KR20100116302A - Apparatus and method for 3d digital noise reduction of camera - Google Patents

Abstract

PURPOSE: A device and a method for eliminating a 3D digital noise of a camera is provided to variably set a tap number by M which is smaller than N, thereby efficiently eliminating a ghost noise due to a 3D digital noise eliminating operation. CONSTITUTION: An illumination value is detected in a mode for eliminating a 3D digital noise. N taps are set to eliminate a 3D digital noise in case of very low illumination. M taps are set in case of low illumination to eliminate a 3D digital noise. M is smaller than N. N taps corresponding to very low illumination and M taps corresponding to low illumination are stored and managed in a lookup table in advance as control information.

Description

Apparatus and method for 3D digital noise reduction of camera

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus and method for removing 3D digital noise of a camera. For example, 3D digital noise removing (DNR) for removing ghost noise generated by various types of cameras for photographing a video is provided. Digital Noise Reduction) apparatus and method.

In general, an image pickup device such as a Charge Coupled Device (CCD) or a Complementary Metal-Oxide Semiconductor (CMOS) is widely used in a camera for shooting a video.

On the other hand, the camera equipped with the image pickup device is provided with a two-dimensional digital noise removal (2D-DNR) device or a three-dimensional digital noise removal (3D-DNR) device, for example, as shown in FIG. In the 2D digital noise canceller, N frames (or fields) sequentially output from the CD are temporarily stored in the buffer 10.

In addition, the average value of luminance level values CCD_Out_Y_Value # n-5 to #n of each of the temporarily stored frames is calculated to remove ghost noise generated in extremely low light or low light. The number of frames temporarily stored in (10), that is, the number of N (for example, six) is commonly referred to as the number of taps (Tep).

On the other hand, as shown in Fig. 2, in the three-dimensional digital noise removing device, for example, each pixel of each N (for example, six) frames (or fields) output sequentially from the CDC (Pixel) ) one, the k for which the temporarily stored in the value of the luminance, k buffers (30 ₁ ~ 30 _k) is is the ssissi D corresponds to the number of pixels.

After calculating the average value of the luminance level values of the pixels of the N frames temporarily stored, the luminance level values of the pixels of each frame are calculated as described above, and thus, extremely low or low illumination is added. This eliminates afterimage noise that occurs in the.

However, in the general three-dimensional digital noise canceling device, a predetermined number of N-taps is fixedly applied. In the case of photographing any one particular subject at an extremely low level, for example, six frames are removed. If the luminance level value of one pixel is 20, 20, 20, 30, 30, 30, the luminance level average value for the first pixel of very low illumination is (20 + 20 + 20 + 30 + 30 + 30) / 6 = 25.

However, when photographing in low light, for example, if the luminance level value of the first pixels of six frames is increased by the same ratio as a whole, the first pixel of low light is The average luminance level for is (50 + 50 + 50 + 75 + 75 + 75) /6=62.5.

Therefore, in the extremely low illumination, the luminance level value of the current first pixel is 30, and the corresponding average value is 25, and the difference value before and after applying the 3D-DNR is 30-25 =. On the other hand, in low light, the luminance level value of the current first pixel is 75, and the corresponding average value is 62.5, and the difference value before and after applying the 3D-DNR is 75-. 62.5 = 12.5.

In other words, since the difference value before and after applying the 3D-DNR becomes 5 in the ultra low light, there is less afterimage noise caused by the difference between the luminance levels, but in the low light, before and after applying the 3D-DNR, Since the later difference value is 12.5, there is a problem that afterimage noise generated due to the difference between the luminance levels is relatively large.

In the present invention, for example, in various types of cameras for photographing a video, if the low light is extremely low, the number of taps set for 3D digital noise reduction (3D-DNR) is set to N, To provide a three-dimensional digital noise removing device and method of the camera to variably set the number of steps to M less than N, to effectively remove the ghost noise that occurs relatively large in low light It is for.

Three-dimensional digital noise reduction method of the camera according to the invention, the first step in the three-dimensional digital noise reduction mode, detecting the illumination value; And setting the number of N steps for removing 3D digital noise when the result of the detection is extremely low, and setting the number of M steps less than N when performing low level of illumination, and performing the 3D digital noise removing operation. Characterized in that comprises a,

The number of N steps corresponding to the extremely low illuminance and the number of M steps corresponding to the low illuminance are stored and managed in advance as control information of the lookup table.

In addition, the N pieces are at least three or more values, and the M pieces are two or more values less than N,

In addition, the three-dimensional digital noise removing device of the camera according to the present invention, the detection means for detecting the illumination value; Signal processing means for performing a three-dimensional digital noise canceling operation; And if the detected illuminance value is extremely low illuminance, set N number of steps for 3D digital noise removal, and if the illuminance value is low, set M number of steps less than N to perform 3D digital noise removal operation. Characterized in that it comprises a control means for

The apparatus may further include storage means for storing and managing the number of N steps corresponding to the extremely low illuminance and the number of M steps corresponding to the low illuminance in advance as control information of a look-up table. ,

In addition, the N pieces are at least three or more values, and the M pieces are two or more values less than N,

In addition, the detection means and the signal processing means, characterized in that composed of one hardware chip, or separately configured by separate hardware.

3D digital noise canceling apparatus and method of the camera according to the present invention, for example, by using an image pickup device such as CCD (CCD), CMOS (CMOS), etc., to detect the illuminance value in the camera to shoot a video For example, if the light is extremely low, the number of taps (Tep) set for 3D digital noise reduction (3D-DNR) is set to N. If the light is low, the number of steps is set to M less than N. It is possible to efficiently remove ghost noise generated by the three-dimensional digital noise removing operation.

Hereinafter, with reference to the accompanying drawings, a preferred embodiment of the three-dimensional digital noise removing device and method of the camera according to the present invention will be described in detail.

The three-dimensional digital noise removing device and method according to the present invention is applied to various types of cameras for photographing a moving image of a subject using an image pickup device such as a CCD or a CMOS.

On the other hand, in the camera to which the present invention is applied, for example, as shown in FIG. 3, the zoom lens 10, the focus lens 11, the iris 12, the image pickup device 13, and the automatic gain adjuster 14 are shown. ), An illumination intensity detector 15, a signal processor 16, a synthesizer 17, a motor driver 18, a microcomputer 19, an OSD generator 20, a memory 21, and the like.

In addition, the illuminance detection unit 15 converts the luminance (Y) signal of the image pickup device amplified and output through the automatic gain controller 14 into a digital signal, and then detects an illuminance value corresponding thereto. For example, the illuminance detector 15 is an optical detector (OPD: Optical Detector), which is included in the signal processor 16 and constitutes one hardware chip or is separate from the signal processor. It can be configured individually as hardware.

The signal processing unit 16 includes a three-dimensional digital noise removal (3D-DNR) block. The microcomputer 19 includes the signal according to the illuminance value detected by the illuminance detection unit 15. If the illuminance value detected by the illuminance detection unit 15 is extremely low illuminance, for example, if the illuminance value detected by the illuminance detection unit 15 is extremely low illuminance, the number of taps for removing the 3D digital noise is set to N. However, in low light, the number of steps is variably adjusted to the number of M steps less than N.

In the memory 21, a nonvolatile memory such as an EEPROM or a Flash memory is used. As shown in FIG. 4, the number of N steps corresponding to extremely low illumination and low illumination are shown. The number of M (M < N) taps corresponding to is stored in advance as control information in the form of a lookup table, wherein N is at least three or more values, and M The dog has two or more values less than N.

 Accordingly, the microcomputer 19 retrieves the illuminance value detected by the illuminance detection unit 15 and the number of steps of the control information stored in the memory 21, and retrieves the three-dimensional image of the signal processing unit 16. The digital noise canceling operation is controlled, which will be described in detail below.

FIG. 5 is a flowchart illustrating an embodiment of a method for removing 3D digital noise of a camera according to the present invention. As described above with reference to FIG. 4, N corresponding to extremely low illumination in the memory 21 is illustrated. The microcomputer 19 when the 3D digital noise canceling mode is set in response to a user's request while the number of taps and the number of M taps corresponding to the low illumination are stored as control information in the form of a lookup table (S10). In operation S11, the illuminance detector 15 is controlled to measure and detect the illuminance value (S11).

Then, the detected illuminance value is compared with the control information stored and managed in the memory, and when it corresponds to extremely low illuminance (S12), the microcomputer 19 applies the number of N steps read from the control information. As a result, the signal processor 16 performs the third-dimensional digital noise removing operation (S13).

On the other hand, when the detected illuminance value is compared with the control information stored and managed in the memory, and corresponds to low illuminance (S14), the microcomputer 19 reads M (M <N) pieces read from the control information. By applying the number of steps (S15), the three-dimensional digital noise removing operation of the signal processor 16 is performed (S16).

For example, as described above with reference to FIG. 2, in the case of extremely low illumination, when six tap numbers are applied, the luminance level values of the first pixels for six frames are 20, 20, 20, 30. , 30, 30, the luminance level average value for the first pixel of extremely low illumination is (20 + 20 + 20 + 30 + 30 + 30) / 6 = 25.

The luminance level value of the current first pixel is 30, and the average value corresponding thereto is 25, and the difference value before and after applying the 3D-DNR is 30-25 = 5.

On the other hand, as shown in FIG. 6, if the brightness level value of the first pixels for four frames is 50, 75, 75, 75 in low light by applying four tap numbers less than six, The luminance level average value for the first pixel of low illumination is (50 + 75 + 75 + 75) /4=68.7.

The luminance level value of the current first pixel is 75, and the corresponding average value is 68.7, and the difference value before and after applying the 3D-DNR is 75-68.7 = 6.3.

If the six tap numbers are applied as it is even in low light, since the luminance level values of the first pixels for the six frames are 50, 50, 50, 75, 75, 75, the first low light is shown. Since the luminance level average value for the pixel becomes (50 + 50 + 50 + 75 + 75 + 75) / 6 = 62.5, the luminance level value of the current first pixel is 75, and the corresponding average value is 62.5, the difference between before and after applying 3D-DNR is 75-62.5 = 12.5.

Therefore, if the number of steps is continuously set to 6 in the low city as in the case of the extremely low city, the difference value before and after applying the 3D-DNR becomes 12.5. If the variable setting is less than 6, the difference value before and after the application of the 3D-DNR becomes 6.3, so that the afterimage noise generated due to the difference between the luminance levels can be efficiently removed.

It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the spirit or scope of the invention as defined in the appended claims. , Alteration, substitution, addition, or the like.

1 shows an embodiment of a typical two-dimensional digital noise cancellation (2D-DNR) device,

2 shows an embodiment of a typical three-dimensional digital noise cancellation (3D-DNR) device,

3 illustrates an embodiment of a camera to which the present invention is applied;

FIG. 4 illustrates an embodiment of control information for three-dimensional digital noise cancellation stored and managed according to the present invention.

5 is a flowchart illustrating an embodiment of a method for removing 3D digital noise of a camera according to the present invention;

FIG. 6 illustrates an embodiment in which the number of steps for low illumination three-dimensional digital noise cancellation is adjusted to a small value according to the present invention.

[Description of Drawings]

10: zoom lens 11: focus lens

12: iris 13: image pickup device

14: automatic gain controller 15: illuminance detection unit

16 signal processing unit 17 synthesis unit

18: motor driver 19: microcomputer

20: OSD generator 21: memory

Claims (7)

Detecting the illuminance value in the 3D digital noise canceling mode; And As a result of the detection, if the light is extremely low, the number of N steps for removing 3D digital noise is set, and if the light is low, setting the number of M steps smaller than N is performed, and performing the 3D digital noise removing operation. 3D digital noise reduction method of the camera, characterized in that it comprises a. The method of claim 1, The number of N steps corresponding to the extremely low illuminance and the number of M steps corresponding to the low illuminance are stored and managed in advance as control information of a look-up table. The method of claim 1, Wherein N is at least three or more values, and M is two or more values less than N. 3. Detecting means for detecting an illuminance value; Signal processing means for performing a three-dimensional digital noise canceling operation; If the detected illuminance value is extremely low illuminance, set N number of steps for 3D digital noise removal, and if the illuminance value is low, set M number of steps less than N to perform 3D digital noise removal operation. 3D digital noise removing device of the camera, characterized in that it comprises a control means for. The method of claim 4, wherein And storage means for storing and storing the number of N steps corresponding to the extremely low illuminance and the number of M steps corresponding to the low illuminance as control information of the look-up table in advance. 3D digital noise canceller. The method of claim 4, wherein Wherein N is at least three or more values, and M is two or more values less than N. 3. The method of claim 4, wherein The detecting means and the signal processing means are composed of one hardware chip or separate hardware, characterized in that the three-dimensional digital noise removing method of the camera.

Images