FIG. Explanatory drawing which shows the example of the image processed by STEP15 of FIG.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Aberration correction apparatus, 2 ... Camera, 11 ... Image input part, 12 ... Frame memory, 13 ... Coordinate process part, 14 ... Conversion coefficient determination part, 15 ... Conversion table database, 16 ... Correction process part

Claims (4)

An aberration correction apparatus that corrects barrel aberration occurring in a wide-angle lens,
An image input unit for inputting an image from an image pickup means including the wide-angle lens;
A coordinate processing unit that associates each pixel of the image input to the image input unit with coordinates on the image;
A conversion coefficient determination unit that determines a coordinate conversion coefficient for converting the coordinates on the ellipse into corrected coordinates by approximating the barrel aberration with an ellipse;
And a correction processing unit that corrects the pixels associated with the coordinates on the ellipse to corrected coordinates according to the coordinate conversion coefficient determined by the conversion coefficient determination unit. Aberration correction device. The aberration correction apparatus according to claim 1,
The conversion coefficient determining unit solves a simultaneous expression of the ellipse expression and a straight line expression connecting the coordinates on the ellipse and the corrected coordinates on condition that the straight line expression passes through fixed coordinates. To determine the coordinate transformation coefficient. The aberration correction apparatus according to claim 1 or 2,
The conversion coefficient determination unit includes a data table in which a part or all of the coordinate conversion coefficient is calculated for each coordinate interval, and determines the coordinate conversion coefficient by referring to the data table. apparatus. An aberration correction method for correcting barrel aberration occurring in a wide-angle lens,
An image input step in which an image from an imaging means including the wide-angle lens is input;
A coordinate conversion step of associating each pixel of the input image with coordinates on the image;
A conversion coefficient determination step for determining a coordinate conversion coefficient for converting the coordinates on the ellipse into corrected coordinates by approximating the barrel aberration with an ellipse;
And a correction processing step of correcting the pixels associated with the coordinates on the ellipse to corrected coordinates according to the coordinate conversion coefficient determined in the conversion coefficient determination step. Aberration correction method.