KR101397182B1 - Image processing device and method for improving horizontal noise - Google Patents

Abstract

An image processing apparatus and a horizontal line noise improvement method are disclosed. The image processing apparatus includes: a noise value calculating unit for calculating an average noise value that is an average pixel value of a plurality of target pixels existing in a target pixel existing region in an optical black region; And a noise removing unit for outputting a pure pixel value obtained by subtracting the average noise value from the pixel value of the effective pixel.

Description

TECHNICAL FIELD The present invention relates to an image processing apparatus and method for improving horizontal noise,

The present invention relates to an image processing apparatus and a method for improving horizontal noise.

Recently, high-resolution image sensors have become mainstream in the application market, and problems that do not occur in conventional low-resolution image sensors occur one by one. One of the new problems in high resolution image sensors is horizontal noise.

FIG. 1 illustrates a structure of a general image sensor. FIG. 2 illustrates an ADC reference voltage varied by a power source noise, and FIG. 3 illustrates an image in which horizontal line noise is generated.

As shown in FIG. 1, the image sensor has a structure in which an analog-to-digital converter (ADC) is operated on a line-by-line basis to convert data into a digital signal after charges are charged in the photodiode.

At this time, the ADC reference voltage at the time when the analog-to-digital converter operates for each line is varied as illustrated in Fig. 2 due to power supply noise.

That is, although the number of charges charged in all the photodiodes is the same, since the reference voltage at the moment of conversion into the digital signal differs, the analog-to-digital converter outputs different data for each line.

Therefore, if the performance of the analog-to-digital converter is the same, horizontal noise due to power supply noise is generated as illustrated in FIG.

Efforts to stabilize the power supply in hardware are required to remove the horizontal line noise generated as described above. For example, additional resistors and capacitors are required to stabilize the power supply, and there may be a need for considerable consideration in signal line placement when designing the PCB pattern.

Therefore, there is a need for a method for eliminating horizontal line noise in a simpler manner.

The present invention eliminates the need for passive elements to stabilize the power supply, so that a camera module with a smaller size can be realized, and an image processing apparatus and a horizontal line noise And to provide an improvement method.

The present invention is to provide an image processing apparatus and a horizontal line noise improvement method capable of performing horizontal line noise correction and black level correction at the same time to realize improved image quality.

The present invention is to provide an image processing apparatus and a horizontal line noise improvement method capable of integrating an analog power source and a digital power source, thereby reducing the number of sensor power sources.

An object of the present invention is to provide an image processing apparatus and a horizontal line noise improvement method capable of extending a use range of a sensor gain and ensuring a brighter image at low illumination.

Other objects of the present invention will become readily apparent from the following description.

According to an aspect of the present invention, there is provided an image processing apparatus comprising: a noise value calculating unit for calculating an average noise value that is an average pixel value of a plurality of target pixels existing in a target pixel existing region in an optical black region; And a noise removing unit for outputting a pure pixel value obtained by subtracting the average noise value from the pixel value of the effective pixel.

The target pixel may be a pixel satisfying a predetermined minimum threshold value and a maximum threshold value condition for defective pixel extraction.

The noise canceller may perform a black level compensation process so that the minimum value of the effective pixel values of one frame is equal to or less than a predetermined reference value.

The noise value calculator may calculate the average noise value on a line or frame basis.

The noise value calculator may use eight or more target pixels to calculate the average noise value.

According to another aspect of the present invention, there is provided a horizontal noise improvement method performed in an image processing apparatus, comprising: receiving pixel values for each pixel from an image sensor; Calculating an average noise value that is an average pixel value of a plurality of target pixels existing in a target pixel existing region in an optical black region; And outputting a pixel value of the effective pixel by subtracting the average noise value from the pixel value of the effective pixel.

The target pixel may be a pixel satisfying a predetermined minimum threshold value and a maximum threshold value condition for defective pixel extraction.

The outputting step may precede the black level compensation process so that the minimum value of the effective pixel values of one frame is equal to or less than a predetermined reference value.

The calculating step may be performed in units of lines or frames.

Other aspects, features, and advantages will become apparent from the following drawings, claims, and detailed description of the invention.

According to the embodiment of the present invention, there is no addition of passive elements (e.g., resistors, capacitors, etc.) for stabilizing the power supply, a more compact camera module can be realized, and furthermore, And the improvement of the recognition rate can be achieved.

In addition, horizontal line noise correction and black level correction can be performed at the same time, thereby achieving improved image quality.

In addition, it is possible to integrate the analog power and the digital power supply, thereby reducing the number of sensor power sources. In general, an analog power supply and a digital power supply are separated from each other in order to stabilize the power supply (that is, to suppress generation of horizontal noise generated by noise caused by the operation of the digital circuit due to influence of the analog circuit) , It is possible to eliminate the horizontal line noise by software, so that there is no problem that the analog circuit is interfered with by the digital circuit.

In addition, the range of the sensor gain can be extended, so that a brighter image can be obtained in low light. Generally, at night, the image becomes dark due to lack of light, so the sensor gain is amplified by amplifying the signal. In this case, not only the normal signal but also the noise (power supply noise, etc.) are amplified and horizontal noise is generated. However, according to the present embodiment, it is possible to remove the horizontal line noise by software, so that the limit of amplification can be increased, and thus a brighter image can be obtained.

BRIEF DESCRIPTION OF THE DRAWINGS Fig.
2 is a diagram showing an ADC reference voltage varied by power supply noise.
3 is a diagram illustrating an image in which a horizontal line noise is generated;
4 is a diagram schematically showing a configuration of an image processing apparatus according to an embodiment of the present invention.
FIG. 5 is a diagram showing a region where an optical black pixel exists and a region where a valid pixel exists in a general image sensor. FIG.
6A and 6B are diagrams for explaining a concept of calculating a noise value according to an embodiment of the present invention;
7 is a diagram showing a difference in image quality before and after black level correction;
8 and 9 are flowcharts illustrating a method of improving horizontal line noise according to an exemplary embodiment of the present invention.
10 is a diagram illustrating a difference in image quality before and after improvement of the horizontal noise according to an exemplary embodiment of the present invention.

While the invention is susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the drawings and will herein be described in detail. It is to be understood, however, that the invention is not to be limited to the specific embodiments, but includes all modifications, equivalents, and alternatives falling within the spirit and scope of the invention.

It is to be understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, . On the other hand, when an element is referred to as being "directly connected" or "directly connected" to another element, it should be understood that there are no other elements in between.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The singular expressions include plural expressions unless the context clearly dictates otherwise. In this specification, the terms "comprises" or "having" and the like refer to the presence of stated features, integers, steps, operations, elements, components, or combinations thereof, But do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, or combinations thereof.

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

Also, the terms " part, "" unit," " module, "and the like, which are described in the specification, refer to a unit for processing at least one function or operation and may be implemented by hardware or software or a combination of hardware and software .

In the following description of the present invention with reference to the accompanying drawings, the same components are denoted by the same reference numerals regardless of the reference numerals, and redundant explanations thereof will be omitted. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description, well-known functions or constructions are not described in detail since they would obscure the invention in unnecessary detail.

FIG. 4 is a diagram schematically illustrating a configuration of an image processing apparatus according to an exemplary embodiment of the present invention. FIG. 5 is a block diagram of a conventional image sensor in which the presence of an optical black pixel and the presence of a valid pixel Fig. FIGS. 6A and 6B are diagrams for explaining the concept of noise value calculation according to an embodiment of the present invention, and FIG. 7 is a diagram illustrating differences in image quality before and after black level correction.

Referring to FIG. 4, after noise is measured using pixels included in an optical black region, noise is measured in valid pixels to output noise, The apparatus may include an input unit 410, a noise value calculating unit 420, a noise removing unit 430, and an output unit 440.

The input unit 410 receives an image signal corresponding to each pixel from an image sensor (not shown) that charges the photodiode and outputs a corresponding digital signal.

The noise value calculation unit 420 calculates a noise value for optical black pixels (i.e., target pixels) that are not dead pixels or white pixels among the pixels disposed in the optical black region An average value of pixel values is calculated and an average noise value is calculated.

The optical black region is an area formed on the periphery of the pixel regions of the image sensor as illustrated in Figs. 5A to 5D. The optical black pixel disposed in the optical black region is covered with a metal or the like, Is not influenced by the light. Of course, it is natural that the formation position of the optical black region is not limited to the peripheral portion or the like. Accordingly, the pixel value of the optical black pixels can be understood as a noise value, and the noise-removed pure pixel value can be calculated by subtracting the corresponding value from the pixel value of the valid pixels.

Referring to FIG. 4, the noise value calculation unit 420 may include a bad pixel extraction unit 422 and a noise value calculation unit 424.

When the optical black pixel for which the average noise value is calculated by the noise value calculating unit 424 is equal to or smaller than a predetermined minimum threshold value or is equal to or greater than the maximum threshold value, the defective pixel extracting unit 422 outputs the optical black pixel to the noise value calculating unit 424 in the target pixel.

The horizontal noise considered in this embodiment is a random noise caused by a power source, so that the value may vary for each line or / and frame. Therefore, it is necessary to find the value by which the noise is most removed through tuning.

In order to determine the target optical black pixel for which the average noise value is to be calculated by the noise value calculating unit 424, the defective pixel extracting unit 422 determines whether the pixel value that the pixel can have is 0 to 256, The minimum threshold value can be set to 30, and the maximum threshold value can be set to 100. [

Here, an optical black pixel having a pixel value smaller than the minimum threshold value can be regarded as a dead pixel, and an optical black pixel having a pixel value greater than or equal to the maximum threshold value can be regarded as a white pixel.

This means that an optical black pixel having a pixel value that does not satisfy the conditions of the minimum threshold value or the maximum threshold value can be judged to be malfunctioning or insensitive to light, As shown in Fig.

However, when the noise value calculation unit 420 extracts (excludes) defective pixels and calculates the average noise value only for the target pixels, only a part of the entire area of the optical black area may be targeted.

That is, as shown in FIG. 6A, the optical black area can be divided into the area a and the area b, which is different from the area b in the case of the area a, as shown in FIG. 6B, Is generated. Herein, the region b is referred to as a target pixel existing region.

Referring again to FIG. 4, the noise value calculator 424 calculates the noise value of the pixels recognized as non-defective pixels by the determination of the defective pixel extractor 422 (I.e., the average noise value) of the pixel values of the respective pixels.

The average noise value calculation method by the noise value calculation unit 424 can be summarized as shown in the following equation (1).

Here, n is the number of target pixels to be calculated for calculating the average noise value, and x _k is the pixel value of each target pixel.

The noise removing unit 430 calculates the pure pixel value of the effective pixel by subtracting the average noise value calculated by the noise value calculating unit 424 from the pixel value of the effective pixel input from the image sensor.

The noise eliminator 430 may further perform black level compensation (BLC) processing in calculating a pure pixel value.

The black level compensation compensates the image so that the pure pixel value having the lowest value among the pure pixel values for the frame is a preset reference value (for example, 0 (zero) or any natural number larger than this) Image processing technique.

7 shows a difference in image quality before and after black level correction. That is, referring to FIG. 7, it can be seen that the image (b) in which the black level is corrected is clearer than the image (a) in which the black level is not corrected.

As described above, the noise remover 430 according to the present embodiment not only removes horizontal line noise by reflecting the average noise value calculated using the pixel values of the target pixels, but also performs black level correction to provide a clearer image can do.

The output unit 440 outputs the pixel value of the effective pixel subjected to horizontal noise elimination and black level correction by the noise removing unit 430. [

FIG. 8 and FIG. 9 are flowcharts illustrating a method of improving horizontal line noise according to an embodiment of the present invention, and FIG. 10 is a diagram illustrating a difference in image quality before and after improvement of horizontal line noise according to an embodiment of the present invention.

Referring to FIG. 8, in operation 810, the image processing apparatus performs setting for horizontal noise improvement processing. The setting in step 810 may be at least one of setting an image size, setting a target pixel presence area for calculating an average noise value, setting a minimum and maximum threshold value for defective pixel extraction, setting a reference value for black level compensation, .

If an image signal is input from the image sensor in step 820, the image processing apparatus proceeds to step 830 to calculate an average noise value.

Referring to FIG. 9, in which step 830 is embodied, in step 910, the image processing apparatus starts calculating an average noise value for pixels existing in the target pixel existing area set in step 810.

In step 920, the image processing apparatus determines whether the determination of whether or not a defective pixel is performed on all the pixels existing in the target pixel existing region, pixel value summing, and the like is completed.

If there is a pixel not yet determined in the target pixel presence area, the image processing apparatus determines in step 930 whether the pixel value of the pixel satisfies the minimum / maximum threshold value for the defective pixel extraction specified in step 810.

If the predetermined minimum / maximum threshold value is not satisfied (i.e., the pixel is a defective pixel), the process returns to step 920.

However, if the predetermined minimum / maximum threshold value is satisfied (that is, if the target pixel is a predetermined pixel), the process proceeds to step 940, where pixel values of the target pixel are summed to calculate the average noise value, And then proceeds to step 920 again.

If it is determined in step 920 that all pixels existing in the target pixel existing area are bad pixels and pixel value summing is completed, the process proceeds to step 950. In step 950, the image processing apparatus calculates the pixel value of the sum pixel and the pixel value of the target pixel And calculates the average noise value using the number.

9 shows a method of calculating an average noise value with respect to all the pixels included in the target pixel existing region. However, in order to calculate the average noise value, it is not necessarily limited to all pixels included in the target pixel existing region .

Referring back to FIG. 8, in step 840, the image processing apparatus outputs a pure pixel value calculated by subtracting the average noise value calculated by the above-described process from the pixel value of the effective pixel to the receiving end.

The image processing apparatus can perform the black level compensation process before outputting the pure pixel value to the receiving end as described above.

The difference in the image quality in which the horizontal line noise is removed by the above-described process and the black level compensation process is illustrated in FIG.

10A illustrates an image in which horizontal line noise is not removed and black level compensation is not performed. In contrast, FIG. 10B illustrates an image in which horizontal line noise is removed and black level compensation is performed. .

10 (a) and 10 (b), it can be easily confirmed that the clear image quality can be obtained because the horizontal stripe pattern is removed from the image b and the black level compensation process is performed.

It is a matter of course that the above-described horizontal line noise improvement method may be performed by an automated procedure in a time-series sequence by a built-in or installed program in the digital processing apparatus. The codes and code segments that make up the program can be easily deduced by a computer programmer in the field. In addition, the program is stored in a computer readable medium readable by the digital processing apparatus, and is read and executed by the digital processing apparatus to implement the method. The information storage medium includes a magnetic recording medium, an optical recording medium, and a carrier wave medium.

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 following claims And changes may be made without departing from the spirit and scope of the invention.

410:
420: Noise value calculating unit
422: defective pixel extracting unit
424: Noise value calculation unit
430: Noise elimination
440: Output section

Claims (10)

An image processing apparatus comprising:
A plurality of pixels satisfying a predetermined minimum threshold value and a predetermined maximum threshold value among a plurality of pixels existing in a target pixel existence region excluding a region where noise is generated in a predetermined pattern irrespective of a horizontal stripe pattern in an optical black region, A noise value calculation unit for calculating an average noise value that is an average pixel value of the pixels; And
And a noise removing unit for outputting a pure pixel value obtained by subtracting the average noise value from a pixel value of an effective pixel in order to remove horizontal line noise, which is a random noise due to a power source,
Wherein the noise eliminator further performs a black level compensation process so that the minimum value has the reference value when the minimum value of the pure pixel values of the effective pixels for one frame is less than a predetermined reference value Processing device.
delete delete The method according to claim 1,
Wherein the noise value calculating unit calculates the average noise value on a line or frame basis.
The method according to claim 1,
Wherein the noise value calculator uses eight or more target pixels to calculate the average noise value.
A method for improving horizontal noise in an image processing apparatus,
Receiving a pixel value for each pixel from an image sensor;
A plurality of pixels satisfying a predetermined minimum threshold value and a predetermined maximum threshold value among a plurality of pixels existing in a target pixel existence region excluding a region where noise is generated in a predetermined pattern irrespective of a horizontal stripe pattern in an optical black region, Calculating an average noise value that is an average pixel value of the pixel; And
A black level compensation process is performed so that the minimum value of the pure pixels of the effective pixels for one frame is equal to or less than a predetermined reference value and the horizontal level compensation is performed so that a horizontal noise And outputting a pure pixel value obtained by subtracting the average noise value from a pixel value of an effective pixel in order to remove the horizontal pixel noise.
delete delete The method according to claim 6,
Wherein the calculating step is performed on a line or a frame basis.
10. A recording medium on which a program of instructions executable by a digital processing apparatus for performing the method of improving horizontal line noise according to any one of claims 6 to 9 is recorded.

Images