KR101271738B1 - Apparatus and method for reducing noise and recording media written by the method for reducing noise - Google Patents

Abstract

A noise reduction apparatus of the present invention comprises a line memory (12) in which data is stored; a line memory (12) for storing pixel values of externally input image data, the number of which is smaller than the number of bits of the pixel value, An encoding section 11 for converting the code obtained by the conversion into a plurality of lines in the line memory 12 and storing the codes obtained by the conversion in the line memory 12 for a plurality of lines; Unit 13 'and a noise reduction processing unit 14 for reducing noise included in image data obtained by decoding by the decoding unit 13'.
The noise reduction apparatus of the present invention enhances the noise reduction effect on image data without changing the capacity of the memory.

Description

TECHNICAL FIELD [0001] The present invention relates to a noise reduction apparatus, a noise reduction method, and a recording medium recording a noise reduction method.

The present invention relates to a noise reduction apparatus, a noise reduction method, and a recording medium on which a noise reduction method is recorded.

Conventionally, in order to reduce noise included in image data, noise reduction processing is performed on image data temporarily stored in a line memory for each row (line) of image data. It is necessary to increase the use efficiency of the line memory in order to increase the noise reduction effect. In general, the image is processed for a predetermined number of rows (divided into rectangles (blocks)), thereby reducing the amount of memory in the line memory storing the images for each row.

In order to increase the effect of reducing the noise of the image data, it is necessary to average the pixel values for a larger number of pixels. For this purpose, it is necessary to increase the number of lines of image data that the line memory can hold at one time. Increasing the number of lines increases the capacity of the line memory, resulting in an increase in circuit area and an increase in cost.

As a method of reducing the capacity of the line memory, Japanese Patent Laid-Open No. 2000-69418 discloses that the resize process is divided horizontally and vertically so that the hardware executes the resize process in the horizontal direction while the software performs the resize in the vertical direction. Japanese Patent Laying-Open No. 2002-44459 discloses a method of compressing image data for a plurality of lines.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a noise reduction device, a noise reduction method, and a recording medium on which a noise reduction method is recorded enabling a noise reduction effect on image data to be enhanced without changing the capacity of a memory .

According to an aspect of the present invention, there is provided a noise reduction apparatus comprising: a storage unit for storing data; a storage unit for storing, for each line of image data input from outside, An encoding unit for converting a code of a line of the line into a code of the same number of bits and storing the code obtained by the conversion in the storage unit for a plurality of lines; And a noise reduction processor for calculating an average of pixel values in a predetermined area with respect to the image data obtained by the decoding to reduce the noise.

Wherein the encoding unit stores the pixel value of the externally input image data in the storage unit by the number of lines obtained by dividing the number obtained by multiplying the maximum number of lines storable by the storage unit by the number of bits of the pixel value, And the decoding section reads out the code from the storage section by the number of lines stored in the storage section.

Wherein the apparatus is configured to reduce noise by using the image data as it is, based on a selection signal for selecting whether or not noise reduction is to be increased, which is input from the outside, and using the image data obtained by decoding by the decoding unit, The switching control unit controlling to switch whether or not to reduce the power consumption of the vehicle; A first switching unit for storing in the storage unit any one of the image data and the code converted by the coding unit based on the control by the switching control unit; And a control unit that reads out the image data for a predetermined number of lines smaller than the plurality of lines from the storage unit based on the control by the switching control unit and outputs the read image data to the noise reduction processing unit, And a second switch for switching between outputting the image data obtained by the decoding by the noise reduction processing unit to the noise reduction processing unit by the number of lines, wherein the noise reduction processing unit, based on the control by the switching control unit, And the noise of the data is reduced based on the number of lines of the data output from the second switching unit.

And a judging unit for judging whether or not the noise included in the image data exceeds a predetermined threshold value based on photographing information inputted from the outside, The control unit controls the first switching unit and the second switching unit.

The apparatus includes a noise detector for detecting a noise amount of the image data, and the determination unit determines whether or not the noise included in the input image data exceeds a predetermined threshold value based on the shooting information and the noise amount Or not.

Wherein the noise reduction processing unit includes: a small averaging unit for reducing noise of the output data when the predetermined number of lines of data are outputted from the second switching unit; A large averaging unit for reducing noise of the output data when the number of lines of data is output from the second switching unit; And a third switching unit for switching whether to output noise-reduced data by the large-averaging unit or to output noise-reduced data by the small-averaging unit based on the control by the switching control unit do.

According to an aspect of the present invention, there is provided a noise reduction method performed by a noise reduction apparatus having a storage unit for storing data, which is an embodiment of the present invention, Converting the code into a code which is smaller than the number of bits and is represented by the same number of bits in all the pixels of the line and stores the code obtained by the conversion in the storage unit for a plurality of lines; A decoding step of reading the code from the storage unit by the number of lines and decoding the code; And a noise reduction processing step of calculating an average of pixel values within a predetermined area with respect to the image data obtained by the decoding to reduce noise.

According to an aspect of the present invention, there is provided a recording medium for recording a noise reduction method, comprising: a computer, which is a noise reduction apparatus having a storage unit for storing data, Converting each pixel value of the pixel value into a code which is smaller than the bit number of the pixel value and which is represented by the same number of bits in all the pixels of the line and stores the code obtained by the conversion for the number of lines in the storage section; A decoding step of reading the code from the storage unit by the number of lines and decoding the code; And a noise reduction processing step of calculating an average of pixel values within a predetermined area with respect to the image data obtained by the decoding to reduce noise.

The noise reduction apparatus of the present invention can improve the noise reduction effect by increasing the number of lines of image data input to the noise reduction processing unit and increasing the number of pixels to be subjected to the noise reduction processing by increasing the compression ratio at the time of encoding by the encoding unit have.

Further, the noise reduction apparatus of the present invention includes a coding unit that performs coding without changing the storage capacity of the storage unit and a decoding unit that performs decoding, thereby improving the noise reduction effect by the noise reduction processing unit without changing the size of the line memory .

1 is a block diagram of a noise reduction apparatus according to a first embodiment of the present invention.
2 is a flowchart showing the processing flow of the noise reduction apparatus of FIG.
3 is a block diagram of a noise reduction apparatus according to a second embodiment of the present invention.
Fig. 4 is a block diagram showing the outline of processing of the noise reduction apparatus in the normal operation mode of Fig. 3; Fig.
5 is a flowchart showing a process flow of the noise reduction apparatus of FIG.
Fig. 6 is a modification of the noise reduction processing unit of Fig.
7 is a block diagram of a noise reduction apparatus according to the third embodiment of the present invention.
8 is a flowchart showing the processing flow of the noise reduction apparatus of FIG.

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

≪ Embodiment 1 >

The noise reduction apparatus 10 of the first embodiment increases the number of lines capable of compressing the data amount and storing it in the line memory by encoding the data of each pixel without changing the capacity of the line memory. As a result, the noise reduction processing section 14 can increase the number of pixels subjected to the noise reduction processing at once for the image data of the increased number of lines, thereby enhancing the noise reduction effect.

1 is a block diagram of a noise reduction apparatus 10 according to a first embodiment of the present invention.

The noise reduction apparatus 10 includes a coding unit 11, a line memory (storage unit) 12, a decoding unit 13 ', and a noise reduction processing unit 14. Here, the decoding unit 13 'includes ten decoders 13_i (i is an integer from 1 to 10) composed of decoders 13_1, 13_2, ..., 13_10.

The encoding unit 11 performs DPCM (Differential Pulse Code Modulation) coding on 12-bit image data input from the outside of the noise reduction apparatus 10 for each row (line) of image data, Into a 6-bit code.

The encoding unit 11 stores the code obtained by the conversion on a line by line basis in the line memory 12. [

For example, in the case of image data whose resolution is SVGA (Super Video Graphics Array), since one line becomes 800 pixels, the encoding unit 11 outputs 2400 (6 [bit] x 800) bits of image data for one line And stores the sign in the line memory 12.

At this time, the encoding unit 11 sequentially stores 6-bit encoded code data per line in the line memory 12 in order. When the encoding unit 11 stores the image data of the 10th line in the line memory 12, the image data of the 11th line is stored next in the memory space of the line memory 12 storing the image data of the 1st line. By repeating the above processing, the line memory 12 stores the code of 10 lines obtained by encoding, which is closest to the line memory 12.

The encoding method of the encoding unit 11 is not limited to DPCM but may be an equal length encoding in which the number of bits after encoding is always the same in all pixels included in one line.

For example, when the image data input to the encoding unit 11 is 12 bits, pixel values from the head line are 100, 120, 115, 118, ... As shown in FIG.

The encoding unit 11 encodes the pixel value of the line head pixel value (for example, 100) as it is. The encoding unit 11 calculates a value (difference value) obtained by subtracting the immediately preceding pixel value from the target pixel value with respect to the second and subsequent pixels. In the above example, the encoding unit 11 calculates a value (differential value) 20, -5, 3,... .

When the calculated difference value is converted into a predetermined number of bits (for example, 6 bits), the absolute value of the differential value is converted from a predetermined number of bits to a minus sign (For example, 32 (= 2 ^ 5) steps) equivalent to the number of bits (for example, 5 bits) obtained by subtracting one bit for representing the data. The encoding unit 11 performs conversion so that the smaller the absolute value (difference) of the calculated difference value, the smaller the range of the differential value that becomes the same after the conversion.

In other words, for each line of externally input image data, the encoding unit 11 converts each pixel value of the line into a code that is smaller than the number of bits of the pixel value and is displayed in the same number of bits in all pixels of the line , And the code obtained by the conversion is stored in the line memory 12 for a plurality of lines.

In the encoding unit 11, as the difference between adjacent pixel values becomes smaller, the larger the difference between adjacent pixel values becomes, the more cramped encoding is performed. Therefore, the degradation of the image data after conversion in the image region where the pixel value changes smoothly, such as the gradation included in the image data, .

The line memory 12 holds the 6-bit codes stored (stored) by the encoding unit 11 for 10 lines.

The decoding unit 13 'reads out the code from the line memory 12 by the number of lines and decodes the code.

More specifically, each decoding unit 13_i constituting the decoding unit 13 'performs the following processing in parallel. The decoding unit 13_i reads out the code corresponding to the i-th line among the image data stored from the line memory 12. [ The decoding unit 13_i generates 12-bit image data equal to the original data size by decoding the code corresponding to the i-th line of the image data, and outputs the image data to the noise reduction processing unit 14. [

The above processing is summarized. The encoding unit 11 encodes the pixel values of the image data for which compression encoding has not been performed to the maximum number of lines (for example, 5 [lines]) that the line memory 12 can store The number of lines (for example, (5 12) / 6 = 10 lines (for example, 12 [bits]) obtained by dividing the number obtained by multiplying the number of bits of the value ) In the line memory 12 as shown in Fig.

The decryption unit 13'including the decryption unit 13_i for the number of lines (for example, ten lines) stored in the line memory 12 is controlled by the number of lines stored in parallel from the line memory 12 For ten lines) code.

The noise reduction processing unit 14 reduces the noise included in the image data obtained by the decoding by the decoding unit 13_i. Specifically, the noise reduction processing unit 14 calculates an average value of pixel values for a predetermined number of rows (for example, 10 columns) of image data input simultaneously in ten lines. Specifically, the noise reduction processing section 14 calculates the sum of the pixel values of pixels of a predetermined number of columns (for example, 10 columns) of the image data of the input line, and outputs the sum to the number of pixels (for example, 100 ) To calculate the average value of the pixel values.

The noise reduction processing unit 14 changes the pixel value of the input image data to the average value of the calculated pixel values and outputs the image data (for example, 12 bits) having the changed pixel value to the outside of the noise reduction apparatus 10. [

Here, the total number of pixels used for noise reduction is called a tap number. In the example of this embodiment, the number of taps is 100 because of 10 lines x 10 columns.

The noise reduction processing unit 14 can use the pixel values as many as the number of lines of the input image data as the noise reduction processing tap. As the number of input image data lines increases, the number of taps can be increased and the noise reduction effect can be increased.

2 is a flowchart showing a process flow of the noise reduction apparatus 10 of FIG.

First, the encoding unit 11 performs DPCM encoding for each line on the input image data (S101).

Next, the encoding unit 11 stores the codes obtained by the DPCM encoding in the line memory 12 for each line (S102).

Next, each decoding unit 13_i reads the code obtained by the DPCM coding from the line memory 12 in parallel with the decoding unit for each line of the image data, and performs DPCM decoding on the code obtained by the read DPCM coding (S103).

The noise reduction processing unit 14 calculates an average value using the image data for 10 lines obtained by the DPCM decoding and changes the pixel value of the image data obtained by the decoding to the average value (S104). Thus, the processing of this flowchart is terminated.

≪ Effects of First Embodiment >

The noise reduction apparatus 10 of the present embodiment increases the number of lines of image data input to the noise reduction processing unit 14 by compressing the image data in the encoding unit 11 and increases the number of taps of the noise reduction processing, Can be improved.

Further, by using the technique shown in this embodiment, it is possible to increase the number of lines of image data input to the noise reduction processing section 14 by increasing the compression ratio at the time of coding by the encoding section 11, so that the noise reduction processing section 14 The capacity of the memory can be reduced with the same number of taps, that is, with the same noise reduction effect. For example, when 12 bits are encoded into 6 bits, the noise reduction processing unit 14 can reduce the capacity of the memory by 50%.

The noise reduction apparatus 10 of the present embodiment is provided with a coding unit 11 for performing isochronous coding without changing the storage capacity of the line memory 12 and a decoding unit 13_i for performing decoding, It is possible to improve the noise reduction effect by the noise reduction processing unit 14 without changing the size of the noise reduction processing unit 12.

In addition, since the noise reduction apparatus 10 of the present embodiment can greatly reduce the line memory scale compared to the apparatus that simply constitutes the memory capacity of the line memory 12 twice, the circuit area can be reduced.

In the noise reduction apparatus 10 of the present embodiment, the DPCM coding performed by the coding unit 11 hardly deteriorates the image quality, and the noise reduction effect can be increased, so that the quality of the final image quality can be improved.

The noise reduction apparatus 10 of the present embodiment can improve the efficiency of use of the line memory 12 by using the compression of image data by the encoding unit 11 to improve the efficiency of use of the encoding unit 11 and the decoding unit 13_i And the algorithm itself of the noise reduction processing need not be changed.

≪ Second Embodiment >

In the noise reduction apparatus 10 of the first embodiment, it is difficult to prove that image quality deterioration due to DPCM compression is not visually conspicuous under all image conditions, so that the compression ratio can not be set too high.

The noise reduction apparatus 20 in the second embodiment employs a configuration for switching between a normal operation mode for performing noise reduction processing on input image data and a DPCM operation mode for performing noise reduction processing on compressed image data.

Here, deterioration of picture quality due to DPCM coding is mainly caused by a phenomenon in which a step is visible in a gradation area of an image. The phenomenon tends to be visually conspicuous when the noise contained in the input image data is small. Therefore, the noise reduction apparatus 20 does not perform DPCM coding by switching from the DPCM operation mode to the normal operation mode when noise is small in the input image.

In addition, in the normal operation mode, the number of lines of image data input to the noise reduction processing section 14b is reduced compared to the DPCM operation mode, and the number of taps is reduced, thereby reducing the noise reduction processing effect. However, if there is less noise in the image data, the necessity of performing the noise reduction processing is low. Therefore, even if the noise reduction effect by the noise reduction processing section 14b is lowered, the image quality of the image data output from the noise reduction apparatus 20 is hardly affected.

Thus, when the noise reduction apparatus 20 can visually recognize deterioration of image quality due to DPCM coding in the DPCM operation mode, the noise reduction apparatus 20 can switch to the normal operation mode. Therefore, in the DPCM operation mode, The compression rate can be set high.

3 is a block diagram of the noise reduction apparatus 20 according to the second embodiment.

The noise reduction apparatus 20 includes a coding section 11b, a line memory 12b, a decoding section 13b, a noise reduction processing section 14b, a switching control section 21, a first switching section 22, (23). The noise reduction processing unit 14b includes a small averaging unit 24, a large averaging unit 25, and a third switching unit 26.

The encoding unit 11b performs DPCM encoding on the input 12-bit image data in the same manner as the encoding unit 11 in Fig. 1, and outputs a 6-bit code obtained by DPCM encoding to the second input of the first switch 22 Port.

The switching control section 21 switches between the first switching section 22 and the second switching section 22 in accordance with the selection signal s inputted from the outside of the noise reduction apparatus 20 to select whether it is the normal operation mode or the DPCM operation mode 23) and a switching signal (k) for switching the normal operation mode and the DPCM operation mode to the third switching unit (26).

Thereby, the switching control section 21 controls to switch the output data among the data input to the first switching section 22, the second switching section 23, and the third switching section 26 to switch them.

The first switching unit 22, the second switching unit 23 and the third switching unit 26 output or switch any one of two pieces of data inputted in accordance with the switching signal k. The first switching unit 22, the second switching unit 23, and the third switching unit 26 have first and second ports for receiving input data, respectively.

When the switching signal k for switching to the normal operation mode is input, the first switching unit 22, the second switching unit 23, and the second switching unit 23 are turned on until the switching signal k for switching to the DPCM operation mode is input. The third switching unit 26 outputs data input through the first port.

On the other hand, when the switching signal k for switching to the DPCM operation mode is input, the first switching unit 22 and the second switching unit 23 are operated until the switching signal k for switching to the normal operation mode is input. , And the third switching unit 26 outputs the data input through the second port.

The first switching unit 22 stores, in the line memory 12b, image data input from the outside through the first port in the normal operation mode. In the DPCM operation mode, the first switching unit 22 stores the code inputted from the coding unit 11b through the second port in the line memory 12b for each line.

The line memory 12b holds (stores) the image data input from the first switching unit 22 on a line-by-line basis, similar to the line memory 12 of Fig. Specifically, in the normal operation mode, the line memory 12b holds five lines of 12-bit data per pixel. In the DPCM operation mode, the line memory 12b holds 10 lines of 6-bit data per pixel.

The decoding unit 13b reads codes from the line memory 12b in the same manner as the decoding unit 13 in Fig. 1, decodes the codes, generates 12-bit image data, To the second switching unit (23).

When the switching signal (k) for switching to the normal operation mode is input, the second switching unit 23 reads the image data in five lines in parallel through the first port in the line memory 12b, And outputs it to the small averaging unit 24 in five lines in parallel. When the switching signal (k) for switching to the DPCM operation mode is input, the second switching unit 23 switches the image data obtained by the decoding inputted through the second port in 10 lines in parallel in the decoding unit 13b to 10 lines And outputs it to the large averaging unit 25 in parallel.

The small averaging unit 24 calculates the average value of pixel values for a predetermined number of columns (for example, five columns) of the image data for the five lines input by the second switching unit 23. Specifically, the small averaging unit 24 calculates the sum of the pixel values of pixels of a predetermined number of columns (for example, five columns) of the input five-line image data, and outputs the sum to the number of pixels (for example, 25) to calculate the average value of the pixel values.

The small averaging unit 24 changes the pixel value of the input image data to an average value of the calculated pixel values and outputs image data (for example, 12-bit image data) having the changed pixel value to the third switching unit 26 do.

The large averaging unit 25 calculates an average value of pixel values for a predetermined number of rows (for example, ten columns) of the image data for ten lines input from the second switching unit 23. Specifically, the large averaging unit 25 calculates the sum of the pixel values of pixels of a predetermined number of columns (for example, ten columns) of the inputted 10 lines of image data, and outputs the sum to the number of pixels (for example, 100), thereby calculating the average value of the pixel values.

The large averaging unit 25 changes the pixel value of the input image data to the average value of the calculated pixel values and outputs image data (for example, 12-bit image data) having the changed pixel value to the third switching unit 26 do.

When the switching signal k for switching to the normal operation mode is input, the third switching unit 26 switches the image data input from the small averaging unit 24 via the first port to the outside of the noise reduction apparatus 20 . When the switching signal k for switching to the DPCM operation mode is inputted, the third switching unit 26 outputs the image data (for example, 12-bit image) input via the second port in the large averaging unit 25 Data to the outside of the noise reduction apparatus 20. [

Thereby, the switching control section 21 can switch the normal operation mode and the DPCM operation mode by outputting the switching signal k to the first switching section 22, the second switching section 23, and the third switching section 26 have.

4 is a block diagram showing the outline of processing of the noise reduction apparatus 20 in the normal operation mode.

As shown in Fig. 4, in the normal operation mode, the image data is stored in the line memory 12b on a line-by-line basis. The noise reduction processing section 14b reads the image data for five lines stored in the line memory 12b. The small averaging unit 24 of the noise reduction processing unit 14b calculates the average value of the pixel values every 5 lines by 5 columns and changes the pixel value of the input image data to the average value. Then, the small averaging unit 24 outputs the image data having the changed pixel value to the outside of the noise reduction apparatus 20.

In the DPCM operation mode, as in the first embodiment, the noise reduction processing section 14b reduces the noise contained in the image data input in the processing flow shown in Fig.

5 is a flowchart showing a process flow of the noise reduction apparatus 20 according to the second embodiment.

5 (a) is a flowchart showing the processing flow of the noise reduction apparatus 20 in the normal operation mode. First, the line memory 12b of the noise reduction apparatus 20 holds (stores) image data inputted from the outside on a line-by-line basis (S201). Next, the noise reduction processing section 14b reduces the noise of the image data by averaging the pixel values with taps of 5 lines x 5 columns (S202). Thus, the processing of this flowchart is terminated.

5 (b) is a flowchart showing a process flow of the noise reduction apparatus 20 in the DPCM operation mode. The processing from step S301 to step S304 is the same as the processing from step S101 to step S104 in Fig.

First, the coding unit 11b performs DPCM coding on the input image data (S301).

Next, the encoding unit 11b stores the encoded image data for each line in the line memory 12b (S302).

Next, the decoding unit 13b reads the image data in 10 lines in parallel in the line memory 12b, and decodes the image data for each read line in parallel (S303).

Next, the large averaging unit 25 of the noise reduction processing unit 14b performs noise reduction processing on the image data obtained by the decoding inputted by the second switching unit 23, with the taps of 10 lines x 10 columns (S304). Thus, the processing of this flowchart is terminated.

The noise reduction apparatus 20 of this embodiment can change the number of lines stored in the line memory 12b of the same size by switching the normal operation mode and the DPCM operation mode.

As a result, when the image deterioration due to the DPCM compression coding is visually conspicuous, it is possible to change to the normal operation mode in which the DPCM compression is not performed, so that the compression ratio of the coding unit 11b can be set high. As a result, since the number of taps of the noise reduction processing section 14b can be increased, the noise reduction effect by the noise reduction processing section 14b can be improved.

That is, the noise reduction apparatus 20 can improve the noise reduction effect by employing a configuration in which the compression ratio of the encoding unit 11b is increased and the number of lines subjected to the noise reduction processing at a time by the noise reduction processing unit 14b is increased.

<Modification of Noise Reduction Processing Unit>

Fig. 6 is a modification of the noise reduction processing section 14b of the second embodiment. The noise reduction processing section 14c includes a small averaging section 24c and a large averaging section 25c.

In the normal operation mode, the small-averaging section 24c, based on the control of the switching control section 21, controls the second switching section 23 to switch the image data And outputs the image data having the pixel values averaged by the 5 x 5 taps to the outside of the noise reduction apparatus 20 like the above-described small averaging unit 24 of Fig.

In the DPCM operation mode, the large averaging unit 25c receives the image data obtained by the decoding through the 10-line wiring from the second switching unit 23 based on the control of the switching control unit 21, The image data having a pixel value averaged by a tap of 10 x 10 is output to the outside of the noise reduction apparatus 20. [

Thus, the noise reduction apparatus 20 can reduce the circuit area of the noise reduction processing unit 14c by adopting a configuration in which the small averaging unit 24c is included in the large averaging unit 25c. As a result, the entire circuit area of the noise reduction apparatus 20 can be reduced.

&Lt; Third Embodiment >

The noise reduction apparatus 20 of the second embodiment has described that the normal operation mode and the DPCM operation mode are switched according to the selection signal s input from the outside to the switching control unit 21. [ In the third embodiment, the noise reduction apparatus 30 switches between the normal operation mode and the DPCM operation mode based on at least one of the amount of noise included in the image input from the outside or the image pickup information input from the external image pickup apparatus Explain.

7 is a block diagram of the noise reduction apparatus 30 according to the third embodiment. 3 are denoted by the same reference numerals and the description thereof is omitted.

The configuration of the noise reduction apparatus 30 of Fig. 7 differs from the configuration of the noise reduction apparatus 20 of Fig. 3 in that the noise detection unit 31 and the determination unit 32 are added and the switching control unit 21 controls the switching control unit The first switching unit 22 is connected to the first switching unit 22d, the second switching unit 23 is connected to the second switching unit 23d, the third switching unit 26 is connected to the third switching unit 23d, (26d).

The noise detecting unit 31 detects the amount of noise included in the image data input from the outside of the noise reducing apparatus 30 and outputs the information indicating the detected amount of noise to the determining unit 32. [ More specifically, for example, the noise detecting unit 31 performs an Hadamard Transform (Hadamard Transform) on the input image data to perform an operation of separating the intensity of each frequency included in the image, And the noise amount is calculated.

The judging unit 32 receives photographing information (for example, information indicating the ISO sensitivity) from a photographing apparatus (not shown) outside the noise reducing apparatus 30. [

The judging unit 32 judges whether or not a large amount of noise is contained in the image data inputted based on at least any one of the information indicating the amount of noise or the photographing information per predetermined image frame.

When the information indicating the amount of noise exceeds the predetermined amount of noise, the determining unit 32 determines that the input image data contains a large amount of noise. On the other hand, when the information indicating the amount of noise is equal to or less than the predetermined amount of noise, the determining unit 32 determines that the amount of noise is not included in the input image data.

If the information indicating the ISO sensitivity exceeds the predetermined ISO sensitivity (for example, exceeds ISO 400), only the information indicating the ISO sensitivity is input, the determining section 32 includes a large amount of noise in the input image data . This generally utilizes the fact that increasing the ISO sensitivity increases the amount of noise included in the image data. On the other hand, when the information indicating the ISO sensitivity is not more than the predetermined ISO sensitivity, the determining section 32 determines that the input image data does not contain much noise.

When both the information indicating the amount of noise and the information indicating the ISO sensitivity are inputted and the information indicating the amount of noise or the information indicating the ISO sensitivity exceeds the predetermined threshold value set in each of them, It is determined that the input image data contains a large amount of noise. On the other hand, when both the information indicating the amount of noise and the information indicating the ISO sensitivity are equal to or less than a predetermined threshold value, the determining section 32 determines that the input image data does not contain much noise.

Then, the determination section 32 outputs information r indicating the determination result after the determination to the switching control section 21d.

When the input control unit 21d receives the information r indicating that the input image data contains a large amount of noise, the first switching unit 22d, the second switching unit 23d, the third switching unit 23d, And outputs a switching signal (k) for switching to the normal operation mode.

When the input control unit 21d receives the information r indicating the determination result that the input image data contains a large amount of noise, the switching control unit 21d controls the first switching unit 22d, the second switching unit 23d, (K) for switching to the DPCM operation mode.

The first switching unit 22d outputs image data input from the outside through the first port based on the switching signal k input from the switching control unit 21d as in the first switching unit 22 in Fig. And stores the code in the memory 12b or the code input from the encoding unit 11b through the second port in the line memory 12b.

Specifically, when the switching signal k input from the switching control section 21d is the switching signal k for switching to the normal operation mode, the first switching section 22d switches the image data input from the outside through the first port To the line memory 12b.

On the other hand, when the switching signal k inputted from the switching control section 21d is the switching signal k switching to the DPCM operation mode, the first switching section 22d outputs the code inputted through the second port in the coding section 11b To the line memory 12b.

The second switching unit 23d switches the data to be output to the noise reduction processing unit 14d based on the switching signal k input from the switching control unit 21d in the same manner as the second switching unit 23 in Fig. 3 .

Specifically, when the switching signal k input from the switching control section 21d is the switching signal k for switching to the normal operation mode, the second switching section 23d switches the image data from the line memory 12b to five lines Through the first port in parallel, and outputs the read image data to the small averaging unit 24 of the noise reduction processing unit 14d in five lines in parallel.

On the other hand, when the switching signal k input from the switching control section 21d is the switching signal k for switching to the DPCM operation mode, the third switching section 26d outputs the image data from the decoding section 13b to the 10- Through the second port, and outputs the read image data to the large averaging unit 25 of the noise reduction processing unit 14d in 10 lines in parallel.

8 is a flowchart showing a process flow of the noise reduction apparatus 30 according to the third embodiment.

First, the noise detection unit 31 detects the amount of noise contained in the input image data (S401).

Next, the determining section 32 determines whether any of the information indicating the amount of noise or the information indicating the ISO sensitivity per predetermined image frame exceeds a predetermined threshold set in each of the image frames (S402).

When the noise amount does not exceed the predetermined noise amount and the ISO sensitivity does not exceed the predetermined ISO sensitivity, the determination section 32 determines that the input image data does not contain much noise amount. The switching control section 21d controls the first switching section 22d, the second switching section 23d and the third switching section 23d so as to operate in the normal operation mode and outputs the switching signal k (S403).

The first switching unit 22d stores the externally input image data in the line memory 12b for each line (S404).

Next, the second switching section 23d reads the image data in five lines in parallel in the line memory 12b, and reads the read image data in five lines in parallel to the small averaging section 24 of the noise reduction processing section 14d (S405).

The small averaging unit 24 of the noise reduction processing unit 14d performs noise reduction processing on the image data input from the second switching unit 23d with a 5x5 tap (S406).

On the other hand, when the noise amount exceeds the predetermined noise amount or the ISO sensitivity exceeds the predetermined ISO sensitivity, the determination section 32 determines that the input image data contains a large amount of noise. The switching control section 21d switches the first switching section 22d, the second switching section 23d and the third switching section 26d so as to operate in the DPCM operation mode, (S407).

Since the processing from step S408 to step S411 is the same as the processing from step S301 to step S304 shown in Fig. 5B, the description of the processing will be omitted. Thus, the processing of this flowchart is terminated.

As described above, the noise reduction apparatus 30 of the present embodiment determines whether there is a large amount of noise included in the input image based on at least any one of the amount of noise included in the image input from the outside or the image input from outside . The noise reduction apparatus 30 adopts a configuration in which the line memory 12b and the noise reduction processing unit 14d are combined by switching between the normal operation mode and the DPCM operation mode based on the determination result, The number of lines to be stored can be changed.

Thereby, the noise reduction apparatus 30 can change to the normal operation mode in which the DPCM compression coding is not performed when the image deterioration due to the DPCM compression coding becomes visually conspicuous, so that the compression ratio of the coding unit 11b can be set high. As a result, since the number of taps of the noise reduction processing section 14d can be increased, the noise reduction effect by the noise reduction processing section 14d can be improved.

In addition, the noise reduction apparatus 30 of the present embodiment may change the noise reduction processing unit 14d to the noise reduction processing unit 14c shown in FIG.

The determination unit 32 in the present embodiment is an information input from the outside, and the noise included in the image input based on at least one of the information indicating the amount of noise included in the image data or the image- It may be judged whether or not there is a large amount.

In each of the embodiments, the encoding units 11 and 11b encode 6-bit image data from 12-bit image data. However, the present invention is not limited to this, and the number of bits of the image data of the first number of bits It may be encoded as image data. For example, the encoding units 11 and 11b may be encoded as 5-bit image data from 10-bit image data or 6-bit image data from 14-bit image data.

In each of the embodiments, the encoding units 11 and 11b may change the number of bits after encoding according to the area of the image. For example, the encoding units 11 and 11b may encode 12-bit image data into 6-bit image data in any line, and may encode 12-bit image data into 8-bit image data in other lines.

In each of the embodiments, the encoding units 11 and 11b use DPCM as a compression method, but the present invention is not limited thereto, and any encoding method may be used as long as it is an isochronous encoding. Further, the compression method by the encoding units 11 and 11b is preferable as long as it can implement a small circuit scale. Further, the compression method by the encoding units 11 and 11b is better if the image quality deterioration is small.

In the method of dividing a processing unit for each column integrated by a predetermined number of columns such as a general rectangular (block) processing, noise reduction processing can not be performed over the integrated row. Since the noise reduction apparatuses 10, 20, and 30 in the embodiment of the present invention are not divided into such processing units, the noise reduction processing units 14, 14b, and 14d can freely set the widths of noise reduction rows.

In the rectangular (block) processing, the processing clock rises due to overlapping (redundancy) between the integrated rows by the predetermined number of columns, which causes a problem of increase in power consumption and heat generation. There is also a process in which a rectangular (block) process can not be performed by the noise reduction algorithms of the noise reduction processing units 14, 14b, and 14d. In addition, rectangle (block) processing also complicates system control, which takes time to design and verify. The noise reduction apparatuses 10, 20 and 30 of the respective embodiments have an advantageous effect that such a problem does not occur.

The noise reduction apparatuses 10, 20, and 30 of the respective embodiments can be all mounted on hardware, and the processing by software does not increase, so that the burden on a CPU (Central Processing Unit) not shown increases.

The functions of the noise reduction apparatuses 10, 20, 30 of the respective embodiments of the present invention or a part of the functions thereof may be realized by a computer. In this case, a computer program for realizing the function may be recorded on a computer-readable recording medium, and the computer program recorded on the recording medium may be loaded into a computer system and executed. The term &quot; computer system &quot; as used herein includes OS (Operating System) and peripheral hardware. The "computer-readable recording medium" refers to a storage device such as a flexible disk, a magneto-optical disk, an optical disk, a portable recording medium such as a memory card, or a hard disk built in a computer system. The term &quot; computer-readable recording medium &quot; refers to a medium in which a program is dynamically maintained for a short period of time, such as a communication line when a program is transmitted through a communication line such as a network such as the Internet or a telephone line, And may include maintaining a program for a certain period of time such as a volatile memory in the computer system. The computer program may be for realizing a part of the functions described above, or may be realized by a combination with the computer program already recorded in the computer system.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, the invention is not limited to the disclosed exemplary embodiments, but includes various designs and the like that do not depart from the gist of the invention.

10, 20, 30; Noise reduction device
11, 11b; The encoding unit
12, 12b; Line memory (storage unit)
13_1, 13_2, ... , 13_10, 13b; The decoding unit
13 '; The decoding unit
14, 14b, 14c, 14d; The noise-
21, 21d; [0050]
22, 22d; The first switching unit
23, 23d; The second switching portion
24; Small averager
25; Large averaging unit
26, 26d; The third switching unit
31; The noise detector
32; [0031]

Claims (8)

A storage unit for storing data;
For each line of externally input image data, each pixel value of the line is converted into a code which is smaller than the number of bits of the pixel value and which is represented by the same number of bits in all pixels of the line, An encoding unit for storing the number of lines as a plurality of lines;
A decoding unit that reads out the code from the storage unit by the number of lines and decodes the code; And
And a noise reduction processing unit for calculating an average of pixel values within a predetermined area with respect to the image data obtained by the decoding to reduce noise. The method according to claim 1,
The encoding unit stores the pixel value of the externally input image data in the storage unit by the number of lines obtained by dividing the maximum number of lines storable by the storage unit by the number of bits of the pixel value, And,
Wherein the decoding unit reads out the code from the storage unit by the number of lines stored in the storage unit. The method according to claim 1,
Whether to reduce noise by using the image data as it is, based on a selection signal for selecting whether or not noise reduction is to be increased, which is input from the outside, or to reduce noise by using image data obtained by decoding by the decoding unit A switching control unit for controlling the switching control unit;
A first switching unit for storing in the storage unit any one of the image data and the code converted by the coding unit based on the control by the switching control unit; And
The control unit reads out the image data for a predetermined number of lines smaller than the plurality of lines from the storage unit based on the control by the switching control unit and outputs the read image data to the noise reduction processing unit, And a second switching unit for switching between outputting the image data obtained by the decoding by the number of lines to the noise reduction processing unit,
Wherein the noise reduction processing section reduces the noise of the data based on the number of lines of the data outputted from the second switching section based on the control by the switching control section. Claim 4 has been abandoned due to the setting registration fee. The method of claim 3,
And a determination unit that determines whether noise included in the image data exceeds a predetermined threshold value based on shooting information input from the outside,
Wherein the switching control section controls the first switching section and the second switching section on the basis of the determination result by the determination section. Claim 5 has been abandoned due to the setting registration fee. 5. The method of claim 4,
And a noise detector for detecting a noise amount of the image data,
Wherein the determination unit determines whether noise included in the input image data exceeds a predetermined threshold value based on the shooting information and the noise amount. Claim 6 has been abandoned due to the setting registration fee. 6. The image processing apparatus according to any one of claims 3 to 5,
A small averaging unit for reducing noise of the output data when the predetermined number of lines of data are output from the second switching unit;
A large averaging unit for reducing noise of the output data when the number of lines of data is output from the second switching unit;
And a third switching unit for switching whether to output noise-reduced data by the large-averaging unit or to output noise-reduced data by the small-averaging unit based on the control by the switching control unit Noise reduction apparatus. A noise reduction method executed by a noise reduction apparatus having a storage unit in which data is stored,
For each line of externally input image data, each pixel value of the line is converted into a code which is smaller than the bit number of the pixel value and which is represented by the same number of bits in all pixels of the line, An encoding step of storing a plurality of lines as a plurality of lines;
A decoding step of reading the code from the storage unit by the number of lines and decoding the code; And
And a noise reduction processing step of calculating an average of pixel values in a predetermined area with respect to the image data obtained by the decoding to reduce noise. Claim 8 has been abandoned due to the setting registration fee. A computer, which is a noise reduction apparatus having a storage unit for storing data,
Each pixel value of the line is converted into a code which is smaller than the number of bits of the pixel value and is displayed in the same number of bits in all pixels of the line for each line of externally input image data, An encoding step of storing the number of lines in the storage unit;
A decoding step of reading the code from the storage unit by the number of lines and decoding the code; And
And a noise reduction processing step of calculating an average of pixel values in a predetermined area with respect to the image data obtained by the decoding so as to reduce the noise. The computer readable recording medium according to claim 1, Recording medium.

Images