KR20070079215A - Image processing device - Google Patents

Abstract

An image processing device having a differential application filter and a method therefor are provided to improve the good degree of image quality in comparison to a compression rate by performing filtering in consideration of region-classified importance. N LPFs(Low Pass Filters)(211,212) have inter-different cutoff frequencies. A region designation register(115) stores a predetermined register value corresponding to one of the N LPFs(211,212) according to classification regions of image data. The classification regions as partial regions configuring the image data have a certain horizontal pixel number and a certain vertical pixel number. A selection unit(220) extracts the register value for the classification region corresponding to a position of a certain pixel of the inputted image data from the region designation register(115), and selects and outputs result pixel data by the LPF corresponding to the extracted register value among N result pixel data generated after certain pixel data are differentially filtered by the N LPFs(211,212).

Description

Image processing device and method having a differential application filter unit

1 is a block diagram of an image processing apparatus according to an exemplary embodiment of the present invention.

FIG. 2 is a diagram illustrating a differential applying filter unit in the image processing apparatus of FIG. 1. FIG.

3 is a diagram illustrating a region classification method of image data for performing differential image filtering according to an exemplary embodiment of the present invention.

4 is a diagram illustrating a register value preset for each division area in the case of FIG.

FIG. 5 is a flowchart illustrating an image filtering process in the differential applying filter unit illustrated in FIG. 2.

6 to 8 are views comparing the good quality of the compression ratio compared to the case of the image filtering through the differential application filter unit according to an embodiment of the present invention and the image filtering in the conventional manner.

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

100: image processing device

110: preprocessing unit

115: differential application filter

120: compression unit

130: storage unit

211: weak low pass filter

212 strong low pass filter

220: selection unit

230: area designation register

The present invention relates to an image processing apparatus and method, and more particularly, to an image processing apparatus and method having a differential application filter.

Image processing apparatuses are used in various digital processing apparatuses, such as digital cameras, mobile phones, and computers, and are particularly related to image compression technology. Recently, with the development of wireless communication technology, the communication environment using a still image or a moving image is rapidly spreading, while users want a high quality image, but the amount of data to be transmitted to realize this is increasing. In order to solve this problem, various image compression methods including JPEG (Joint Photographic Experts Group) and Moving Picture Experts Group (MPEG) have been proposed, and these image compression techniques have been internationally standardized and used. Accordingly, an image processing apparatus generally includes a codec for compressing or reconstructing image data.

However, the image processing apparatus does not immediately compress the input image data, and generally performs image processing (hereinafter, referred to as 'preprocessing') of the image data before performing image compression. For example, the image processing apparatus performs preprocessing of the image data, such as scaling the image data or removing noise included in the image data through the preprocessor. Accordingly, the preprocessor of the image processing apparatus includes a scaler for adjusting the size of the image data, a filter unit for removing noise, and the like. In this case, the filter unit is generally implemented using a low pass filter in consideration of the characteristics of the noise composed of high frequency components.

However, the filter unit included in the conventional image processing apparatus is implemented with only one low pass filter. When the strength of the lowpass filter is strong (that is, when the cutoff frequency of the lowpass filter is relatively small), the ability to remove high frequency noise is improved, but at the same time, the image data having a frequency component larger than the cutoff frequency of the lowpass filter is shown. Since the image information is also removed, there is a problem that the distortion of the image is severe. On the other hand, when the strength of the low pass filter is weak (that is, when the cutoff frequency of the low pass filter is relatively large), the image data can be preserved without significant distortion, but the noise canceling ability is degraded and subsequent image compression process is performed. There is a problem that the compression ratio of the compressed image data generated through the drop.

In addition, the conventional image filtering method through one low pass filter has a limitation in that it is impossible to consider the importance of each region in the image data. For example, as the person to be photographed usually exists in the center of the captured image, the user generally pays attention to the image data displayed in the center portion of the screen, and the importance of the image of the peripheral portion of the screen is low. Therefore, it is necessary to preserve the areas of high importance in the image data without distortion as much as possible than the areas of low importance. However, in the conventional method, since all regions of the image data are filtered with the same intensity by one filter, the importance of the regions of the image data cannot be considered.

Accordingly, an object of the present invention is to provide an image processing apparatus and method capable of performing differential filtering according to the importance of each region of image data.

Another object of the present invention is to provide an image processing apparatus and method capable of improving the goodness of image quality compared to the compression rate by performing filtering considering the importance of each region.

Still another object of the present invention is to provide an image processing apparatus and method capable of generating various image effects such as highlighting a subject in an image by performing differential filtering for each region.

Another object of the present invention can be easily implemented by simply adding one or more low pass filters and a selection unit to an existing image processing device, and can reduce the burden on the image storage device by improving the image compression rate. It is to provide a processing apparatus and method.

Other objects of the present invention will be readily understood through the following description.

According to one aspect of the invention, there are provided N low pass filters, each having a different cutoff frequency, wherein N is at least two natural numbers; An area designation register that stores a predetermined register value corresponding to any one of the N low pass filters of each of the N low pass filters for each division area of the image data. And regions having a predetermined number of vertical pixels; And N result pixel data generated by extracting a register value for a division region corresponding to a position of an arbitrary pixel of the input image data from the region designation register, and random pixel data being differentially filtered by the N low pass filters. An image processing apparatus having a differential applying filter unit including a selector for selecting and outputting result pixel data by a low pass filter corresponding to the extracted register value may be provided.

Here, the low pass filter may be implemented as a finite impulse response (FIR) filter, and the cutoff frequency of the low pass filter may be different through adjustment of the coefficient of the FIR filter.

Here, the selection unit may include a horizontal pixel counter and a vertical pixel counter, but may calculate the position of the pixel through calculation through the horizontal pixel counter and the vertical pixel counter.

An image processing apparatus having a differential applying filter unit according to the present invention includes two low pass filters, wherein a register value stored in an area designation register is different from a register value of a first region and a second region of image data. The register value of the first region corresponds to a low pass filter having a relatively large cutoff frequency of two low pass filters, and the register value of the second region corresponds to a low pass filter having a relatively small cutoff frequency of two low pass filters. May correspond to

Here, the first region may be formed in the center portion of the image data, and the second region may be formed in the region other than the center portion of the image data.

Here, the number of division areas constituting the first area and the second area may be determined by the number of area designation registers and the number of bits in each register.

According to another aspect of the invention, (a) generating N result pixel data differentially filtered by the N low pass filters for any one pixel data of the input image data, wherein N low pass Filters have different cutoff frequencies; (b) extracting from the region designation register a preset register value for the segmentation region corresponding to the position of any pixel; And (c) selecting the resulting pixel data by the low pass filter corresponding to the extracted register value among the N low pass filters, wherein the register value corresponds to the low pass filter of any one of the N low pass filters. There may be provided an image processing method in an image processing apparatus including-, but having a differential applying filter unit in which steps (a) to (c) are repeatedly performed on all pixels of the image data.

The following merely illustrates the principles of the invention. Thus, although not clearly described or illustrated herein, one of ordinary skill in the art would be able to invent various methods that implement the principles of the invention and are included in the concepts and scope of the invention and apparatuses using the same. In other words, it is to be understood that all details, including the principles, aspects, and embodiments of the present invention, as well as listing specific embodiments, are intended to include structural and functional equivalents. In addition, in describing the present invention, when it is determined that the detailed description of the related known technology may unnecessarily obscure the gist of the present invention, the detailed description thereof will be omitted.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings, and the same or corresponding components will be denoted by the same reference numerals regardless of the reference numerals and redundant description thereof will be omitted.

1 is a block diagram of an image processing apparatus according to an exemplary embodiment of the present invention. The image processing apparatus 100 according to the present invention may be used for any device that requires processing of digital images such as a mobile communication terminal, a digital camera, a computer, a game terminal, and the like.

Referring to FIG. 1, the image processing apparatus 100 according to the present invention includes a preprocessing unit 110 including a differential applying filter unit 115, a compression unit 120, and a storage unit 130.

The preprocessor 110 performs image processing on the raw image data input from the image sensor (not shown) before compressing the image data. The data processed by the preprocessor 110 may be stored in the storage 130.

Here, the raw image data input to the preprocessor 110 may be a digital image signal converted from a raw image captured by an image sensor such as a CCD sensor.

The preprocessor 110 includes a differential application filter 115. The differential application filter unit 115 may increase the compression rate of the compressed image data generated through the subsequent compression process by varying the intensity of the image filtering according to the importance of each region of the raw image data. The differential application filter unit 115 will be described in detail with reference to FIG. 2.

In FIG. 1, the differential filter unit 115 is included in the pretreatment unit 110, but it may be provided separately from the pretreatment unit 110.

The compression unit 120 encodes image data processed by the preprocessor 110 or decodes the compressed image data into original image data. The compression unit 120 may have an image codec for compressing or restoring image data. The image codec may be any one of various image codecs such as JPEG and MPEG. Image data compressed by the compression unit 120 may be stored in the storage unit 130. As such, the compression unit 120 may reduce the load of the storage unit 130 by reducing the storage capacity of the image data to be stored.

The storage unit 130 includes a plurality of memories, and stores image data processed by the preprocessor 110 or the compressor 120. Each data processed by the preprocessor 110 or the compressor 120 may be stored in a separate storage space having different addresses in the storage 130.

In addition to the components illustrated in FIG. 1, it is obvious that the image processing apparatus 100 according to the present invention may further include various components for image processing. For example, the image processing apparatus 100 may include a scaler for adjusting the size of the image data, a Bayer filter for arranging the image data in a Bayer pattern, and an interpolation unit for performing inter-pixel interpolation of the image data ( Interpolation), gamma correction and color / luminance correction for image data, and color correction and color conversion unit for converting RGB image data into YUV image data (RGB to YCbCr). Converter) may be further included. All or some of these components may be included in the preprocessor 110 or may be included in the image processing apparatus 100 separately from the preprocessor 110. Of course, it may be stored in the image processing apparatus 100 in the form of a program that performs the function of such a component.

In order to perform the above-described image processing process, the image processing apparatus 100 may be controlled by a controller that controls the operation of the entire digital processing apparatus. Of course, the image processing process may be controlled by providing a separate image controller inside the image processing apparatus 100.

FIG. 2 is a diagram illustrating a differential applying filter unit in the image processing apparatus of FIG. 1. In the description of all the drawings below FIG. 2, the case where two low-pass filters having different cutoff frequencies are provided in the differential application filter unit 115 as shown in FIG. 2 will be described. It will be readily apparent to those skilled in the art that the following description may be implemented in the form of three or more low pass filters having a frequency. In the following description, since the configuration and function of the low pass filter and the multiplexer are obvious to those skilled in the art, a detailed description thereof will be omitted.

Referring to FIG. 2, the differential application filter unit 115 according to the present invention includes two low pass filters 211 and 212, a selection unit 220, and an area designation register 230.

The two low pass filters 211, 212 have different cut-off frequencies. The weak low pass filter 211 means a low pass filter having a relatively large cutoff frequency of the two low pass filters, and the strong low pass filter 212 means a low pass filter having a relatively small cutoff frequency. When the image data passes through the strong low pass filter 212, more of the high frequency components included in the image data are blocked than when passing through the weak low pass filter 211.

In more detail, since the low pass filter cuts without passing a frequency component above the cutoff frequency of the filter, the lower the cutoff frequency, the smaller the cutoff frequency is. Information). Therefore, the more the use of the strong low pass filter, the greater the amount of high frequency image information and noise removed, and the smaller the size of the compressed image data generated through the subsequent compression process. That is, when a strong low pass filter is used, the compression rate of the image data can be increased. However, a large amount of video information removed from the original video data also means that the image-processed data is severely distorted from the original video data. On the contrary, when the weak low pass filter is used, the degree of image distortion can be reduced, but the compression rate and noise removal efficiency are inferior. For this reason, when only one low pass filter is used, the two requirements of improving the compression ratio of image data and preventing image distortion cannot be simultaneously achieved. This, in turn, means that the manufacturer has no choice but to select a cutoff frequency that will make the most of this requirement in the manufacture of the image processing device.

However, the differential application filter unit 115 according to the present invention may select more image characteristics required for each region according to the importance of each region of the image data by using two low pass filters 211 and 212. The differential application filter 115 selects the resultant image by the weak low pass filter 211 to prevent distortion of the image in an area of high importance of information among all image data, and selects an image in an area of low importance of information. Since distortion prevention is not greatly required, the resultant image of the strong low pass filter 212 may be selected to improve the compression ratio.

For example, a person to be photographed is usually present in the center of the photographed image. This is because the user places the photographing target in the center of the screen when the user photographs using the camera or the like. For this reason, more important information is located in the center of the captured image than in the peripheral. Also, when viewing a photographed image, people generally pay more attention to the center portion of the image than the surrounding portion, regardless of the importance of the above-described information. Since the user places more emphasis on the image data located in the center of the image, the user can say that the image data of the center portion of the image is of high importance and the image data of the peripheral portion of the image is of low importance. . Therefore, the differential filter unit 115 according to the present invention can prevent distortion of the image by selecting the resultant image filtered through the weak low pass filter 211 for the central portion of the image having high information importance. In addition, the compression ratio may be improved by selecting a resultant image filtered through the strong low pass filter 212 for the peripheral portion of the image having low information importance. Through the differential filtering process using the differential filter unit 115 as described above, the user may feel that the image data of the image data having the same compression ratio is better than the image filtered using one low pass filter (FIG. 6). To FIG. 8).

Here, the low pass filters 211 and 212 may be implemented by various digital filters including a finite impulse response (FIR) filter. In particular, for FIR filters, their output depends only on current and past inputs, eliminating the need for a feedback loop to ensure filter stability. Despite the disadvantage of increasing the order of Infinite Impulse Response (IIR) filters to achieve the same amplitude characteristics, FIR filters are suitable for use in the design of lowpass filters in terms of their implementation simplicity and filter stability. can do.

In addition, the FIR filter has an advantage of changing the strength of the low pass filter implemented by simple adjustment of the weight coefficient of the impulse response indicating the characteristic of the filter. Here, the strength of the low pass filter is related to the cutoff frequency of the filter.

For example, the impulse response of the 3rd order FIR filter is expressed by Equation 1 below, and by adjusting different coefficient values of the impulse response, it is possible to simply select the magnitude of the cutoff frequency of the low pass filter. Coefficient adjustment of the impulse response of the FIR filter can be easily performed by changing a setting value of a register (not shown) separately provided in the image processing apparatus 100 without changing the filter configuration.

y [n] = {A x [n] + B x [n-1] + C x [n-2]} / (A + B + C)

Here, x [n], x [n-1] and x [n-2] mean each pixel sequentially input to the FIR filter, and A, B and C are coefficients of an impulse response.

The selector 220 includes a horizontal pixel counter 221, a vertical pixel counter 222, and three multiplexers (hereinafter, 'MUX1', as shown in FIG. 2). 'MUX2', short for 'MUX3').

The horizontal pixel counter 221 calculates the horizontal order in which any input pixel is positioned throughout the image data, and the vertical pixel counter 222 locates the same pixel across the horizontal pixel counter 221 in the entire image data. Calculate the vertical order. The selector 220 may determine a position where any input pixel exists in the entire image data using the horizontal pixel counter 221 and the vertical pixel counter 222. In addition, the selector 220 may extract, from the region designation register 230, a preset register value corresponding to the position of the pixel identified above using MUX1 and MUX2. According to the extracted register value, the selector 220 may select one of the resultant images that pass through the weak low pass filter 211 and the strong low pass filter 212 using MUX3.

The function of the differential applied filter unit 115 shown in FIG. 2 and the organic connection relationship between the components thereof will become more apparent through the description of FIGS. 3 to 5 below.

FIG. 3 is a diagram illustrating an area classification method of image data for performing differential image filtering according to an exemplary embodiment of the present invention, and FIG. 4 is a diagram illustrating a preset register value for each division area in FIG. 3. One drawing.

As shown in Fig. 3, it is assumed that the total image data has a size of 640 horizontal pixels and 480 vertical pixels (640 x 480). To this end, the size of the image data may be adjusted to have a size of 640 × 480 through a scaler (not shown) of the image processing apparatus 100. In addition, it is assumed that an area corresponding to a size of 32 horizontal pixels and 32 vertical pixels (32 × 32) is selected as a unit area for area division of image data. In the following description, (m, n) indicates the position of the pixel with respect to the origin (0, 0) of the image data. Here, m means the number of horizontal pixels having a value between 0 and 639, and n means the number of vertical pixels having a value between 0 and 479.

All image data illustrated in FIG. 3 is divided into 20 regions (640 (total horizontal pixels) / 32 (unit horizontal pixels)) in total in the horizontal direction (hereinafter, referred to as a 'horizontal line'), and is vertical. Direction is divided into 15 regions (480 (total number of vertical pixels) / 32 (number of vertical pixels)) in the direction (hereinafter referred to as 'vertical line'). Accordingly, the image data may be divided into 300 divided regions (20 horizontal lines x 15 vertical lines). The division areas mean partial areas of the image data having a size (32 × 32) of the unit area.

All pixels corresponding to a total of 15 vertical lines (from the first vertical line to the fifteenth vertical line) of the image data are each assigned to separate registers. That is, all pixels corresponding to the first vertical line (ie, pixels 0 (m, 0) to 31 (m, 31)) based on the origin of the image data are allocated to the AREA_SEL_00 [19: 0] registers. All pixels corresponding to the second vertical line (that is, pixels 32 (m, 32) to 63 (m, 63)) are allocated to the AREA_SEL_01 [19: 0] register and finally correspond to the 15th vertical line. All pixels (ie, pixels 448 (m, 448) to 479 (m, 479)) are assigned to the AREA_SEL_14 [19: 0] register.

Here, the AREA_SEL_00 [19: 0] to AREA_SEL_14 [19: 0] registers each consist of 20 bits, and each 1 bit constituting one register has a horizontal line (first'horizontal line) of image data. To 20 'horizontal line). For example, in the case of the AREA_SEL_00 [19: 0] register, the AREA_SEL_00 [0] bits belong to the first vertical line, and all pixels corresponding to the first 'horizontal line with respect to the origin of the image data (that is, zero (0) k) to 31 (31, k) pixels are allocated, and in the AREA_SEL_00 [1] bits, all pixels corresponding to the second 'horizontal line as belonging to the first vertical line (that is, 32 (32, k)) 63 (63, k) pixels are allocated, and finally, in the AREA_SEL_00 [19] bit, all pixels corresponding to the 20 'horizontal line as belonging to the first vertical line (ie, 608 (608, k) ~) 639 (639, k) pixels) are allocated. Here, k means the number of vertical pixels having a value between 0 and 31.

As illustrated in FIG. 4, a predetermined value of "0" or "1" is stored in each bit of the AREA_SEL_00 [19: 0] to AREA_SEL_14 [19: 0] registers allocated to each division area of the image data.

For example, bits AREA_SEL_02 [0] to AREA_SEL_02 [19] configuring the AREA_SEL_02 [19: 0] register of FIG. 4 are set to "1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 1 1 1 1 1 Values such as 1 "are stored in order one bit at a time. A value of "0" or "1" (hereinafter, referred to as a 'register value') stored in each bit corresponds to either the weak low pass filter 211 or the strong low pass filter 212, respectively. By setting the register value as described above, the differential application filter unit 115 may perform differential image filtering for each region of the image data. The differential image filtering method for each region will be described in detail with reference to the flowchart of FIG. 5.

In FIG. 3 and FIG. 4, the size of the image data is assumed to be 640 × 480, and the size of the unit area is 32 × 32. However, the present invention is not limited thereto and various applications are possible. That is, the number and size of registers required may be different according to the size of the image data and the size of the set unit region. If the number and size of commercial registers is limited, the number and size of necessary registers may be reduced by increasing the size of the unit area (that is, by dividing each divided area of the image data into larger ones).

In addition, since only two low pass filters are used in FIGS. 3 and 4, only one bit is required for setting a corresponding register value ("0" or "1"). Of course, more than two bits may be required to set the corresponding register value corresponding to the number of low pass filters.

FIG. 5 is a flowchart illustrating an image filtering process in the differential applying filter unit illustrated in FIG. 2. The image filtering process (steps S501 to S505) through the differential applying filter of FIG. 5 is performed in such a manner that the pixels are sequentially input one by one until all the pixels constituting the image data are filtered. Only the filtering process for data of any one pixel (hereinafter referred to as 'raw pixel data') will be described. The image data input to the differential application filter unit 115 may have a YCrCb form converted from RGB color image data through a preprocessing process of the image processing apparatus 100.

In S501, the raw pixel data is input to each of the weak low pass filter 211 and the strong low pass filter 212, and simultaneously to the horizontal pixel counter 221 in the selector 220.

The raw pixel data is filtered by different cutoff frequencies, respectively, passing through two low pass filters 211 and 212. Two different pixel data (hereinafter, referred to as 'result pixel data') that are filtered for the input raw pixel data are generated. The generated two result pixel data will be selected according to the corresponding register value only through the step S504 and step S505 to be described later.

At the same time, the raw pixel data passes through the horizontal pixel counter 221, and the horizontal pixel counter 221 calculates the horizontal order in which the raw pixel data is located in the entire image data. The raw pixel data that has passed through the horizontal pixel counter 221 passes through the vertical pixel counter 222 again, and the vertical pixel counter 222 calculates an order of the vertical direction in which the raw pixel data is positioned in the image data. As described above, the selection unit 220 may check the position of the raw pixel data existing in the entire image data while passing through the horizontal pixel counter 221 and the vertical pixel counter 222.

In step S502, when the position of the raw pixel data is confirmed, the selector 220 extracts from the region designation register 230 a corresponding register value preset in accordance with the position of the pixel.

For example, the raw pixel data input by the selector 220 is positioned 620th in the horizontal direction and 70th in the vertical direction with respect to the origin (0, 0) of the entire image data (ie, (620). , 70). In this case, the raw pixel data 620 and 70 belong to a 20 'horizontal line (ie, pixels 608 to 639) and a third vertical line (ie, pixels 64 to 95). Accordingly, the selector 220 determines that the corresponding bit in the register included in the 20 'horizontal line as belonging to the third vertical line (ie, corresponding to the AREA_SEL_02 [19: 0] register) is the AREA_SEL_02 [19] bit. (See FIG. 4). As a result, the selector 220 may extract a preset register value (“1” in FIG. 4) stored in the corresponding bit of the corresponding register corresponding to the position of the raw pixel data identified using MUX1 and MUX2. Will be.

In step S503, the selector 220 determines whether the extracted register value is "1". When the extracted register value is "1", the selector 220 selects the resultant pixel data by the strong low pass filter 212 using MUX3 (step S504), and the extracted register value is "1". If not (ie, "0"), the resultant pixel data by the weak low pass filter 211 is selected (step S505).

That is, in general, considering that the video data for a video call is located in the center and the center information exists in the center of the video, the pixel data located at the center of the screen is processed by the weak low pass filter 211. In addition, the pixel data located at the periphery may be processed by the strong low pass filter 212. In this way, by opening the iris and taking a thinner depth when taking a picture, it may be possible to create an effect similar to the picture that emphasizes the subject. Of course, in addition to this, by setting various image areas to be processed by each filter, various effects may be produced.

6 to 8 are diagrams comparing the goodness of the image quality with the compression ratio when the image filtering through the differential application filter unit according to an embodiment of the present invention and the image filtering by the conventional method. FIG. 6 shows a raw image, FIG. 7 shows a processed image when the image is filtered according to a conventional method, and FIG. 8 shows a processed image when the image is filtered through the differential applying filter 115 according to the present invention. Display the image. Here, FIG. 8 applies register values for respective regions of the image data illustrated in FIG. 4 in filtering an image through the differential applying filter 115. In addition, the processed images shown in FIGS. 7 and 8 are all generated after the image filtering process and the subsequent JPEG compression process.

The size of the raw image shown in FIG. 6 is 31 kilobytes (KB), and the processed images of FIGS. 7 and 8 each have a size of 24 kilobytes (Kilo Byte). That is, the processed images of FIGS. 7 and 8 have the same image data size after JPEG compression processing.

However, in the case of FIG. 7, the entire raw image data is image filtered by the same intensity (ie, the same cutoff frequency) by one low pass filter, and it is understood that the processed image is generally blurred compared to the raw image of FIG. 6. Can be.

In the case of FIG. 8, the raw image data is subjected to differential image filtering for each region by the region-specific register values illustrated in FIG. 4, and it can be seen that the center portion of the processed image is clearer than that of FIG. 7.

That is, although FIG. 7 and FIG. 8 have the same ratio with respect to the compression efficiency, the user feels that the processed image of FIG. 8 is much higher in quality than the processed image of FIG. Therefore, the image filtering method using the differential applied filter unit according to the present invention has the effect of improving the quality of image quality compared to the compression ratio that the user feels. Such an improvement in the quality of the image quality compared to the compression ratio will be more apparent in the case of a video in which an image of more than 24 frames per second is changed instantly.

In addition, the image filtering method according to the present invention prints the image effect (for example, when the target object is photographed with the aperture of the camera open, such as to highlight the object of the center portion of the entire image than its peripheral portion) As the depth of the entire picture becomes thinner, the peripheral part of the entire image is blurred.

As described above, according to the image processing apparatus and method having the differential application filter unit according to the present invention, there is an effect that can consider the importance of each area of the image data in the image processing.

In addition, the present invention has an effect of improving the good quality of the compression ratio compared to the compression ratio by performing filtering considering the importance of each region.

In addition, the present invention has an effect that can produce a variety of image effects, such as highlighting the subject of the subject in the image by performing differential filtering for each region.

In addition, the present invention can be easily implemented by simply adding one or more low pass filters and a selection unit to the existing image processing apparatus, and has an effect of reducing the burden on the image storage device by improving the image compression ratio. .

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.

Claims (8)

N lowpass filters, each having a different cutoff frequency, wherein N is a natural number of two or more; An area designation register for storing a predetermined register value corresponding to any one of the N low pass filters for each of the division areas of the image data, wherein the division area is a partial area constituting the image data. Means regions having a horizontal number of pixels and a predetermined vertical number of pixels; And N registers generated by extracting the register value for the divided region corresponding to the position of any pixel of the input image data from the region designation register, and the arbitrary pixel data being differentially filtered by the N low pass filters. And a selector for selecting and outputting result pixel data by a low pass filter corresponding to the extracted register value among result pixel data. The method of claim 1, The low pass filter is an image processing device having a differential applied filter unit, characterized in that implemented as a Finite Impulse Response (FIR) filter. The method of claim 2, And a cutoff frequency of the low pass filter is different by adjusting a coefficient of the FIR filter. The method of claim 1, The selection unit includes a horizontal pixel counter and a vertical pixel counter, And a differential applying filter unit to calculate a position of the pixel through calculation through the horizontal pixel counter and the vertical pixel counter. The method of claim 1, The number of the low pass filter is two, The register value stored in the region designation register is different from the register value of the first region and the register value of the second region of the image data, and the register value of the first region is a relatively large cutoff of the two low pass filters. A low pass filter having a frequency, and a register value of the second region corresponds to a low pass filter having a relatively small cutoff frequency among the two low pass filters. . The method of claim 5, And the first area is formed at a center portion of the image data, and the second area is formed at a region other than the center portion of the image data. The method of claim 5, And the number of division areas respectively constituting the first area and the second area is determined by the number of the area designation registers and the number of bits of each register. (a) generating N result pixel data differentially filtered by the N low pass filters for any one pixel data of the input image data, wherein the N low pass filters each have a different cutoff frequency. Has-; (b) extracting from the region designation register a preset register value for the division region corresponding to the position of the arbitrary pixel; And (c) selecting the resultant pixel data by the low pass filter corresponding to the extracted register value among the N low pass filters, wherein the register value is a low pass filter of any one of the N low pass filters. Matches-, but And (c) the step (a) to the step (c) are repeated for all the pixels of the image data.

Images