KR20140117808A - Device and method for calibrating image distortion of all around survillance camera - Google Patents

Abstract

The present invention relates to a device and a method for correcting image distortion of a surveillance camera. More specifically, the device and the method are capable of correcting image distortion of a photographed image through distortion correction based on a distortion coefficient of a lens, distortion correction by parameter adjustment, and distortion correction based on planarization image processing. The device for correcting image distortion of a surveillance camera according to the present invention comprises: a photographing unit to receive an image by a wide-angle lens, and to photograph an omnidirectional image to be outputted as picture data; an image processing unit to extract a distortion coefficient for distortion of the image generated by the wide-angle lens, extract the outer most line of a camera module based on the photographed image, select a planarization area from the image-distorted picture data, set the selected area as an interest area, determine a center coordinate value and a focal distance for an image area based on the outer most line of the extracted image area, determine a parameter for correcting distortion, overlap the interest area with the image for cropping processing, and unwrapping-process a distant image curved by cropping, to convert into a planarization image; and a distortion correcting unit to correct the image distortion by the extracted distortion coefficient, adjust the center coordinate value and the focal distance to calculate an optimum parameter, correct image distortion, and restore to an original image through an image calibration process by the unwrapping-processing after the unwrapping-processing process.

Description

Technical Field [0001] The present invention relates to an image distortion correction apparatus and a distortion correction method for a surveillance camera,

The present invention relates to an image distortion correction apparatus and a distortion correction method of a surveillance camera, and more particularly to a distortion correction method and a distortion correction method, And more particularly, to an apparatus and method for correcting image distortion of an image.

The application fields using cameras are various across the industry and their application fields are becoming more and more widespread. For example, cameras are being used for automation in industry instead of human eyes, such as robot control of an assembly line, product quality inspection, medical diagnosis, security system, and image system recognition.

In general, if you want to obtain a high-precision image of a subject, use a narrow-angle lens at the front of the camera. If you need to observe a wide area with a large subject, use a wide-angle lens Device.

When a wide angle lens is used, it has a merit that a wide field of view can be secured. On the other hand, the central portion of the lens has a high resolution and the resolution gradually decreases toward the outer portion of the lens.

In addition, with the wide-angle lens, the distortion of the image, that is, the radial distortion is remarkable as the resolution of the outer portion decreases. This radial distortion is also a major factor in resolution degradation.

Korean Patent Laid-Open No. 10-2006-0041116 discloses an apparatus and method for compensating for a distorted image and an image display system using the apparatus.

Korean Patent Application Publication No. 10-2006-0041116 discloses a distortion image correcting apparatus comprising: a camera for obtaining a three-dimensional image having a view angle of 180 degrees or more through a fish-eye lens as a two-dimensional pre-correction image; A distortion coefficient acquiring unit for acquiring and storing the parameters of the fish-eye lens, and storing specific coordinates of the pre-correction image reflecting the display time; A user interface unit for acquiring specific coordinates of the pre-correction image reflecting the display time and providing the specific coordinates to the distortion coefficient obtaining unit when the user selects a new display time point; And a correction unit configured to correct the spherical aberration and the spherical aberration of the spherical aberration correcting unit to correct the spherical aberration and the spherical aberration of the spherical aberration correction unit, An image correcting unit for obtaining a posterior image; And an image buffer that separately stores the pre-correction image and the post-correction image in separate areas.

However, the image of the surveillance camera includes not only the image distortion caused by the lens but also various factors such as image distortion due to various parameters and image distortion generated in the process of converting the curved surface image into the plane image. Therefore, as disclosed in Korean Patent Laid-Open Publication No. 10-2006-0041116, there is a limit to the image distortion correction method by the distortion coefficient of the lens.

Korean Patent Publication No. 10-2006-0041116

An object of the present invention is to provide an image processing apparatus and method capable of simultaneously outputting an optimal image by correcting a distortion coefficient of a lens, a parameter adjustment for correcting distortion of a photographed image, An image distortion correction apparatus and a distortion correction method of a surveillance camera.

According to an aspect of the present invention, there is provided an apparatus for correcting an image distortion of a surveillance camera, comprising: an imaging unit that receives an image with a wide-angle lens, captures an omnidirectional image and outputs the image data; And extracting an outline of the camera module based on the captured image, selecting a flattened area among the image distorted image data, and selecting the selected area as a region of interest Determines a parameter for distortion correction by determining a center coordinate value and a focal distance of the image region based on the outline of the extracted image region, overlaps the region of interest with an image, crops the image, An image processor for converting an image of a long distance curved by ping into a plane image by unwrapping the far image; Correcting the image distortion due to the extracted distortion coefficient, correcting the image distortion by calculating the optimal parameter by adjusting the center coordinate value and the focal length, and performing the image calibration process by the unlamping process after the unlapping process And a distortion correcting unit for reconstructing the original image.

According to another aspect of the present invention, there is provided a method of correcting image distortion of a surveillance camera, comprising: capturing an image with a wide-angle lens, capturing an omnidirectional image and outputting the image data as image data; Angle lens, extracts a distortion coefficient for distortion of the image generated by the wide-angle lens, extracts the outline of the image area by dividing the housing area and the image area of the camera module based on the sensed image, A distortion correction pre-setting step of selecting a flattened area and setting the selected area as a region of interest; Determining a center coordinate value for the image region based on the outline of the extracted image region, overlapping and overlapping the region of interest on the image, and unwrapping the far-field image, which is curved by the cropping, To a distortion correction step Correcting the distortion caused by the extracted distortion coefficients, obtaining a corrected image using the center coordinate value while changing a focal length value of the image, and determining a focal distance based on straight lines appearing in the respective corrected images And a distortion correction ending step of restoring a plane image to an original image through an image calibration process after the unwrapping process.

According to the configuration of the present invention described above, in order to correct the distortion of the photographed image, it is possible to simultaneously correct the distortion coefficient of the lens, the parameter adjustment for distortion correction of the photographed image, It is possible to provide an image distortion correction apparatus and a distortion correction method of a surveillance camera.

FIG. 1 is a block diagram of an internal block of a surveillance camera for implementing a method of correcting image distortion of a surveillance camera according to the present invention.
2 is a flowchart of a method for correcting image distortion of a surveillance camera according to the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Reference will now be made in detail to the preferred embodiments of the present invention, examples of which are illustrated in the accompanying drawings.

FIG. 1 is a block diagram of an internal block of a surveillance camera for implementing a method of correcting image distortion of a surveillance camera according to the present invention.

Referring to FIG. 1, a surveillance camera system according to the present invention includes an imaging unit 100, an image processing unit 200, a distortion correction unit 300, and a monitor 400.

The image sensing unit 100 includes a wide angle lens, and the incident image is sensed as an electrical signal image through an image sensor, and the image processing unit 200 corrects the electrical signal image.

The image processing unit 200 is a processor that processes an image input through the imaging unit 100. The processor 200 corrects radial distortion generated by the wide-angle lens, corrects distortion of the captured area by parameter adjustment, And performing a ping and unwrapping to correct the captured area.

The image processing unit 200 includes a distortion coefficient extracting unit 210, a parameter adjusting unit 220, and a plane image transforming unit 230.

The distortion coefficient extracting unit 210 extracts a distortion coefficient for image distortion generated by the wide-angle lens, and then causes the distortion correction unit 300 to correct the image through the backward mapping method using the extracted distortion coefficient.

The parameter adjusting unit 220 extracts an outermost line of the photographed image, determines a center coordinate value therefor, obtains a corrected image using the center coordinate value, determines a focal distance from the corrected image, And includes an outermost line extracting unit 221 and a center coordinate / focal length determining unit 222.

The planar image transform unit 230 maps the photographed image to a region of interest and corrects the distorted image through a chroming process and an unwrapping process, and includes a chroming unit 231 and an unwrapping unit 232 .

The distortion correction unit 300 is a processor for correcting a distorted image using data obtained through the distortion coefficient extraction unit 210, the parameter adjustment unit 220, and the planar image conversion unit 230.

The monitor 400 is an apparatus for displaying an image processed through the image processor 200 and the distortion correction unit 300, and may be an LCD, a CRT monitor, or the like.

2 is a flowchart of a method for correcting image distortion of a surveillance camera according to the present invention.

As shown in FIG. 2, the image distortion correction method of the surveillance camera according to the present invention includes a method of extracting a distortion coefficient value of a wide-angle lens and correcting the image using the extracted distortion coefficient value (S211) (S212) and an image correction method (S213) by plane image conversion are simultaneously applied.

First, at least one image to be acquired by the camera module having the wide-angle lens is captured and acquired (S214).

The distortion coefficient extracting unit 210 in this embodiment can use the following formula (1) as a lens distortion equation for obtaining a distortion coefficient (S215).

Here, R and R ₀ are coordinate distances from the center of the corrected image and the distorted image, R _d is the maximum coordinate distance from the center of the distorted image to the outermost edge, k is a distortion coefficient, and a is a constant of 2 or more.

The lens distortion equation can be defined as a high-order term with a large number of distortion coefficients for fine distortion correction (in this case, a numerical calculation error due to an increase in the distortion coefficient to be measured may occur) Lens distortion can be measured. In the case of radial distortion, only one distortion factor can be sufficiently modeled (in this case, modeling is not sufficient for sophisticated distortion correction, but errors in numerical computation can be minimized).

In the present invention, since the distortion model is assumed to be radial distortion, the optimal distortion coefficient becomes positive and the maximum value is determined by R _d and a. Thus, the optimal distortion factor can be extracted by manual adjustment according to [Equation 1] within a defined range.

Next, the housing area and the image area of the camera module are separated based on the obtained image, and the outermost line of the image area is extracted (S216).

The image area is divided into a housing area and an image area, and a boundary line (outermost line) between the image area and the housing area can be identified. The outermost line can be obtained using an edge detection method.

As described above, when the housing area and the image area are distinguished from each other, the boundary lines are formed as the outermost lines of the image area, and the outermost lines can be displayed as a set of coordinate values.

After the image acquisition (S214), the flattened region is selected from the image distorted image data, and the selected region is set as the region of interest (S217).

The distortion coefficient extraction (S215), the extraction of the outline of the acquired image region (S216), and the setting of the region of interest of the planarization region (S217) may be performed at the same time.

Next, distortion correction of the image is primarily performed using the distortion coefficient obtained in the distortion coefficient extraction (S215) (S218).

When the optimal distortion coefficient is extracted through the above-described step S215, the distortion correction unit 300 can correct the image using the extracted distortion coefficient using the backward mapping method.

When the outermost line of the image area is extracted from the step S216, the center coordinate value and the focal length value for the image area are determined based on the outline of the extracted image area (S219).

The determination of the center coordinate value can be made by the following process. Since the outline of the image area is an elliptical shape, an elliptic equation can be used.

The center coordinates of the ellipse can be obtained by using the ellipse equation, and the center coordinates of the ellipse equation can be obtained by expressing the center coordinates of the ellipse, and the center coordinates can be determined as the center coordinate value. The coefficient values of the ellipses can be obtained by the undetermined method.

In step S214, the image is acquired, the straight lines are determined by the corrected image, and the focal distance is adjusted based on the determined straight lines. Thus, the image is captured by photographing the straight line patterns intersecting in the horizontal and vertical directions .

The corrected image is obtained while changing the focal distance value with respect to the obtained image, the straight lines appearing in the respective corrected images are discriminated, and the focal distance is determined based on the determined straight lines.

Next, the cropping process and the unwrapping process are performed after the ROI setting step S217 (S220).

First, the ROI set in step S217 is overlaid on the image, and then the ROI is cropped to generate a curved image of the desired area.

After the cropping process, the process goes to the unwrapping process, and the remote image curved by the cropping is unwrapped and converted into a plane image.

Finally, after the image distortion correction (S218), the correction by the parameter determination (S219), and the correction by the flattening image processing (S220), the calibration process is finally performed (S221).

When the correction by the image distortion correction (S218) is performed, the image quality of the outer portion of the corrected image may be deteriorated (blurring phenomenon), and a calibration process for minimizing such phenomenon may be added.

As described above, the focal distance and the center coordinate are important parameters for performing the distortion correction in step S219. The center position of the acquired image is adjusted and the focal distance is adjusted by adjusting the center position when the center position is adjusted. Although the focal length is not used when adjusting the optical center position, the center position value should be used when adjusting the focal distance. Accordingly, in order to calculate a sophisticated parameter, it is possible to calculate the parameter by adjusting the center position of the acquired image first and adjusting the focal distance value in a state in which the adjusted center position is reflected, and more precisely performing the distortion correction by using these parameters .

After the unlapping process in step S220, the normal image restoration process may be performed, and the coarse planar image may be restored to the original image through the image calibration process by the unlamping process.

While the present invention has been described with reference to exemplary embodiments, it is to be understood that the invention is not limited to the disclosed embodiments, but, on the contrary, As will be understood by those skilled in the art. Therefore, it should be understood that the above-described embodiments are to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than the foregoing description, It is intended that all changes and modifications derived from the equivalent concept be included within the scope of the present invention.

100: image pickup unit 200: image processing unit
210: Distortion coefficient extracting unit 220:
221: outermost line extracting unit 222: center coordinate / focal length determining unit
230: plane image conversion unit 231: chroming unit
232: unwrapping part

Claims (2)

An image pickup section for picking up an image with a wide-angle lens, picking up an omnidirectional image and outputting it as image data;
Angle lens, extracts a distortion coefficient for distortion of the image generated by the wide-angle lens, extracts an outline as a boundary line between the housing area and the image area of the camera module based on the sensed image, Determining a parameter for distortion correction by determining a center coordinate value and a focal length of the image region based on the outline of the extracted image region to determine a parameter for distortion correction, An image processing unit for overlapping and cropping the cropped long-distance image by cropping, and converting the curved long-distance image into a flat image by unlapping;
Correcting the image distortion due to the extracted distortion coefficient, correcting the image distortion by calculating the optimal parameter by adjusting the center coordinate value and the focal length, and performing the image calibration process by the unlamping process after the unlapping process And a distortion correcting unit for correcting the distortion of the video signal to an original image. An image capturing step of receiving an image with a wide-angle lens, capturing an omnidirectional image and outputting it as image data;
Angle lens, extracts a distortion coefficient for distortion of the image generated by the wide-angle lens, extracts the outline of the image area by dividing the housing area and the image area of the camera module based on the sensed image, A distortion correction pre-setting step of selecting a flattened area and setting the selected area as a region of interest;
Determining a center coordinate value for the image region based on the outline of the extracted image region, overlapping and overlapping the region of interest on the image, and unwrapping the far-field image, which is curved by the cropping, To a distortion correction step
Correcting the distortion caused by the extracted distortion coefficients, obtaining a corrected image using the center coordinate value while changing a focal length value of the image, and determining a focal distance based on straight lines appearing in the respective corrected images And a distortion correction completion step of restoring a plane image to an original image through an image calibration process after the unwrapping process.

Images