KR100927236B1 - A recording medium that can be read by a computer on which an image restoring method, an image restoring apparatus and a program for executing the image restoring method are recorded. - Google Patents

Abstract

Disclosed is a recording medium on which an image restoration method, an apparatus, and a program for executing the method are recorded. According to an embodiment of the present invention, in a method for reconstructing an image by an image reconstructing apparatus, (a) a plurality of element image sets obtained from a predetermined object having a different distance from the image reconstructing transmissive body through an image reconstructing transmissive body; Imaging on a plane of to generate a reconstructed image on each plane; (b) calculating an enlargement ratio by analyzing correlations with respect to the reference image and the reconstructed images, respectively; and (c) enlarging the set of element images at the enlargement ratio, and masking the reference image to the enlarged image to remove obstruction information. Accordingly, the reconstructed image from which the information of the obstacle is removed can be provided by using a masking method.

Description

Method, device for restoring of image and computer readable record-medium on which program for executing method

The present invention relates to an image restoration, and more particularly, to an image restoration method capable of restoring an object to be recognized by removing obstacle information through masking, and an apparatus and a program for executing the method, on a recording medium having recorded thereon. It is about.

Research and system implementation on three-dimensional object recognition have been advanced for centuries. Recognizing three-dimensional objects inherently involves detecting positional information of three-dimensional objects in space. Recently, an integrated imaging technique has been proposed as an approach for 3D object recognition. An integrated imaging technique, which is one of methods for recording and displaying a 3D image, is a method of obtaining multi-view 2D images of an object through a microlens array. Approaches to 3D object recognition using integrated image technique have been accompanied by a study of detecting position information of 3D object in space.

 In general, the integrated imaging system is largely composed of a pickup process, a display process and a restoration process. In the pickup process, a reduced image of a multi-view made through the lens array or the pinhole array is recorded by the CCD camera. These recorded images are called element images. In contrast, the display and restoration process is the reverse of the pickup process. The recorded element images of the display panel are reconstructed into three-dimensional images through the lens array or the pinhole array. In recovering an integrated image, there are an optical integrated image restoration technique and a computer integrated image restoration technique. The optically integrated image reconstruction technique is capable of reconstructing low-resolution three-dimensional images due to insufficient overlap between element images caused by physical limitations of optical devices such as diffraction and aberration, and reduced image quality of reconstructed three-dimensional images. Have. To overcome this drawback, a computer integrated imaging reconstruction (CIIR) based on a pinhole array model, which can reconstruct three-dimensional images computer-aided through geometrical digital simulation, was introduced. That is, the 3D image may be computerly restored on the depth plane from the overlap between all the element images mapped inductively using the pinhole array model. However, the existing computer integrated image reconstruction technique also reconstructed low resolution images due to lattice structure and contrast irregularities due to insufficient overlap between element images in the reconstructed image plane. Since then, a lens array model-based computer integrated image reconstruction technique has been proposed to improve the resolution of reconstructed three-dimensional images.

The resolution of an optically or computerly reconstructed image is an important factor in applications such as the detection and recognition of three-dimensional objects and the extraction of location information. In other words, the correlation between the images and features of a 3D object essentially acts as a sub-processor for object recognition, so the critical correlation peak in the correct output plane is highly dependent on the resolution of the images or features. Therefore, improving the resolution of the image reconstructed from the integrated image is an issue. Recently, several digital methods have been proposed to improve the resolution of images reconstructed from integrated images. One of them is to generate intermediate element images between the element images obtained by using an intermediate-view-reconstruction (IVR) method, thereby obtaining an image having an improved resolution due to sufficient overlap. to be.

Similarly, another method first rearranges the acquired element images into sub-images, generates intermediate sub-images between the rearranged sub-images using an IVR, and then converts them into element images to obtain an image having improved resolution. That's how. The number of element images increases as a result of the IVR, but the low resolution of element images or rearranged sub-images obtained from the lens array may cause a lot of false-negatives and false-positives in the IVR stage, resulting in a decrease in resolution. There is this. Therefore, even though there are no occlusion regions between adjacent element images or sub-images, the methods proposed above require additional subpixel level parallax correction and the baseline between adjacent element images or sub-images should be far enough. .

If the restoration is performed in a state where the distance between adjacent image elements or sub-images is not sufficiently given, the obstacle information serves as a noise in the reconstructed image of the object to be recognized, thereby reducing the recognition rate.

An object of the present invention is to provide an image restoration method capable of providing a restored image from which obstructions have been removed using a masking method, an apparatus, and a recording medium having recorded thereon a program for executing the method.

In addition, the present invention is to provide an image restoration method that can improve the recognition rate of the object to be recognized by removing the obstacle information to restore the object to be recognized, an apparatus and a recording medium on which the program for executing the method is recorded.

According to an aspect of the present invention, there is provided a method for restoring an image by removing obstacle information, and a recording medium recording a program for performing the method.

According to an embodiment of the present invention, in a method for reconstructing an image by an image reconstructing apparatus, (a) a plurality of element image sets obtained from a predetermined object having a different distance from the image reconstructing transmissive body through an image reconstructing transmissive body; Imaging on a plane of to generate a reconstructed image on each plane; (b) calculating an enlargement ratio by analyzing correlations with respect to the reference image and the reconstructed images, respectively; and (c) enlarging the set of element images at the enlargement ratio, and masking the reference image to the enlarged image to remove obstruction information.

The image restoration transmissive body may be a pinhole array including a plurality of pinholes or a lens array including a plurality of lenses.

Before generating the reconstructed image on the plane, the method may further include acquiring the element image set from the predetermined object through an image acquisition transmitting body.

Determining an enlargement ratio by analyzing correlations with respect to the reference image and the reconstructed images, respectively, calculating a correlation value for each reconstructed image and the reference image; Determining a distance corresponding to the reconstructed image corresponding to the correlation value having the largest correlation value as a pickup distance; And calculating an enlargement ratio using the determined pickup distance.

The enlargement ratio is calculated using the following equation.

M = L / g

L is a distance from the pinhole array to the reconstructed image, and g is a distance from the element image to the pinhole array.

Enlarging the elementary image set at the enlargement ratio, and masking the reference image to the enlarged image to remove the obstacle information, (a) opening the elementary image set pinhole array one by one to enlarge the elementary image; Expanding at a rate; (b) masking the enlarged image and the reference image to remove the obstacle information from the enlarged image.

(c) further generating a new element image by reducing the enlarged image from which the obstacle information is removed using the enlargement ratio, wherein steps (a) to (c) are repeated n times. N is the number of element images included in the element image set.

The method may further include reconstructing the object to be recognized using the plurality of new element images.

In the recording medium recording a program that can be read by the digital processing device is implemented in the form of a program of instructions that can be executed by the digital processing device to perform a method for removing the obstacle contained in the reconstructed image,

Generating a reconstructed image on each of the planes by forming a set of element images obtained from a predetermined object on a plurality of planes having a different distance from the image reconstructing transmission body through the image reconstructing transmission body;

Calculating an enlargement ratio by analyzing correlations with respect to the reference image and the reconstructed images, respectively; And

A recording medium may be provided that records a program for enlarging the set of element images at the enlargement ratio and removing the obstacle information by masking the reference image on the enlarged image.

According to another aspect of the present invention, there is provided a restoring apparatus capable of restoring an image by removing the obstacle information.

According to an embodiment of the present invention, in the image restoring apparatus, an element image set obtained from a predetermined object is imaged on a plurality of planes having different distances from the image restoring transmission body through an image restoring transmission body. An image generating unit generating a reconstructed image on a plane; A calculator configured to calculate a magnification ratio by analyzing correlations of the reference image and the reconstructed images, respectively; And a remover configured to enlarge the element image set at the enlargement ratio, and to remove the obstacle information by masking the reference image on the enlarged image.

The apparatus may further include a pickup unit configured to acquire the set of element images from the predetermined object through the image acquisition transmissive member.

The calculator calculates a correlation value for each of the reconstructed image and the reference image, and determines a pickup distance as a distance corresponding to the reconstructed image corresponding to the correlation value having the largest correlation value. The enlargement ratio may be calculated using the determined pickup distance.

The removal unit may enlarge the element image at the enlargement ratio by opening the pinhole arrays one by one, and remove the obstacle information from the enlarged image by masking the enlarged image and the reference image.

The removal unit reduces the enlarged image from which the obstacle information has been removed to generate a new element image by using the enlargement ratio, and repeats n times, where n is the number of element images included in the element image set.

The apparatus may further include a restoration unit for restoring an object to be recognized using the plurality of new element images.

By providing a recording medium on which an image restoring method, an apparatus, and a program for executing the method according to the present invention are recorded, there is an effect of providing a restored image from which information of an obstacle is removed by using a masking method.

In addition, the present invention has an effect of improving the recognition rate of the object to be recognized by removing the obstacle information to restore the object to be recognized.

As the invention allows for various changes and numerous embodiments, particular embodiments will be illustrated in the drawings and described in detail in the written description. However, this is not intended to limit the present invention to specific embodiments, it should be understood to include all transformations, equivalents, and substitutes included in the spirit and scope of the present invention. In the following description of the present invention, if it is determined that the detailed description of the related known technology may obscure the gist of the present invention, the detailed description thereof will be omitted.

Terms such as first and second may be used to describe various components, but the components should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another.

The terminology used herein is for the purpose of describing particular example embodiments only and is not intended to be limiting of the present invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. In this application, the terms "comprise" or "have" are intended to indicate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, and one or more other features. It is to be understood that the present invention does not exclude the possibility of the presence or the addition of numbers, steps, operations, components, components, or a combination thereof.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

For convenience of understanding and description, a general 3D integrated imaging system will be briefly described with reference to FIG. 1.

1 is a diagram illustrating the principle of a general three-dimensional integrated imaging system.

In general, the integrated image is composed of a pickup process and a display process. The pickup process of the integrated image consists of a two-dimensional sensor and a lens array. Here, a limited amount of rays entering through the three-dimensional objects are picked up through the lens array. At this time, the rays are stored as digital images as element images having information of each 3D object through a 2D image sensor.

On the other hand, the display process of the integrated image is the reverse of the pickup process. The digital image having the element image is displayed on the display device for reconstruction of the three-dimensional images, and displayed in the front portion of the lens array.

In addition, a computer-based reconstruction method that models the optical reconstruction method of element images is reported. This reconstruction method reconstructs a three-dimensional image in a plane placed at a certain distance.

FIG. 2 is a block diagram of an image restoration apparatus according to an embodiment of the present invention. FIG. 3 is a diagram illustrating a restoration technique in an image generating unit according to an embodiment of the present invention. FIG. 5 is a diagram illustrating an image reconstructed according to a distance according to an embodiment of the present invention. FIG. 5 is a diagram illustrating masking according to an embodiment of the present invention.

Referring to FIG. 2, the image restoration apparatus 200 according to the present invention includes a pickup unit 210, an image generator 215, a calculator 220, a remover 225, and a restorer 230. It is composed.

The pickup unit 210 acquires a set of element images from a predetermined object by using an integrated image method. Here, the element image set is a set of element images generated by different parallaxes in an integrated image method. In more detail, the pickup unit 210 acquires the element image set by recording the intensity and the direction of the light reflected from a predetermined object to an optical sensor such as a charge-couple camera through each lens of the lens array. can do.

Here, the CCD camera is a kind of digital camera and may acquire a 2D element image by converting a signal of an object incident through the lens array into an electrical signal.

That is, the integrated image method may acquire a set of element images for a 3D image by using a 2D image acquisition device such as a CCD camera.

The lens array is positioned between the predetermined object and the image capturing apparatus, and passes the light reflected from the object to the image capturing apparatus. In this case, a pinhole array may be used instead of the lens array, and the pinhole array may include a plurality of pinholes, and may pass light reflected to an object through each pinhole to be transmitted to the image acquisition device.

In addition to the lens array or the pinhole array as described above, it is obvious that the same may be applied to the present invention in the case of a transmissive body capable of passing the light reflected from the object at various angles to the image acquisition device.

Hereinafter, in the present specification, for convenience of understanding and explanation, it is assumed that the transmission body for acquiring an image is a lens array, and the transmission body for image reconstruction is assumed to be a pinhole array and will be mainly described. However, as described above, the image acquisition transmission body and the image restoration transmission body may be other transmission bodies other than the lens array and the pinhole array, and the present invention is not limited thereto.

The image generator 215 generates a reconstructed image by imaging the set of element images acquired through the pickup unit 210 on a plane using a pinhole array. Here, the reconstructed image is an image in which each element image is inversely imaged on a plane at a predetermined distance through the pinhole array.

In addition, as described above, the pinhole array may include all the transmissive bodies capable of transmitting an image of an element and forming an image on a plane as a lens array as the transmissive body for image restoration.

One elementary image of the elementary image set passes through one pinhole and is imaged on a plane at a predetermined distance from the pinhole array. Different element images of the element image set pass through different pinhole arrays and are imaged on the same plane. Thus, one reconstructed image may be generated by forming an image by overlapping element images on a plane.

This will be described in more detail with reference to FIG. 3.

As shown in FIG. 3, the reconstructed image may be generated by forming a set of element images on a plane through a pinhole arrangement.

The element image set is a set of a plurality of element images captured by an image capturing apparatus (for example, a CCD camera) as described above.

One reconstructed image may be generated by forming an element image superimposed on a plane.

For example, when generating a reconstructed image on a plane having a distance L from the pinhole array, each element image passes through the pinhole array and is imaged on the plane. In this case, the element image is reversed (ie, rotated by 180 degrees) and overlapped when the distance g between the pinhole array and the element image is smaller than L. Another element image is superimposed on the position where one element image is formed.

크기로 확대될 수 있다. 각각의 요소 영상은 각 픽셀 별로 중첩되어 더해진다. 여기서, a 및 b는 영상의 사이즈이다.Accordingly, the superimposed image elements include different viewpoints and distance information about an object and constitute one reconstructed image on a plane. In addition, the ratio of L and g is called an enlargement ratio (M = L / g) below. If the image reconstructed on the plane has an enlargement ratio (M) greater than 1, the element image is determined according to the enlargement ratio (M).

Can be enlarged to size. Each element image is added to be superimposed on each pixel. Here, a and b are the size of the image.

Referring back to FIG. 2, the calculator 220 calculates an enlargement ratio by analyzing a correlation between the plurality of reconstructed images and the reference image.

For example, the calculator 220 calculates a correlation value for each reconstructed image and a reference image, respectively. The calculator 220 may determine a distance corresponding to the reconstructed image corresponding to the largest correlation value among the calculated correlation values as the pickup distance. Subsequently, the calculator 220 may calculate the enlargement ratio using the lens formula for the determined pickup distance. For convenience of understanding and explanation, the enlargement ratio calculated using the pickup distance will be referred to as a reference ratio.

Here, the correlation is used to analyze the relationship between variables and to determine how closely one variable changes with other variables. The calculator 220 calculates a _{correlation using} the coordinates P _R (x _r , y _r , z _r ) of the reference image and the coordinates P _o (x _o , y _o , z) of the reconstructed reconstructed image. Can be. That is, the correlation is calculated as a correlation coefficient, which is a numerical value representing the degree of correlation between two variables, and the correlation coefficient has a value between -1 and 1, and the closer the absolute value is to 1, the stronger the correlation is. . Since the calculation of the correlation coefficient is already well-known numerical value, in order to facilitate the understanding and explanation of this invention, detailed description of the part which is not related to the summary of this invention is abbreviate | omitted.

This correlation is calculated repeatedly according to the change of the distance parameter Z 'until a point having a correlation peak is derived. As described above, the reconstructed planar images 142 of the target object include a sharply focused partial image reconstructed from the Z _o ' _-plane where the original target object was located and blurred partial images reconstructed from the Z _o ' -plane. Consists of Therefore, if we calculate the correlation between the reconstructed images reconstructed in each output plane according to the change of the distance parameter Z 'between the reference image and the reconstructed plane image, the remarkable correlation peak in the Z _o ' -plane where the target object is located is calculated. Will be In addition, the three-dimensional positional information (x _o , y _o , z _o ) of the target object in space can be detected based on the point having this correlation peak.

Referring to FIG. 4, images reconstructed according to a distance from a reference image are illustrated. FIG. 4A illustrates a preset reference image (ie, obstacle), and (b) to (f) are reconstructed images reconstructed for each distance. As shown in (b) to (f) of FIG. 4, it can be seen that the focus becomes clear at a certain point of time (ie, distance) among the reconstructed images reconstructed according to each distance. As such, a well-focused distance (ie, a picked-up distance and a well-reconstructed distance) is determined as the picked-up distance.

The remover 225 opens the pinhole arrays one by one to enlarge the element images one by one at a reference ratio and then masks the reference image to remove the obstacle information. Here, the reference image may be an image of an obstacle.

For example, the remover 225 reconstructs the image by transmitting the element images from the element image set through the pinhole array one by one and then forming an image on a plane according to the determined distance. Here, the reconstructed image is an image corresponding to one element image. For convenience of understanding and explanation, the image reconstructed by the remover 225 corresponding to one element image will be referred to as a reconstructed element image. The remover 225 removes the obstacle information by masking the reconstructed element image and the mask image. Here, the masking image may be an enlarged image of the reference image according to the determined distance. The remover 225 reduces the reconstructed element image from which the obstacle information is removed by using the determined distance. Here, the reduced element image from which the obstacle information has been removed will be referred to as a new element image for convenience of explanation and explanation.

The remover 225 may generate a new element image set by generating respective new element images by removing obstacle information by performing the element images included in all the element image sets.

A method of removing the obstacle information by the remover 225 will be described in more detail with reference to FIG. 5 as follows.

One element image, such as 510 illustrated in FIG. 5, is enlarged using the distance determined through the pinhole arrangement (see FIG. 515). By masking the enlarged element image and the masking image as described above, the obstacle (the barbed wire in the example of FIG. 5) information is removed to leave only information on the object to be recognized. As described above, when the enlarged element image and the masking image are masked, only the object information to be recognized is included in the enlarged element image (see 520). The remover 225 reduces the size using the distance determined by the enlarged distance to generate a new element image 525.

Referring back to FIG. 2, the reconstructor 230 may generate a reconstructed image by enlarging the new element image set from which the obstruction information is removed by the control unit by an enlargement ratio using a pinhole arrangement. In this case, since the obstacle image is already removed by the remover 225, the reconstructed image includes only information on the object to be recognized.

6 is a flowchart illustrating a method of restoring an image in the integrated image restoration apparatus according to an embodiment of the present invention. Each step described below is performed by each internal component of the integrated image restoration apparatus, but will be collectively referred to as an integrated image restoration apparatus for the convenience of understanding and explanation.

In operation 610, the image reconstruction apparatus 200 obtains a set of element images from a predetermined object. Acquisition of a set of element images may be obtained by recording to an image acquisition device (eg, a CCD camera) through a lens array as described above. In this case, one element image may be generated through each lens, and the element image set may be generated by arranging the generated element images in two dimensions.

In operation 615, the image reconstruction apparatus 200 forms a reconstructed image on each plane by imaging the obtained element image set on a plurality of planes having different distances through a pinhole array. As described above, the image reconstructing apparatus 200 may generate a reconstructed image by forming a set of element images by varying a distance and forming an image on each plane.

In operation 620, the image reconstruction apparatus 200 calculates an enlargement ratio by analyzing a correlation between the reconstructed images and the reference image.

For example, the image reconstruction apparatus 200 calculates a correlation value between the reconstructed image and the reference image, respectively. The image reconstruction device 200 determines a distance corresponding to the reconstructed image corresponding to the correlation value having the largest correlation value as the pickup distance. Subsequently, the image reconstruction apparatus 200 may calculate the enlargement ratio using the determined formula by using the lens formula.

In operation 625, the image reconstruction apparatus 200 opens the pinhole arrays one by one to enlarge the element images according to an enlargement ratio, and masks the reference images to remove the obstacle information to generate a new element image.

In more detail, the image reconstructing apparatus 200 removes the obstacle information by opening the pinhole arrays one by one, magnifying the element images one by one according to an enlargement ratio, and masking the reference images. In this manner, the image reconstruction apparatus 200 generates a new element image by masking all element images of the element image set to remove obstacle information.

As described above, a set of new element images from which obstruction information is removed through a masking process will be referred to as a new element image set.

In operation 630, the image restoration apparatus 200 restores the new element image set according to the computer integrated image restoration method.

7 is a diagram illustrating an image reconstructed by an image reconstruction device according to an embodiment of the present invention, and FIG. 8 is an exemplary diagram of a correlation graph for finding a distance of an obstacle according to an embodiment of the present invention.

Referring to FIG. 7, (a) of FIG. 7 is a set of element images. The element image set illustrated in (a) of FIG. 7 is a set of picked up element images. FIG. 7B illustrates a reference image (ie, obstruction) reconstructed at a distance of 9 mm. FIG. 7C illustrates an image of an object to be recognized which is reconstructed at a distance of 12 mm.

8 is a graph showing a correlation value between a reference image (ie, an obstacle) and reconstructed images reconstructed by changing a reconstruction distance. As shown in FIG. 8, it can be seen that the correlation value is the highest when the restoration distance is 9 mm. Through this result, it is possible to extract the position of the obstruction is located 9mm away from the lens array, the reference ratio can be calculated as 3 using this.

FIG. 9 is a diagram illustrating an image restored by an image restoration apparatus according to another embodiment of the present invention, and FIG. 10 is a graph showing a correlation before and after removing an obstacle according to an embodiment of the present invention.

FIG. 9 (a) is a novel element image set from which obstruction information is removed for the picked-up element image sets illustrated in FIG. 7 (a), and FIG. 9 (b) is a reconstruction for the obstruction restored at a distance of 9 mm. An image (that is, a reconstructed image of an obstruction from which obstruction information is removed) and FIG. 9C is a reconstructed image of an object to be recognized from which obstruction information has been removed.

Comparing FIG. 7 (c) with FIG. 9 (c), it can be seen that most information on the obstacles has been removed. The reason why the information on the obstacle is not completely removed in FIG. 9 (c) is caused by the inability to completely remove the boundary point due to the blocking phenomenon that occurs when the element image is enlarged.

10 (a) is a graph showing the correlation before removing the information on the obstacle, Figure 10 (b) is a graph showing the correlation after removing the information on the obstacle. When comparing (a) and (b) of FIG. 10, it can be seen that the correlation graph in FIG. 10 (a) has many peaks, and it is not easy to distinguish between desired values and other values. However, FIG. 10 (b) has a steep peak, and it can be seen that the recognition rate is improved after removing the obstacle information through the graph observation.

Although the above has been described with reference to a preferred embodiment of the present invention, those skilled in the art to which the present invention pertains without departing from the spirit and scope of the present invention as set forth in the claims below It will be appreciated that modifications and variations can be made.

1 illustrates the principle of a typical three-dimensional integrated imaging system.

2 is a block diagram of an image restoration apparatus according to an embodiment of the present invention.

3 is a diagram illustrating a reconstruction technique in an image generating unit according to an embodiment of the present invention.

4 is a diagram illustrating an image reconstructed according to a distance according to an embodiment of the present invention.

5 is a diagram for explaining masking according to an embodiment of the present invention;

6 is a flowchart illustrating a method of restoring an image in an integrated image restoration apparatus according to an embodiment of the present invention.

7 is a diagram illustrating an image restored by an image restoration apparatus according to an embodiment of the present invention.

8 is an exemplary diagram of a correlation graph for finding a distance of an obstacle in accordance with an embodiment of the present invention.

9 is a diagram illustrating an image reconstructed by an image reconstruction device according to another embodiment of the present invention.

10 is a graph showing a correlation before and after removing an obstruction according to an embodiment of the present invention.

<Description of the symbols for the main parts of the drawings>

210: pickup section

215: image generating unit

220: output unit

225: removal unit

230: restoration unit

Claims (17)

In the method for restoring an image by the image restoration apparatus, (a) generating a reconstructed image on each of the planes by forming a set of element images obtained from a predetermined object on a plurality of planes having a different distance from the image reconstructing transmission body through the image reconstructing transmission body; (b) calculating an enlargement ratio by analyzing correlations with respect to the reference image and the reconstructed images, respectively; And (c) enlarging the set of element images at the enlargement ratio, and masking the reference image to the enlarged image to remove obstruction information. According to claim 1, And the image reconstructing transmission body is a pinhole array including a plurality of pinholes or a lens array including a plurality of lenses. According to claim 1, Before generating the reconstructed image on the plane, And acquiring the element image set from the predetermined object through an image acquisition transmitting body. According to claim 1, Determining an enlargement ratio by analyzing correlations with respect to the reference image and the reconstructed images, Calculating a correlation value for each reconstructed image and the reference image, respectively; Determining a distance corresponding to a reconstructed image corresponding to a correlation value having the largest correlation value as a pickup distance; And And calculating an enlargement ratio using the determined pickup distance. According to claim 1, The enlargement ratio is calculated using the following equation. M = L / g L is a distance from the pinhole array to the reconstructed image, and g is a distance from the element image to the pinhole array. According to claim 1, Enlarging the element image set at the enlargement ratio, and masking the reference image to the enlarged image to remove the obstacle information, (a) opening the pinhole arrays one by one to enlarge the element images in the element image set at the enlargement ratio; And and (b) masking the enlarged image and the reference image to remove obstruction information from the enlarged image. The method of claim 6, (c) further comprising generating a new element image by reducing the enlarged image from which the obstacle information is removed using the enlargement ratio, The steps (a) to (c) are repeated n times, wherein n is the number of element images included in the element image set. The method of claim 7, wherein And reconstructing the object to be recognized using the plurality of new element images. In the image restoration apparatus, An image generation unit configured to form a set of element images obtained from a predetermined object on a plurality of planes having a different distance from the image restoration transmission body through an image restoration transmission body to generate a restoration image on each plane; A calculator configured to calculate a magnification ratio by analyzing correlations of the reference image and the reconstructed images, respectively; And And a remover configured to enlarge the element image set at the enlargement ratio and to remove the obstacle information by masking the reference image on the enlarged image. The method of claim 9, And the image reconstructing transmission body is a pinhole array including a plurality of pinholes or a lens array including a plurality of lenses. The method of claim 9, And a pickup unit for acquiring the element image set from the predetermined object through an image acquisition transmitting body. The method of claim 9, The calculator calculates a correlation value for each of the reconstructed image and the reference image, and determines a pickup distance as a distance corresponding to the reconstructed image corresponding to the correlation value having the largest correlation value. And an enlargement ratio is calculated using the determined pickup distance. The method of claim 9, The magnification ratio is calculated using the following equation M = L / g L is a distance from the pinhole array to the reconstructed image, and g is a distance from the element image to the pinhole array. The method of claim 9, The removal unit opens the pinhole arrays one by one to enlarge the element images in the element image set at the enlargement ratio, and masks the enlarged image and the reference image to remove the obstacle information from the enlarged image. Device. The method of claim 14, The removal unit reduces the enlarged image from which the obstacle information is removed using the enlargement ratio to generate a new element image. n is repeatedly performed, and n is the number of element images included in the element image set. The method of claim 15, And a reconstruction unit for reconstructing the object to be recognized using the plurality of new element images. A program of instructions that can be executed by a computer is tangibly embodied to perform a method of removing an obstacle included in a reconstructed image, and in the computer-readable recording medium recording the program, Generating a reconstructed image on each of the planes by forming a set of element images obtained from a predetermined object on a plurality of planes having a different distance from the image reconstructing transmission body through the image reconstructing transmission body; Calculating an enlargement ratio by analyzing correlations with respect to the reference image and the reconstructed images, respectively; And And recording a program for enlarging the set of element images at the enlargement ratio and removing the obstacle information by masking the reference image on the enlarged image.

Images