KR20120007639A - Image processing apparatus and method of extracting high resolution brightness image in charge-coupled device - Google Patents

Abstract

PURPOSE: An image processing apparatus for extracting a brightness image with high definition by a CCD(Charged Coupled Device) using an interlace scan complementary color filter and a method thereof are provided to effectively maintain high definition by adaptive filtering in consideration of the features of an input image. CONSTITUTION: A low band filtering unit(220) performs low band filtering on a first input image and a second input image. A rate calculating unit(230) calculates the rate of the second input image to the first input image. A pixel predicting unit(240) predicts a pixel. A pixel combining/extracting unit(250) combines a pixel of high definition by the predicted pixel. The pixel combining/extracting unit extracts a brightness image of high definition.

Description

IMAGE PROCESSING APPARATUS AND METHOD OF EXTRACTING HIGH RESOLUTION BRIGHTNESS IMAGE IN CHARGE-COUPLED DEVICE}

The present invention relates to an image processing apparatus and method for analyzing a current input image and extracting a high resolution luminance image by adaptively performing low band filtering on an image having a high band frequency and an image having only a low band frequency. will be.

Background Art [0002] In recent years, imaging devices have been used to acquire images using an electrical light receiving device such as a charge coupled device (CCD) or a complementary metal oxide semiconductor (CMOS) instead of a film.

As digital cameras and digital camcorders have been developed as a representative photographing device using an electric light receiving element, a technology of photographing a subject by adjusting the charge accumulation time of the light receiving element is being used instead of a mechanical shutter that blocks the film photosensitive surface from light. .

The CCD (CHARGE-COUPLED DEVICE) using the interlaced scanning CMYG complementary color filter array (CFA) is more sensitive to light brightness than the CCD using the primary color CFA.

Due to this characteristic of the complementary color CFA, the complementary color CFA is mainly used in the field of video security than the primary color CFA.

However, in order to overcome the limitation of the interlaced scanning method, a loss may occur in terms of resolution due to the characteristics of the complementary CFA that outputs a signal by summing signals of upper and lower pixels.

For example, even when a monochromatic region is photographed, the output signal of the CMYG CCD may have high and low results because the color sensitivity of each CFA pixel is different.

Therefore, when the low band filter is not used, there is a possibility that the characteristics of the output signal are reflected in the image.

In the extraction of luminance, it is common to use a low pass filter such that such a characteristic pattern is not seen. However, if the high frequency part is attenuated due to the characteristics of the low band filter, a loss in terms of resolution is inevitable since the attenuation effect also appears for an image captured with the actual high frequency part.

As the high frequency attenuation of the low band filter disappears, the high frequency portion of the original image disappears. To reduce the problem, conventionally, the size of the low band filter is increased.

Increasing the size of the low band filter may reduce the attenuation of the high frequency region, but the ripple and ringing effects caused by the large size of the filter may be problematic.

An image processing apparatus according to an embodiment of the present invention includes an input image collecting unit for collecting an input image and a luminance image extracting apparatus for checking a frequency characteristic of the collected input image and extracting a high-resolution luminance image through low band filtering. Characterized in that it comprises a.

An image processing method according to an embodiment of the present invention includes the steps of collecting an input image, checking frequency characteristics of the collected input image, and extracting a high-resolution luminance image through low-band filtering. do.

According to an embodiment of the present invention, in extracting a luminance image of a CCD using a complementary color CFA of an interlaced scanning method, it is possible to effectively maintain a high resolution through adaptive filtering considering characteristics of an input image.

1 is a diagram illustrating a digital photographing apparatus to which an image processing apparatus is applied according to an embodiment of the present invention.
2 is a block diagram illustrating an apparatus for extracting luminance images according to an embodiment of the present invention.
FIG. 3 is a diagram for explaining a format of an input image photographed by a CCD (CHARGE-COUPLED DEVICE) according to an embodiment of the present invention.
FIG. 4 is a diagram illustrating a signal separation unit separating an input image for each pixel according to an embodiment of the present invention.
FIG. 5 illustrates a case where a current pixel does not include a high band signal in an input image according to an embodiment of the present invention, and FIG. 6 illustrates a case where a current pixel includes a high band signal.
7 is a flowchart illustrating a method of extracting a luminance image according to an embodiment of the present invention.

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

In describing the present invention, when it is determined that detailed descriptions of related known functions or configurations may unnecessarily obscure the subject matter of the present invention, the detailed description thereof will be omitted. Terminology used herein is a term used to properly express a preferred embodiment of the present invention, which may vary according to a user, an operator's intention, or a custom in the field to which the present invention belongs. Therefore, the definitions of the terms should be made based on the contents throughout the specification. Like reference numerals in the drawings denote like elements.

1 is a diagram illustrating a digital photographing apparatus 100 to which an image processing apparatus 130 according to an embodiment of the present invention is applied.

The digital photographing apparatus 100 according to the exemplary embodiment of the present invention may include a lens 110, a light receiving element 120, and an image processing apparatus 130.

The digital photographing apparatus 100 acquires light incident on the lens 110 through the light receiving element 120, and the light receiving element 120 transmits the obtained light to the image processing apparatus 130 in the form of an image signal. Can be.

The image processing apparatus 130 may signal-process the received input image, generate an output image, and output the output image to a storage medium.

In this case, the image processing apparatus 130 may check frequency characteristics of the collected input image and extract a high resolution luminance image through low band filtering.

In other words, the image processing apparatus 130 may extract the high resolution luminance image by performing adaptive low band filtering on the image having the high band frequency and the image having the low band frequency only by analyzing the characteristics of the current image. have.

As a result, the digital photographing apparatus 100 according to an embodiment of the present invention effectively maintains a high resolution through adaptive filtering considering characteristics of an input image in extracting a luminance image of a CCD using a complementary color CFA of interlaced scanning. Can be.

2 is a block diagram illustrating a luminance image extracting apparatus 200 according to an embodiment of the present invention.

The luminance image extracting apparatus 200 according to an embodiment of the present invention may check a frequency characteristic of the collected input image and extract a high resolution luminance image through low band filtering.

The input image of the present invention is photographed through a CCD (CHARGE-COUPLED DEVICE) using a CMYG complementary color filter array (CFA), which will be described in detail with reference to FIG. 3.

FIG. 3 is a diagram illustrating an input image 310 photographed by a CCD (CHARGE-COUPLED DEVICE) according to an embodiment of the present invention.

3, the CFA structure and the signal transmission method of the CCD using the interpolation scanning complementary filter can be confirmed.

In detail, CMYG complementary colors may be disposed in each pixel of the input image 310. Particularly, the G and Mg filters are reversed in the even line (Even field), and the upper and lower pixels are summed in the form of column A in the odd field and B in the even line. 1] can be output as S1 and S2 signals for each column.

[Equation 1]

는 A1의 영역에 대한 S1 신호이고,

는 A1 영역에 대한 S2 신호이며,

는 A2 영역에 대한 S1 신호이고,

는 A2 영역에 대한 S2 신호로 해석될 수 있다.At this time,

Is the S1 signal for the region of A1,

Is the S2 signal for the A1 region,

Is the S1 signal for the A2 region,

Can be interpreted as an S2 signal for the A2 region.

The output S1 and S2 signals may be transmitted to the luminance image extracting apparatus 200 according to an embodiment of the present invention through the horizontal register 320.

Referring back to FIG. 2, the luminance image extractor 200 according to an exemplary embodiment of the present invention may include a signal separator 210, a low band filter 220, a ratio calculator 230, and a pixel predictor 240. ), And the pixel combining / extracting unit 250.

First, the signal separator 210 may separate the input image into a first input image and a second input image for differential processing of the input image.

For example, the signal separator 210 according to an embodiment of the present invention separates an odd-numbered pixel of the input image into the first input image, and divides an even-numbered pixel of the input image into the second input image. Can be separated.

The operation of the signal separator 210 will be described in detail with reference to FIG. 4.

4 is a diagram illustrating that the signal separator 210 separates an input image for each pixel according to an exemplary embodiment.

The signal separator 210 separates the S1 signal and the S2 signal of the input image. This basically serves to bind S1 signals to S1 signals and S2 signals to S2 signals in order to prevent high and low signals of the input signal from being discriminated.

The signal separator 21 may be separated in the form of a simple switch with respect to the specific pixel 410 as shown in FIG. 3. In the case of even-numbered S2, as shown by reference numeral 430, the second input image GS2 may be switched to separate a signal with respect to the entire input image.

The two signals separated into the first input image and the second input image have characteristics of a pure image without a high and low width, and these signals can be expressed as values representing characteristics of the image through respective low-band filters. have.

Referring back to FIG. 2, the low band filtering unit 220 may perform low band filtering on each of the separated first input image and the second input image.

To this end, the low band filtering unit 220 may perform a first low band filter 221 for low band filtering the separated first input image and a second low band filter 222 for low band filtering the second input image. ) May be included.

Each low band filter performing low band filtering based on the average value of the input image may be designed to pass only a low band so as to sufficiently include the characteristics of the surrounding image.

To this end, in order to sufficiently express the characteristics of the surrounding image as shown in [Equation 2], the first low band filter 221 and the second low band filter 222 may be designed through the average filter.

[Equation 2]

In Equation 2, Gs means an isolated signal. That is, Gs1 may mean the separated first input image, and Gs2 may mean the separated second input image. In addition, i means the position of the pixel, for example 'i = 0' indicates the current input pixel. N is the number of pixels to be referred to, the size of the low-band filter, the larger the N, the larger the number of samples to represent the characteristics of the image, the quality of the expression value can be improved.

The ratio calculator 230 may calculate a ratio of the low band filtered first input image and the second input image.

For example, the ratio calculator 230 may calculate a ratio of the first input image and the second input image as shown in [Equation 3].

&Quot; (3) "

That is, the value obtained by the ratio calculation unit 230 is the ratio of the relative signals that passed through each low pass filter, and Ratio1 is the ratio of the S1 signal to the S2 signal, and Ratio2 is the ratio of the S2 signal to the S1 signal. Can mean.

In other words, the respective characteristics of S1 and S2 through the obtained peripheral pixels are calculated by the ratio of the opposite pixel properties of the current pixel in the ratio calculation unit and used to estimate the pixels.

Ideally, if the relative pixel value is known at each pixel position, a luminance image of perfect quality can be derived. That is, when the current input pixel is the S1 pixel, if the S2 pixel at the corresponding position is known, the perfect luminance value at this position can be known.

However, since this is impossible in reality, the luminance value is derived with reference to the surrounding pixels. However, in the present invention, in order to predict the relative pixel value of the current pixel instead of the left and right pixels as shown in [Equation 3], the S1 and S2 pixels of the neighboring pixels are predicted. The ratio of can be calculated.

The pixel predictor 240 may predict the pixel based on the calculated ratio.

In detail, the pixel predictor 240 may predict the relative pixel value by using Ratio1 and Ratio2 obtained through the ratio calculator 230.

To this end, Equation 4 may be used to predict the relative pixel value.

&Quot; (4) "

In the method of estimating the relative pixel in the pixel predictor 240, a pixel value corresponding to the relative ratio with respect to the current pixel may be calculated by calculating a relative ratio between the S1 and S2 pixels of the neighboring pixels. As expressed in Equation 4, S2 ', which is an estimated value of the S2 pixel at the current pixel S1 position, is a product of the ratio of the representative value S1 "of the peripheral S1 pixels to the current pixel S1". Can be calculated.

In this case, the ratio value of the neighboring pixels may be replaced by the ratio calculated by the low pass filter and the ratio calculator 230 of each of the pixels.

Similarly, even when the current pixel is S2, the estimated value S1 'of S1 can be calculated as shown in [Equation 4].

Based on the relative pixels predicted by the pixel predictor 240, the pixel combiner / extractor 250 may generate a luminance signal of a substantial current pixel position.

That is, the pixel synthesizing / extracting unit 250 may extract high-resolution luminance images by synthesizing high-resolution pixels using the predicted pixels.

This is a function performed by the low pass filter in the conventional method. The luminance signal can be extracted using the additive mixing property of the three primary colors of light by adding the S1 pixel and the S2 signal.

The luminance signal finally obtained may be calculated through Equation 5.

[Equation 5]

Referring to [Equation 5], when the current input pixel is S1, the pixel predictor 240 averages the S2 'predicted, and when S2 is the input pixel, averages the S2' and extracts the final luminance value. can do.

As a result, when using the luminance image extracting apparatus 300 according to an embodiment of the present invention, in extracting the luminance image of the CCD using the complementary color CFA of interlaced scanning, through adaptive filtering considering the characteristics of the input image It is possible to effectively extract high resolution luminance images.

FIG. 5 illustrates a case where a current pixel does not include a high band signal in an input image according to an embodiment of the present invention, and FIG. 6 illustrates a case where a current pixel includes a high band signal.

FIG. 5 illustrates a case where only the heights according to the characteristics of S1 and S2 are not changed over all the pixels, and FIG. 6 illustrates a case where the pixel value is increased due to a characteristic change in the current pixel.

As shown in FIG. 5, when the current pixel has a similar value to the neighboring pixel, the estimated S2 'value matches the value of the neighboring pixel and the luminance value may be calculated as an average of the S1 signal and the S2 signal.

As shown by reference numeral 610 of FIG. 6, when the current S1 value has a change in image, the estimated value of S2 'is positioned at a distance away from the current S1 signal by the ratio of the surrounding S1 and S2 signals, and in this case, the luminance calculated. The value may be located where the characteristic of the S1 signal is included.

As a result, when the current signal is different from the characteristics of the surrounding component values, the current signal value is included in the luminance signal at a higher ratio, thereby adaptively performing low pass filtering on the luminance signal.

7 is a flowchart illustrating a method of extracting a luminance image according to an embodiment of the present invention.

In the luminance image extracting method according to an embodiment of the present invention, the input image may be collected, and the high frequency luminance image may be extracted through low band filtering by checking the frequency characteristic of the collected input image.

To this end, in the luminance image extracting method according to an embodiment of the present invention, the input image is collected (step 710), and the input image is divided into a first input image and a second input image for differential processing of the input image. It may be done (step 720).

Each pixel is configured in the form of an A column in an odd-numbered line, and the upper and lower pixels are summed in the form of a B column in an even-numbered line (EVEN field), so that signals S1 and S2 can be confirmed for each column.

In the luminance image extracting method according to an embodiment of the present invention, the S1 signal may be separated into the first input image, and the S2 signal may be separated into the second input image.

The luminance image extracting method according to an embodiment of the present invention may filter the separated first input image and the second input image, respectively (step 730), and the low-band filtered first input image and the second input image. The ratio of the input image may be calculated (step 740).

The ratio of the first input image and the second input image may be calculated as shown in [Equation 3].

In the luminance image extracting method according to an embodiment of the present invention, the pixel may be predicted based on the calculated ratio (step 750), and the above-described Equation 4 may be used to predict the relative pixel value.

In addition, in the luminance image extracting method according to an embodiment of the present invention, a high-resolution luminance image may be extracted by synthesizing a high-resolution pixel using the predicted pixel (step 760).

The luminance image extracting method according to an embodiment of the present invention may be implemented in the form of program instructions that may be executed by various computer means and may be recorded in a computer readable medium. The computer readable medium may include program instructions, data files, data structures, etc. alone or in combination. Program instructions recorded on the media may be those specially designed and constructed for the purposes of the present invention, or they may be of the kind well-known and available to those having skill in the computer software arts. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tape, optical media such as CD-ROMs, DVDs, and magnetic disks, such as floppy disks. Magneto-optical media, and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine code generated by a compiler, but also high-level language code that can be executed by a computer using an interpreter or the like. The hardware device described above may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

As described above, the present invention has been described by way of limited embodiments and drawings, but the present invention is not limited to the above embodiments, and those skilled in the art to which the present invention pertains various modifications and variations from such descriptions. This is possible.

Therefore, the scope of the present invention should not be limited to the described embodiments, but should be determined not only by the claims below but also by the equivalents of the claims.

200: luminance image extractor 210: signal separation unit
220: low band filtering unit 230: ratio calculation unit
240: pixel prediction unit 250: pixel synthesis / extraction unit

Claims (10)

An input image collector configured to collect an input image; And
A luminance image extraction apparatus for checking a frequency characteristic of the collected input image and extracting a high resolution luminance image through low band filtering
Image processing apparatus comprising a. The method of claim 1,
The luminance image extractor,
Signal separation unit for separating the input image into a first input image and a second input image for the differential processing of the input image
Image processing apparatus comprising a. The method of claim 2,
The signal separation unit,
And separating the odd-numbered pixels of the input image into the first input image and separating the even-numbered pixels of the input image into the second input image. The method of claim 2,
A low band filtering unit including a low band filter for filtering the separated first input image and the second input image; And
A ratio calculator configured to calculate a ratio of the low-band filtered first input image and the second input image
The image processing apparatus further comprises. The method of claim 4, wherein
The low-band filter for filtering the separated first input image and the second input image, respectively,
And low-band filtering based on the average value of the input image. The method of claim 4, wherein
A pixel predictor which predicts a pixel based on the calculated ratio; And
A pixel synthesizing / extracting unit for synthesizing a high resolution pixel by using the predicted pixel and extracting a high resolution luminance image
The image processing apparatus further comprises. The method of claim 1,
The luminance image extractor,
And a high resolution luminance image is extracted from a CCD using an interlaced scanning complementary color filter. Collecting an input image; And
Checking a frequency characteristic of the collected input image and extracting a high resolution luminance image through low band filtering
Image processing method comprising a. The method of claim 8,
The step of extracting the high-resolution luminance image,
Dividing the input image into a first input image and a second input image for differential processing of the input image;
Filtering the separated first and second input images, respectively;
Calculating a ratio of the low band filtered first input image and the second input image;
Predicting a pixel based on the calculated ratio; And
Extracting a high-resolution luminance image by synthesizing a high-resolution pixel by using the predicted pixel
Image processing method comprising a. A computer-readable recording medium having recorded thereon a program for performing the method of any one of claims 8 and 9.

Images