Dark shading correction method, dark shading correction system and imaging device

Abstract

The invention relates to a dark shading correction method, a dark shading correction system and an imaging device. A dark shading correction method and an apparatus for implementing the same provide a technique capable of performing correction for respective portions of an image causing a dark shading phenomenon that may occur at various positions as the size of a semiconductor chip increases. The dark shading correction method comprises the following steps: dividing a dark image into a plurality of first divided images; dividing each of the first divided areas to be re-divided according to a predetermined standard into a plurality of second divided areas; and calculating a correction parameter of each divided region. As a result, the dark shading correction method and apparatus perform separate dark shading correction for various portions of an image, resulting in forming an image with improved quality.

Description

Technical Field

[0001] The technology and implementation methods disclosed in this patent document generally relate to a method for correcting dark shadows of an imaging device and a device for implementing the same. Background Art

[0002] An image sensing device is a device that captures at least one image by utilizing the semiconductor characteristics that respond to light incident thereon to generate an image. In recent years, with the continuous development of the information technology (IT) industry and related technologies, the demand for high-quality and high-performance image sensing devices in various electronic devices such as smart phones and digital cameras has increased rapidly.

[0003] Image sensing devices can be roughly classified into CCD (Charge Coupled Device)-based image sensing devices and CMOS (Complementary Metal Oxide Semiconductor)-based image sensing devices. Different from the past, CMOS image sensing devices have been deeply studied and widely used rapidly.

[0004] Recently, with the increasing demand for high-quality pixel image sensing devices, the size of semiconductor chips used in image sensing devices has also increased continuously. There is a problem that as the influence of variables depending on the chip position (e.g., voltage drop) becomes greater, the shadow phenomenon depending on the position of pixels within the semiconductor chip becomes more serious, and shadow correction technology for correcting the shadow phenomenon is required. Summary of the Invention

[0005] According to an implementation method of the disclosed technology, a dark shadow correction method may include: dividing a dark image created under lightless conditions into a plurality of first regions; calculating the standard deviation of each of the plurality of first regions by using the pixel values of at least one pixel included in each of the plurality of first regions; dividing the first regions with a standard deviation greater than or equal to a preset reference standard deviation into a plurality of second regions; and calculating correction parameters for each correction target region, where the correction target regions include not only the first regions with a standard deviation less than the reference standard deviation but also the plurality of second regions.

[0006] In some implementations, the plurality of second regions may be regions where the first regions with a standard deviation greater than or equal to the reference standard deviation are divided into regions of an (m×n) matrix structure (where "m" and "n" are natural numbers), where at least one of "m" and "n" is an integer of 2 or greater.

[0007] In some implementations, the dark shadow correction method may further include: calculating the standard deviation of each of the plurality of second regions by using the pixel values of at least one pixel included in each of the plurality of second regions; and dividing the second regions with a standard deviation greater than or equal to the reference standard deviation into a plurality of third regions.

[0008] In some implementations, the dark shadow correction method may further include calculating a correction parameter for each of a plurality of third regions.

[0009] In some implementations, the plurality of third regions may be regions obtained by dividing a second region having a standard deviation greater than or equal to a reference standard deviation into regions having an (m×n) matrix structure (where "m" and "n" are natural numbers), wherein at least one of "m" and "n" is an integer of 2 or greater.

[0010] In some implementations, each correction parameter may be a value obtained by dividing a preset reference pixel value by the average value of the pixel values of the pixels in each correction target region.

[0011] In some implementations, each correction parameter may be a value obtained by dividing the average value of the pixel values of the pixels in at least one first region located at the center of the dark image by the average value of the pixel values of the pixels in each correction target region.

[0012] In some implementations, the dark shadow correction method may further include: assigning a correction level to each correction target region based on the average value of the pixel values of the pixels in each correction target region.

[0013] In some implementations, each correction parameter is a value obtained by dividing the average value of all the pixel values of the pixels included in a correction target region having a median correction level by the average value of the pixel values of the pixels in each correction target region.

[0014] According to another embodiment of the disclosed technology, an imaging device may include: an image sensing device including a first correction target region configured to have a plurality of pixels and a second correction target region configured to have a smaller number of pixels than the first correction target region; and a memory device including a first correction parameter for correcting the dark shadow of an image created by the first correction target region and a second correction parameter for correcting the dark shadow of an image created by the second correction target region, wherein the dark shadow is caused by noise in an image created under lightless conditions.

[0015] In some other implementations, the imaging device may further include: a third correction target region including a smaller number of pixels than the second correction target region, wherein the memory device further stores a third correction parameter for correcting the dark shadow of the third correction target region.

[0016] In some other implementations, the first correction target region may be one of a plurality of first divided regions obtained by equally dividing an image.

[0017] In some other implementations, the second correction target region may be one of a plurality of second divided regions obtained by equally dividing one of the plurality of first divided regions.

[0018] In some other implementations, the plurality of second divided regions may be arranged in a region of an (m×n) matrix structure (where "m" and "n" are natural numbers), and at least one of "m" and "n" is an integer of 2 or greater.

[0019] In some other implementations, the first correction parameter and the second correction parameter may have different values from each other.

[0020] According to another embodiment of the disclosed technology, a dark shadow correction system may include: an image sensing device configured to generate at least one image under lightless conditions; a dark image generator configured to generate a dark image using the generated image; a dark image dividing circuit configured to divide the dark image into a plurality of first regions and divide a first region having a standard deviation greater than or equal to a preset reference standard deviation into a plurality of second regions; an image division determiner configured to calculate the standard deviation of each of the plurality of first regions by using the pixel values of at least one pixel included in the plurality of first regions and calculate the standard deviation of each of the plurality of second regions by using the pixel values of at least one pixel included in the plurality of second regions; a correction parameter setting circuit configured to calculate a correction parameter for each correction target region, the correction target regions including the plurality of first regions and the plurality of second regions having a standard deviation less than the reference standard deviation; and a memory device configured to store the correction target regions and the correction parameters for each correction target region. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The above and other features and advantageous aspects of the disclosed technology will become apparent from the following detailed description when considered in conjunction with the accompanying drawings.

[0022] Figure 1 is a block diagram showing an example of a dark shadow correction system based on some implementations of the disclosed technology.

[0023] Figure 2 is showing an example based on some implementations of the disclosed technology Figure 1 flowchart of an example of a dark shadow correction method of the shown dark shadow correction system.

[0024] Figure 3 is showing an example based on some implementations of the disclosed technology for determining Figure 2 flowchart of an example of a method for the correction target region.

[0025] Figure 4A diagram showing examples in which a dark image is divided into a plurality of first divided regions based on some implementations of the disclosed technology. Figure 3

[0026] Figure 5 A diagram showing examples of the average value and standard deviation of pixel values included in each of a plurality of first divided regions based on some implementations of the disclosed technology. Figure 4

[0027] Figure 6 A diagram showing examples in which a part of a plurality of first divided regions of a dark image is re-divided to form a plurality of second divided regions based on some implementations of the disclosed technology. Figure 4

[0028] Figure 7 A diagram showing examples of the average value and standard deviation of pixel values included in each of a plurality of second divided regions based on some implementations of the disclosed technology. Figure 6

[0029] Figure 8 A diagram showing examples in which a part of a plurality of second divided regions of a dark image is re-divided to form a plurality of third divided regions based on some implementations of the disclosed technology. Figure 6

[0030] Figure 9 A diagram showing examples of the average value and standard deviation of pixel values included in each of a plurality of second divided regions based on some implementations of the disclosed technology. Figure 8

[0031] Figure 10 A diagram showing examples of results of calculating correction parameters for respective correction target regions of a dark image according to a first embodiment of the disclosed technology. Figure 8

[0032] Figure 11 A diagram showing examples of results of calculating correction parameters for respective correction target regions of a dark image according to a second embodiment of the disclosed technology. Figure 8

[0033] Figure 12 A flowchart showing an example of a method for calculating correction parameters according to a third embodiment of the disclosed technology. Figure 2

[0034] Figure 13 A diagram showing an example of a correction level standard according to a third embodiment of Figure 12

[0035] ​​​​​​​​​​Figure 14 This is a diagram showing an example of the result of calculating correction parameters for each correction target region of the dark image shown in Figure 8 accordance with the third embodiment of the disclosed technology.

[0036] Figure 15A This is a diagram showing an example of a method for determining the median of the correction levels shown in Figure 12 accordance with the third embodiment.

[0037] Figure 15B This is a diagram showing an example of the average pixel value of the pixels in the (to-be-corrected) target regions each having the median of the correction levels determined in Figure 15B accordance with the third embodiment of the disclosed technology.

[0038] Figure 15C This is a diagram showing an example of the result of calculating correction parameters for each correction target region of the dark image shown in Figure 8 accordance with the third embodiment of the disclosed technology. Detailed Embodiments

[0039] This patent document provides embodiments and examples of a method for correcting dark shadows of an imaging device and a device for implementing the same, which can be used in a configuration that can substantially solve one or more technical or engineering problems and alleviate limitations or disadvantages encountered in some other devices. Some embodiments of the disclosed technology relate to a technique capable of performing correction for each part of an image that causes a dark shadow phenomenon, and as the size of a semiconductor chip increases, the dark shadow phenomenon may occur at various positions. Recognizing the above problems, the dark shadow correction method and the device for implementing the same based on some implementations of the disclosed technology can perform separate dark shadow correction for each part of the image and can provide an image with improved quality.

[0040] Now, embodiments of the disclosed technology will be described in detail, and examples thereof are shown in the accompanying drawings. As long as possible, the same reference numerals will be used throughout the drawings to denote the same or similar parts. Although the present disclosure is susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the drawings. However, the present disclosure should not be construed as being limited to the embodiments set forth herein.

[0041] Hereinafter, various embodiments will be described with reference to the accompanying drawings. However, it should be understood that the disclosed technology is not limited to specific embodiments, but includes various modifications, equivalents, and / or alternatives of the embodiments. The embodiments of the disclosed technology can provide various effects that can be directly or indirectly recognized by the disclosed technology.

[0042] When describing the components of the embodiments of the disclosed technology, various terms such as first, second, etc. may be used only for the purpose of distinguishing one component from another, but the nature, order, and sequence of the components are not limited to these terms. Unless otherwise defined, all terms (including technical terms and scientific terms) used in the disclosed technology may have the same meaning as commonly understood by those of ordinary skill in the art to which the disclosed technology pertains. It will be further understood that terms such as those defined in common dictionaries may be interpreted as having meanings consistent with their meanings in the context of the prior art and the disclosed technology, and should not be interpreted in an idealized or overly formal sense unless expressly so defined herein.

[0043] Various embodiments of the disclosed technology relate to a technique capable of performing correction for respective parts of an image that causes a dark shadow phenomenon, which may occur at various positions as the size of a semiconductor chip increases.

[0044] It will be understood that both the above general description and the following detailed description of the disclosed technology are exemplary and explanatory, and are intended to provide further illustration of the claimed disclosure.

[0045] Hereinafter, reference will be made to Figures 1 to 15C describe the embodiments of the disclosed technology in detail.

[0046] Figure 1 is a block diagram showing an example of a dark shadow correction system 1 based on some implementations of the disclosed technology.

[0047] Referring to Figure 1 , the dark shadow correction system 1 may include an imaging device 1000 and a dark shadow correction device 2000.

[0048] The imaging device 1000 may include an image sensing device 1100 and a memory device 1200. The imaging device 1000 may be any kind of electronic device including a shooting function, such as a camera, a smart phone, etc.

[0049] In some implementations, the image sensing device 1100 may include: a pixel array including a plurality of pixels that convert incident light received by each pixel into a pixel signal including image information carried by the incident light, an in-pixel internal circuit for processing the pixel signal, and a logic circuit in the pixel that respectively converts the pixel signal into a digital signal.

[0050] A pixel may include a microlens, a filter, and a photoelectric conversion element. The photoelectric conversion element converts light into electric charge in response to the received light for a pixel signal. The microlens may converge incident light from the outside onto the pixel. The filter may transmit light having a specific wavelength range among the light beams included in the incident light. The photoelectric conversion element may generate photo charge in proportion to the amount of incident light that has passed through the filter and is received by the photoelectric conversion element.

[0051] The pixel internal circuit may include a floating diffusion region that accumulates the photo charge generated by the photoelectric conversion element. The source follower transistor may output an electrical signal to the logic circuit in response to the amount of photo charge accumulated in the floating diffusion region.

[0052] The logic circuit may include an analog-to-digital converter (ADC). The logic circuit may convert the electrical signal into a digital signal to generate pixel data.

[0053] The image sensing device 1100 may obtain a raw image by converting the electrical signal of each of the plurality of pixels into a digital signal. The raw image generated under the lightless condition where incident light is blocked from entering the pixels of the pixel array may be a dark raw image (DRI) representing a background pattern formed by pixel signals in the absence of incident light.

[0054] The memory device 1200 may be a storage device for storing data. The memory device 1200 may store correction parameters (e.g., correction gain) for image correction. The raw image or the dark raw image (DRI) may include a shading phenomenon caused by various factors (e.g., the position of the pixel in the pixel array or the degree of IR drop, etc.). The degree of the shading phenomenon may vary according to the position of the pixel. The shading phenomenon of the pixel may cause a difference in pixel values between pixels under the lightless condition. The memory device 1200 may store dominant data similar to the correction parameters to correct the pixel value difference. The memory device 1200 may be implemented as a non-volatile memory. For example, the memory device 1200 may include various non-volatile memory devices, such as a read-only memory (ROM) that can only read data, a one-time programmable (OTP) memory that can only write data once, an erasable programmable ROM (EPROM) memory that can erase and read the stored data, a NAND flash memory, a NOR flash memory, etc. The memory device 1200 may be provided inside the image sensing device 1100, or may be separately installed from the image sensing device 1100. Figure 1 An embodiment in which the memory device 1200 is separate from the image sensing device 1100 is shown.

[0055] The dark shading correction device 2000 may include a dark image generator 2100, a dark image division circuit 2200, an image division determiner 2300, and a correction parameter setting circuit 2400.

[0056] The dark image generator 2100 can receive a dark raw image (DRI). The dark image generator 2100 can receive at least one dark raw image (DRI) from the image sensing device 1110, and can use the received at least one dark raw image (DRI) to generate a dark image. When receiving a plurality of dark raw images (DRI) from the image sensing device 1110, the method for the dark image generator 2100 to generate a dark image may include calculating an average value of pixel values of pixels corresponding to the same positions of the dark raw images (DRI). In some implementations, each pixel value (hereinafter referred to as "at least one pixel value") of at least one pixel included in the dark raw image (DRI) may be an integer between 0 and 1023. The range of the at least one pixel value may be determined differently according to embodiments.

[0057] The dark image division circuit 2200 can divide the dark image into predetermined division regions. The dark image division circuit 2200 can divide (or segment) the dark image generated by the dark image generator 2100 into divided images. When the image division determiner 2300 determines that additional division of the dark image is necessary, the dark image division circuit 2200 can further divide (or segment) the dark image into division regions. When the image division determiner 2300 determines that additional division of the dark image is not necessary, the dark image divider 2200 can determine that the image division is completed.

[0058] The image division determiner 2300 can determine whether the dark image division circuit 2200 needs to further divide the dark image. For example, when the dark image division circuit 2200 divides the dark image into a plurality of division regions, the image division determiner 2300 can determine whether each division region needs to be further divided. If the image division determiner 2300 determines according to a predetermined criterion that there is a region to be further divided, the dark image division circuit 2200 can further divide the region to be further divided. The predetermined criterion will be described hereinafter with reference to Figure 4 the following drawings.

[0059] The correction parameter setting circuit 2400 can set correction parameters for each correction target region to be corrected. The correction target regions and the data of correction parameters (CPD) for each correction target region can be stored in the imaging device 1000 (for example, the memory device 1200).

[0060] The operation of setting correction parameters will be described hereinafter with reference to Figure 10 the following drawings.

[0061] Figure 2 is a flowchart showing an example of the dark shadow correction method of the dark shadow correction system shown based on some implementations of the disclosed technology. Figure 1 shown.

[0062] Refer toFigure 1 and Figure 2 ,the image sensing device 1100 can generate an image of a scene in a lightless state (lightless condition) where no incident light is provided or allowed to reach the pixel array (S10). According to one embodiment, the image sensing device 1100 can repeatedly generate images of the same scene multiple times (e.g., 20 times) under lightless conditions.

[0063] The dark image generator 2100 can use the images generated under lightless conditions to generate a dark image (S20). When the image sensing device 1100 generates images multiple times (e.g., 20 times), the average value of the pixel values of the pixels corresponding to the same positions of the generated images can be calculated to generate a dark image.

[0064] When the image division determiner 2300 determines that the image division is completed (when the image division determiner 2300 determines that there is no division region to be further divided), the dark image division circuit 2200 can determine the correction target region of the dark image (S30). A detailed description of the method for determining the correction target region of the dark image will be given later with reference to the drawings including Figure 3 .

[0065] Once the correction target region of the dark image is determined, the correction parameter setting circuit 2400 can calculate correction parameters for each division region included in the correction target region (S40). A detailed description of the method for calculating correction parameters will be given later with reference to the drawings including Figure 10 .

[0066] The memory device 1200 can store each division region included in the determined correction target region and the correction parameters (CPD) calculated for each division region (S50).

[0067] Figure 3 is a flowchart showing an example of a method (S30) for determining a correction target region based on some implementations of the disclosed technology. Figure 2 The operations S310 to S370 shown in

[0068] Figure 3 show the operation S30 of Figure 2 in more detail.

[0069] Refer to Figures 1 to 3, the dark image division circuit 2200 can divide the entire dark image generated by the dark image generator 2100 into a plurality of first division regions (S310). The plurality of first division regions can be divided into any shape. For example, the plurality of first division regions can have an (a×b) matrix structure. Assume that "a" and "b" are natural numbers and at least one of "a" or "b" is equal to or greater than 2. For example, "a" is equal to or greater than 2, "b" is equal to or greater than 2, or both "a" and "b" are equal to or greater than 2. For example, the dark image division circuit 2200 can divide the dark image into an image with a (3×4) matrix structure. In another example, the dark image division circuit 2200 can divide the dark image into an image with a (4×6) matrix structure.

[0070] The image division determiner 2300 can calculate the average value and standard deviation of the pixel values of at least one pixel included in each of the plurality of first division regions (S320).

[0071] The image division determiner 2300 can determine whether the standard deviation of the pixel values of the pixels included in each of the plurality of first division regions is greater than or equal to a reference standard deviation (S330). In some implementations, for example, the reference standard deviation can be 0.3. The reference standard deviation can be set differently according to the characteristics of the imaging device 1000 (such as pixel size, semiconductor chip size, etc.). In some implementations, the reference standard deviation can be preset and stored in the dark shadow correction system.

[0072] If the standard deviation of the pixel values of the pixels included in each of the plurality of first division regions is less than the reference standard deviation (S330 is "no"), the dark image division circuit 2200 can determine the plurality of first division regions as correction target regions (S370). The pixel value standard deviation can be the standard deviation of the pixel values of at least one pixel included in each of the plurality of first division regions calculated in operation S320. The average value and standard deviation of the pixel values of the pixels included in each of the first division region, the second division region, and the third division region will be briefly described as "average pixel value" and "pixel value standard deviation" respectively hereinafter.

[0073] When there is a first divided region among multiple first divided regions whose pixel value standard deviation is greater than or equal to the reference standard deviation (operation S330 is "yes"), the dark image dividing circuit 2200 may divide the first divided region whose pixel value standard deviation is equal to or greater than the reference standard deviation into multiple second divided regions (S340). The multiple second divided regions may be divided in a predetermined manner. In some implementations, each of the multiple second divided regions may have a (c×d) matrix structure. Assume that "c" and "d" are natural numbers and at least one of "c" or "d" is equal to or greater than 2. For example, "c" is equal to or greater than 2, "d" is equal to or greater than 2, or both "c" and "d" are equal to or greater than 2. For example, each of the multiple second divided regions may have a (2×2) matrix structure. In another example, each of the multiple second divided regions may have a (3×3) matrix structure.

[0074] The image division determiner 2300 may calculate the average value and standard deviation of the pixel values of at least one pixel included in each of the multiple second divided regions (S350).

[0075] The image division determiner 2300 may determine whether the standard deviation of the pixel values of the pixels included in each of the multiple second divided regions is greater than or equal to the reference standard deviation (S360). The reference standard deviation for operation S350 may be the same as the reference standard deviation for operation S330.

[0076] When the pixel value standard deviation of each of the multiple second divided regions is less than the reference standard deviation (operation S360 is "no"), the dark image dividing circuit 2200 may determine the multiple second divided regions and at least one first divided region whose pixel value standard deviation is less than the reference standard deviation as the correction target regions (S370).

[0077] When there is at least one second divided region among the multiple second divided regions whose pixel value standard deviation is greater than or equal to the reference standard deviation (operation S360 is "yes"), the dark image dividing circuit 2200 may divide the at least one second divided region whose pixel value standard deviation is equal to or greater than the reference standard deviation into multiple third divided regions (S340). The multiple third divided regions may be divided into any shape. In some implementations, each of the multiple third divided regions may have an (e×f) matrix structure. Assume that "e" and "f" are natural numbers and at least one of "e" and "f" is equal to or greater than 2. For example, "e" is equal to or greater than 2, "f" is equal to or greater than 2, or both "e" and "f" are equal to or greater than 2. For example, each of the multiple third divided regions may have a (2×2) matrix structure. In another example, each of the multiple third divided regions may have a (3×3) matrix structure.

[0078] Thereafter, operations S340 to S360 may be repeated for a plurality of second divided regions in the same manner as described above. When the additional division operations S340 to S360 are completed, the dark image division circuit 2200 may determine the final state in which the dark image is divided after the additional division operations S340 to S360 are completed as the correction target region (S370).

[0079] Hereinafter, the process of determining the target region to be used for dark image correction (S30) will be described with reference to Figure 4 the following drawings.

[0080] Figure 4 is a diagram illustrating an example in which a dark image based on some implementations of the disclosed technology is divided into a plurality of first divided regions. Figure 3

[0081] Referring to Figures 1 to 4 , the dark image 10 may be an example of an image generated by the operation S20 of Figure 2 . The dark image division circuit 2200 may divide the dark image into a plurality of first divided regions. In some implementations, the plurality of first divided regions may include first to twenty-fourth regions (101 to 124). Figure 4 is a diagram illustrating an example in which the dark image 10 is divided into first to twenty-fourth regions (101 to 124) arranged in a (4×6) matrix structure.

[0082] Figure 5 is a diagram illustrating an example of the average pixel value and the standard deviation of the pixel values of the pixels included in each of the plurality of first divided regions of Figure 4 .

[0083] In the following description depicted in the drawings including Figure 5 , numerical examples of the average pixel value, the pixel value standard deviation, the correction level, and the correction parameter value will be given as examples only for the convenience of understanding the disclosed technology. The scope of the disclosed technology is not limited to these numerical examples.

[0084] Referring to Figure 1 , Figure 4 and Figure 5 , when the image division determiner 2300 calculates the average pixel value and the pixel value standard deviation of each of the first to twenty-fourth regions (101 to 124), the calculation results as shown in Figure 5 may be obtained.

[0085] The image division determiner 2300 may determine whether there is a region in which the pixel value standard deviation is greater than or equal to the reference standard deviation among the first to twenty-fourth regions (101 to 124). For example, the reference standard deviation may be 0.3.

[0086] ​Reference Figure 5 The image division determiner 2300 can determine a reference standard deviation in which the standard deviation of the pixel values of each of the twenty-fourth region 124, the fourth to sixth regions (104 to 106), and the twelfth and thirteenth regions (112, 113) is equal to or greater than 0.3.

[0087] For example, the range of pixel values assignable to each pixel included in the image sensing device 1100 may be an integer between 0 and 1023. As the amount of light incident on a pixel increases, the pixel value of the corresponding pixel may increase. In other words, the greater the amount of light incident on a pixel, the higher the pixel value of the pixel.

[0088] The ideal pixel value of a pixel on which no light is incident can be defined as an offset value, and the offset value can be a value determined experimentally. For example, the offset value may be 64. In one embodiment where the offset value is 64, for example, the range of pixel values (x) assignable to each pixel included in the dark image 10 may be between 63 and 66 (i.e., 63 ≤ x ≤ 66). For various image sensing devices, the range of pixel values of the pixels of the dark image 10 may be different.

[0089] As the size of the semiconductor chip increases, a voltage drop phenomenon may occur according to the position of the semiconductor chip. Even when a specific voltage is applied to the semiconductor chip, as the semiconductor chip is set farther away from the voltage source, the magnitude of the voltage actually applied to the semiconductor chip may become lower due to the voltage drop phenomenon. The pixel array included in the image sensing device 1100 may be provided in the semiconductor chip.

[0090] When an image is generated under lightless conditions, the pixel values of the respective pixels of the pixel array are assumed to have an offset value of 64. However, in the case of the voltage drop phenomenon, the pixel values of the respective pixels of the pixel array may have a value other than 64. As the size of the semiconductor chip increases, the degree to which the pixel value deviates from 64 may become larger.

[0091] According to one embodiment in which 64 is set as the offset value for each pixel and the range of pixel values is from 0 to 1023, Figure 4 the calculation results of the average value and the standard deviation of the pixel values of at least one pixel included in each of the first to twenty-fourth regions (101 to 124) of the dark image 10 may be as Figure 5 shown in the example.

[0092] Figure 6 is a diagram showing an example of a dark image in which a part of a plurality of first divided regions is re-divided to form a plurality of second divided regions. Figure 4

[0093] Figure 1 Reference Figure 1 , Figure 4 andFigure 6 The dark image partitioning circuit 2200 can partition each of the fourth, fifth, sixth, twelfth, thirteenth, and twenty-fourth regions (104-106, 112, 113, 124) into a plurality of second partition regions. For convenience of description, Figure 6 FIG. shows an embodiment in which the dark image partitioning circuit 2200 partitions each of the fourth, fifth, sixth, twelfth, thirteenth, and twenty-fourth regions (104-106, 112, 113, 124) of the dark image into a plurality of second partition regions arranged in a (2×2) matrix structure. Other implementation manners are also possible, and it should be noted that each region of the dark image can also be partitioned into other structures other than the (2×2) matrix structure. For example, the dark image partitioning circuit 2200 can partition each of the fourth, fifth, sixth, twelfth, thirteenth, and twenty-fourth regions (104-106, 112, 113, 124) into second partition regions arranged in a (3×3) matrix structure, or can partition each of the fourth, fifth, sixth, twelfth, thirteenth, and twenty-fourth regions (104-106, 112, 113, 124) into partition regions arranged in a non-matrix structure. The plurality of second partition regions can include the 25th to 48th regions (201-224). More specifically, the fourth region 104 can be partitioned into the 25th to 28th regions (201-204). The fifth region 105 can be partitioned into the 29th to 32nd regions (205-208). The sixth region 106 can be partitioned into the 33rd to 36th regions (209-212). The twelfth region 112 can be partitioned into the 37th to 40th regions (213-216). The 13th region 113 can be partitioned into the 41st to 44th regions (217-220). The 24th region 124 can be partitioned into the 45th to 48th regions (221-224).

[0094] Figure 7 is a diagram showing Figure 6 the average pixel value and standard deviation of the pixel values of each of the plurality of second partition regions.

[0095] Referring to Figure 1 , Figure 6 and Figure 7 , the image partitioning determiner 2300 can calculate the average pixel value and pixel value standard deviation of each of the 25th to 48th regions (201-224) included in the plurality of second regions, and the calculation results can be the same as those in Figure 7 .

[0096] The image partitioning determiner 2300 can determine that the 30th region 206 among the 25th to 48th regions (201-224) has a pixel value standard deviation (S360) greater than or equal to 0.3 of the reference standard deviation.

[0097] Figure 8 A diagram showing that a part of a plurality of second divided regions is re-divided to form a dark image of a plurality of third divided regions. Figure 6

[0098] Referring to Figure 1 、 Figure 6 and Figure 8 ,the dark image dividing circuit 2200 can divide the 30th region 206 into a plurality of third divided regions. For the sake of description, Figure 8 Exemplarily, an embodiment is shown in which the dark image dividing circuit 2200 divides each 30th region 206 into third divided regions arranged in a (2×2) matrix structure, but the dark image dividing circuit 2200 can be divided into other structures other than the (2×2) matrix structure. For example, the dark image dividing circuit 2200 can divide each 30th region 206 into third divided regions arranged in a (3×3) matrix structure, or can be divided into divided regions of a non-matrix structure.

[0099] The plurality of third divided regions may include the 49th to 52nd regions (301 to 304). The 30th region 206 can be divided into the 49th to 52nd regions (301 to 304).

[0100] Figure 9 A diagram showing the average pixel value and standard deviation of the pixel values of the pixels included in each of the plurality of third divided regions. Figure 8

[0101] Referring to Figure 1 、 Figure 8 and Figure 9 ,the image dividing determiner 2300 can determine the average pixel value and the pixel value standard deviation of each of the 49th to 52nd regions (301 to 304) included in the plurality of third regions. Figure 9 shows exemplary results of calculating the average pixel value and the pixel value standard deviation of the 49th to 52nd regions (301 to 304) by the image segmentation determiner 2300.

[0102] The image dividing determiner 2300 can determine whether there is a region in the 49th to 52nd regions (301 to 304) whose pixel value standard deviation is greater than or equal to the reference standard deviation, and can determine that there is no region in the 49th to 52nd regions (301 to 304) whose pixel value standard deviation is greater than or equal to the reference standard deviation.

[0103] When there are no longer regions each having a pixel value standard deviation greater than or equal to the reference standard deviation, the dark image dividing circuit 2200 can determine the "correction target region". The region to be corrected (i.e., the correction target region) can be Figure 8 ​​The area shown in the dark image 10. More specifically, the area to be corrected (i.e., the correction target area) may include the first to third areas (101-103), the seventh to eleventh areas (107-111), the 14th to 23rd areas (114-123), the 25th to 29th areas (201-205), and the 31st to 52nd areas (207-224, 301-304).

[0104] After determining the correction target area, the correction parameter setting circuit 2400 may calculate correction parameters for each correction target area. The first to third embodiments showing an example method of calculating the correction parameters will be described below with reference to Figure 10 the following drawings.

[0105] Figure 10 is a diagram showing the result of calculating correction parameters for each correction target area of the dark image according to the first embodiment Figure 8 of.

[0106] Referring to Figure 1 , Figure 5 and Figures 7 to 10 , the method of calculating the correction parameters according to the first embodiment may be implemented as Figure 5 the method using the offset value of pixels described in. In the first embodiment, for example, the offset value of pixels may be set to 64. The correction parameter may be set to the value obtained by dividing the offset value 64 by the average pixel value of the pixels in each divided area included in the correction target area. For example, the correction parameter may be the gain value to be used when the imaging device 1000 performs image correction in the image signal processing stage.

[0107] For example, in the case of the first area 101, according to Figure 5 , the average pixel value may be 63.8. Therefore, the correction parameter value of the first area 101 may be 1.003 obtained by dividing 64 by 63.8. Here, 1.003 may be the value rounded to the fourth decimal place, and the correction parameter value will be written as the value rounded to the fourth decimal place hereinafter. For example, in the case of the second area 102, according to Figure 5 , the average pixel value may be 64.0. Therefore, the correction parameter value of the second area 102 may be 1 obtained by dividing 64 by 64. For example, in the case of the third area 103, according to Figure 5 , the average pixel value may be 64.2. Therefore, the correction parameter value of the third area 103 may be 0.997 obtained by dividing 64 by 64.2.

[0108] The correction parameter may refer to a parameter for removing a noise component caused by a dark shadow that may occur under lightless conditions. The correction parameter may be a parameter that is multiplied by the pixel data generated by the ADC of the image sensing device 1100. For example, the correction parameter 1.003 of the first region 101 may be multiplied by each pixel data of the pixels included in the first region 101 to perform correction for the dark shadow.

[0109] The correction parameter setting circuit 2400 may perform the same calculation process in the seventh to eleventh regions (107 to 111), the 14th to 23rd regions (114 to 123), the 25th to 29th regions (201 to 205), and the 31st to 52nd regions (207 to 224, 301 to 304), and the execution result of the calculation process is visible in Figure 10 in.

[0110] When calculating the correction parameter according to the first embodiment, the calculation result of the correction parameter is such that the pixel values of the pixels of the created dark image are closer to the offset value, so that a darker image closer to black can be generated, and the effect of correcting the dark shadow can be increased.

[0111] Figure 11 is a diagram showing the result of calculating the correction parameter for each correction target region of the dark image according to the second embodiment Figure 8 of.

[0112] Referring to Figure 1 , Figure 5 , Figures 7 to 9 and Figure 11 , the method of calculating the correction parameter according to the second embodiment may be a method of using the average value of at least one divided region located at the center of the divided regions included in the correction target region.

[0113] Figure 11 shows an example in which the central region (CA) including the ninth, tenth, fifteenth, and sixteenth regions (109, 110, 115, 116) is shown as an example of at least one divided region located at the center of the dark image. The average value of the central region (CA) may be calculated by the method represented by Equation 1 below.

[0114] [Equation 1]

[0115]

[0116] The result obtained by dividing the sum of the pixel values of all the pixels included in the ninth region 109, the tenth region 110, the fifteenth region 115, and the sixteenth region 116 by the total number of pixels included in the ninth, tenth, fifteenth, and sixteenth regions (109, 110, 115, 116) may be 64.45.

[0117] The correction parameter can be set to a value obtained by dividing the average value of the central area (CA) by the average value of the pixel values of the pixels in each divided area included in the correction target area. For example, the correction parameter can be a gain value used when the imaging device 1000 performs image correction in the image signal processing stage.

[0118] For example, in the case of the first area 101, according to Figure 5 , the average pixel value can be 63.8. Therefore, the correction parameter value for the first area 101 can be 1.010 obtained by dividing 64.45 by 63.8. For example, in the case of the second area 102, according to Figure 5 , the average pixel value can be 64.0. Therefore, the correction parameter value for the second area 102 can be 1.007 obtained by dividing 64.45 by 64. For example, in the case of the third area 103, according to Figure 5 , the average pixel value can be 64.2. Therefore, the correction parameter value for the third area 103 can be 1.004 obtained by dividing 64.45 by 64.2.

[0119] The correction parameter setting circuit 2400 can perform the same calculation process in the seventh to eleventh areas (107 to 111), the 14th to 23rd areas (114 to 123), the 25th to 29th areas (201 to 205), and the 31st to 52nd areas (207 to 224, 301 to 304), and the execution result of the calculation process can be seen in Figure 11 .

[0120] When calculating the correction parameter according to the second embodiment, the correction parameter is calculated based on the average value of the pixel values of the pixels located at the center of the pixel array, so that only some pixels set to be close to the edge area of the dark image can be intensively corrected.

[0121] Figure 12 is a flowchart showing an example of a method for calculating a correction parameter according to a third embodiment of the disclosed technology Figure 2 .

[0122] Referring to Figure 1 , Figure 12 and Figure 14 , operations S410 to S440 are operations that more specifically show an example of operation S40 for calculating the correction parameter for each correction target area in Figure 2 .

[0123] The correction parameter setting circuit 2400 can set the correction level standard by considering the maximum and minimum values of the average pixel values of the pixels in each divided area included in the correction target area (S410).

[0124] More specifically, as an example, the correction parameter setting circuit 2400 can obtain a calculated value obtained by dividing the difference between the maximum value and the minimum value by the number of correction levels, and can round the calculated value to two decimal places, so that the correction level interval to be corrected can be determined.

[0125] The correction parameter setting circuit 2400 can assign correction levels to respective correction target areas according to the determined correction level interval (S420).

[0126] The correction parameter setting circuit 2400 can determine the median of the correction levels of respective correction target areas (S430).

[0127] The correction parameter setting circuit 2400 can calculate the correction parameters of respective correction target areas by using the average value of the pixel values of all pixels included in the correction target area having the median (S440).

[0128] The value obtained by dividing the average value of the pixel values of all pixels of the correction target area having the median by the average value of the pixel values of the pixels of the correction target area can be the correction parameter of the corresponding correction area.

[0129] Figure 13 is a diagram showing the correction level standard determined according to Figure 12 the third embodiment.

[0130] According to Figure 5 , Figure 7 and Figure 9 , the minimum value of the average pixel values among the divided areas included in the correction target area can be 63.8, which is the average pixel value of the first area 101. In addition, the maximum value of the average pixel values among the divided areas included in the correction target area can be 66.0, which is the average pixel value of the 46th area (222). The difference between the maximum value and the minimum value can be calculated as 2.2. For example, the correction levels can be divided into 10 levels. As another example, the correction levels can be divided into eight levels. The correction level standard can be set to have a constant interval length. The interval length of each correction level can be determined by considering the difference between the maximum value and the minimum value.

[0131] Referring to Figure 1 , Figure 12 and Figure 13 , in an embodiment where the correction levels are divided into 10 levels, as an example value, considering that the difference between the maximum value and the minimum value can be 2.2, the correction parameter setting circuit 2400 can determine the interval length of each correction level to be 0.2. Here, 0.2 can be obtained by rounding 0.22 (= 2.2 / 10) to two decimal places. In Figure 13 , a diagram showing the correction level standard when the interval length of each correction level is set to 0.2 is shown.

[0132] Figure 14 This is a diagram showing the results of calculating correction parameters for each correction target region of a dark image according to the third embodiment Figure 8 of.

[0133] Referring to Figure 1 , Figure 5 , Figures 7 to 9 and Figures 12 to 14 , the average value of the first region 101 is 63.8, which is less than 64.2, so that the first region 101 can be assigned "level 1". For example, since the average value of the second region 102 is 64.0, which is less than 64.2, the second region 102 can be assigned "level 1". For example, the average value of the third region 103 is 64.2, which is between 64.2 and 64.4, so that the third region 103 can be assigned "level 2".

[0134] The correction parameter setting circuit 2400 can perform the same calculation process in the seventh to eleventh regions (107-111), the 14th to 23rd regions (114-123), the 25th to 29th regions (201-205), and the 31st to 52nd regions (207-224, 301-304), and the execution results of the calculation process can be seen in Figure 14 .

[0135] Figure 15A This is a diagram showing a method for determining the median of the correction level according to the third embodiment Figure 12 of.

[0136] Referring to Figure 1 , Figure 14 and Figure 15A , there are a total of 45 divided regions in the correction target region, and the correction level corresponding to the median can correspond to "level 4". Figure 15A The number "4" circled in can indicate the correction level corresponding to the 23rd median.

[0137] Figure 15B This is a diagram showing the average pixel value of the correction target region having the median of the correction level determined in Figure 15A .

[0138] Referring to Figure 1 , Figure 5 , Figures 7 to 9 , Figure 12 , Figure 14 , Figure 15B and Figure 15C , the method for calculating the correction parameter according to the third embodiment can be a method of using the average value of all pixel values of the divided regions each having "level 4" corresponding to the median of the correction level in the correction target region.

[0139] For example, referring to Figure 15B , the number of divided regions each having "Level 4" among the divided regions in the correction target region may be 8. That is, the divided regions of Level 4 may include the eleventh region 111, the sixteenth region 116, the 28th region 204, the 29th region 205, the 32nd region 208, the 39th region 215, the 40th region 216, and the 42nd region 218. When calculating the average value of all the pixels included in the eight regions, 64.66 can be calculated as Figure 15B shown.

[0140] Figure 15C is a diagram showing the results of calculating correction parameters for each correction target region of the dark image according to the third embodiment for Figure 9 .

[0141] The correction parameter can be set to the value obtained by dividing the overall average value of the pixels included in the divided regions (each having the median of the correction level) by the average value of the pixel values of the pixels included in each divided region in the correction target region. For example, the correction parameter can be the gain value to be used when the imaging device 1000 performs image correction in the image signal processing stage.

[0142] For example, in the case of the first region 101, according to Figure 5 , the average pixel value may be 63.8. Therefore, the correction parameter value of the first region 101 can be 1.013 obtained by dividing 64.66 by 63.8. For example, in the case of the second region 102, according to Figure 5 , the average pixel value may be 64.0. Therefore, the correction parameter value of the second region 102 can be 1.010 obtained by dividing 64.66 by 64.0. For example, in the case of the third region 103, according to Figure 5 , the average pixel value may be 64.2. Therefore, the correction parameter value of the second region 102 can be 1.007 obtained by dividing 64.66 by 64.2.

[0143] The correction parameter setting circuit 2400 can perform the same calculation process in the seventh to eleventh regions (107 - 111), the 14th to 23rd regions (114 - 123), the 25th to 29th regions (201 - 205), and the 31st to 52nd regions (207 - 224, 301 - 304), and the execution results of the calculation process can be seen in Figure 15C .

[0144] When calculating the correction parameter value according to the third embodiment, the correction level at the median is used so that the third embodiment can obtain a result of overall uniform correction in the first case where the distribution of the pixel values of the respective pixels of the created dark image is extremely biased, or in the second case where there are many correction target regions with a very low standard deviation (for example, 0.01 or less).

[0145] The method of calculating the correction parameter value according to the first to third embodiments described above can provide a method of calculating a correction parameter that can correct the shadow phenomenon caused by the voltage drop phenomenon that may occur according to the position of each pixel in the pixel array. The first to third embodiments are merely examples for explaining the technical idea of the disclosed technology, and the scope of the technical idea of the disclosed technology is not limited thereto. Figures 4 to 15C Only for numerically illustrating specific embodiments for illustration Figure 2 and Figure 3 the flowchart of, and is not intended to limit the technical idea of the disclosed technology to the above exemplary numerical values.

[0146] As is apparent from the above description, the dark shadow correction method and the apparatus for implementing the same based on some implementations of the disclosed technology can perform separate dark shadow correction for each part of the image and can provide an image with improved quality.

[0147] The embodiments of the disclosed technology can provide various effects that can be directly or indirectly recognized through the above patent documents.

[0148] Those skilled in the art will understand that the disclosed technology can be implemented in other specific ways than those described herein. In addition, claims not explicitly set forth in the appended claims can be presented as combinations of embodiments, or can be included as new claims through subsequent amendments after the filing of the application.

[0149] Although multiple exemplary embodiments have been described, it should be understood that modifications and enhancements to the disclosed embodiments and other embodiments can be conceived based on what is described and / or shown in this patent document.

[0150] Cross - reference to related applications

[0151] This patent document claims the priority and benefits of Korean Patent Application No. 10 - 2023 - 0179955, filed on December 12, 2023, the disclosure of which is incorporated herein by reference in its entirety as part of the disclosure of this patent document.

Claims

1. A dark shading correction method for processing an image from an image sensing device, the dark shading correction method comprising the following steps: dividing a dark image captured by the image sensing device having an array of image sensing pixels under a no-light condition into a plurality of first regions; calculating a standard deviation of each of the plurality of first regions by using a pixel value of at least one pixel included in each of the plurality of first regions; Dividing a first region whose standard deviation is greater than or equal to a preset reference standard deviation into a plurality of second regions; as well as Correction parameters are calculated for each correction target area including at least one of the plurality of second areas and a first area having a standard deviation smaller than the reference standard deviation.

2. The dark shadow correction method according to claim 1, wherein: Each of the plurality of second regions has an (m×n) matrix structure, Here, "m" and "n" are natural numbers, and at least one of "m" or "n" is an integer equal to or greater than 2.

3. The dark shadow correction method according to claim 1, further comprising the following steps: calculating a standard deviation of each of the plurality of second regions by using a pixel value of at least one pixel included in each of the plurality of second regions; as well as The second region having a standard deviation greater than or equal to the reference standard deviation is divided into a plurality of third regions.

4. The dark shadow correction method according to claim 3, further comprising the following steps: A correction parameter is calculated for each of the plurality of third regions.

5. The dark shadow correction method according to claim 3, wherein: Each of the plurality of third regions has an (m×n) matrix structure, Wherein, "m" and "n" are natural numbers, and at least one of "m" or "n" is an integer equal to or greater than 2.

6. The dark shadow correction method according to claim 1, wherein: Each of the correction parameters is obtained by dividing a preset reference pixel value by an average value of pixel values ​​of pixels in each of the correction target areas.

7. The dark shadow correction method according to claim 1, wherein: Each of the correction parameters is obtained by dividing an average value of pixel values ​​of pixels of at least one first area located at the center of the dark image by an average value of pixel values ​​of pixels of each of the correction target areas.

8. The dark shadow correction method according to claim 7, further comprising the following steps: A correction level is assigned to each of the correction target areas based on an average value of pixel values ​​of pixels of each of the correction target areas.

9. The dark shadow correction method according to claim 8, wherein: Each of the correction parameters is obtained by dividing an average value of all pixel values ​​of pixels included in the correction target area having the median value of the correction level by an average value of pixel values ​​of pixels of each of the correction target areas.

10. An imaging device, comprising: an image sensing device including an array of pixels for converting incident light into pixel signals representing an image captured in the incident light, the captured image including a first correction target area having a first number of the pixels and a second correction target area having a second number of the pixels, the second number being smaller than the first number; as well as A memory device storing a first correction parameter and a second correction parameter, wherein the first correction parameter is used to correct dark shadows in the first correction target area caused by noise existing in the first number of pixels of the image sensing device when there is no incident light on the image sensing device, and the second correction parameter is used to correct dark shadows in the second correction target area caused by noise existing in the second number of pixels of the image sensing device when there is no incident light on the image sensing device.

11. The imaging device according to claim 10, wherein: The image sensing device further comprises: a third correction target area having a third number of the pixels, the third number being smaller than the second number, The memory device further stores a third correction parameter for correcting dark shadows in the third correction target area caused by noise present in the third number of pixels of the image sensing device when no light is incident on the image sensing device.

12. The imaging device according to claim 10, wherein: The first correction target region corresponds to one of a plurality of first divided regions obtained by equally dividing the image.

13. The imaging device according to claim 12, wherein: The second correction target region corresponds to one of a plurality of second divided regions obtained by equally dividing one of the plurality of first divided regions.

14. The imaging device according to claim 13, wherein: Each of the plurality of second divided regions is arranged in a region of an (m×n) matrix structure, Wherein, "m" and "n" are natural numbers, and at least one of "m" or "n" is an integer equal to or greater than 2.

15. The imaging device according to claim 10, wherein: The first correction parameter and the second correction parameter have values ​​different from each other.

16. A dark shading correction system for processing an image captured by an image sensing device, the dark shading correction system comprising: an image sensing device that generates at least one image under dark conditions with no incident light at the image sensing device; a dark image generator that generates a dark image based on the at least one image; a dark image dividing circuit, the dark image dividing circuit dividing the dark image into a plurality of first regions, and dividing the first regions having a standard deviation greater than or equal to a preset reference standard deviation into a plurality of second regions; an image division determiner that calculates a standard deviation of each of the plurality of first regions by using a pixel value of at least one pixel included in the plurality of first regions, and calculates a standard deviation of each of the plurality of second regions by using a pixel value of at least one pixel included in the plurality of second regions; a correction parameter setting circuit that calculates correction parameters for each correction target area including the plurality of second areas and a first area whose standard deviation is smaller than the reference standard deviation; and A memory device stores the correction target areas and correction parameters for each of the correction target areas.

17. The dark shadow correction system of claim 16, wherein: The dark image is divided into the plurality of first regions arranged in an (m×n) matrix structure, wherein at least one of “m” or “n” is an integer equal to or greater than 2.

18. The dark shadow correction system of claim 16, wherein: The dark image dividing circuit further divides the second region whose standard deviation is greater than or equal to the preset reference standard deviation into a plurality of third regions.

Images