image signal processor

Abstract

An image signal processor for performing image conversion is disclosed. The image signal processor includes: a kernel generator configured to generate a target kernel comprising a pair of adjacent target pixels; a texture determiner configured to determine whether the target kernel corresponds to a flat region or a dark region by analyzing the texture of the target kernel; a threshold setting circuit configured to determine a threshold based on the determination result of the texture determiner; a defective pixel determiner configured to determine whether each target pixel is a defective pixel based on pixel data of a reference pixel having the same attributes as the pair of target pixels; and a defective pixel corrector configured to correct one of the pair of target pixels when the target pixel is determined to be a defective pixel.

Description

Technical Field

[0001] The techniques and embodiments disclosed herein generally relate to an image signal processor capable of performing image conversion. Background Technology

[0002] Image sensing devices are devices that capture optical images by converting light into electrical signals using photosensitive semiconductor materials that react to light. With the development of the automotive, medical, computer, and communications industries, the demand for high-performance image sensing devices is constantly increasing in various fields such as smartphones, digital cameras, game consoles, IoT (Internet of Things), robotics, surveillance cameras, and medical miniature cameras.

[0003] The raw image captured by an image sensing device may include original defects or defective pixels that do not correspond to normal images due to time factors. Typically, when defective pixels are present in the image sensor, a device can be used that interpolates the data of the defective pixel by using data from neighboring pixels located around it. However, when two or more defective pixels are arranged consecutively adjacent to each other in an image sensing device, using data from another defective pixel when correcting the defective pixel may result in incorrect correction of the defective pixel. Summary of the Invention

[0004] Various embodiments of this disclosure relate to an image signal processor capable of effectively correcting the pixel values ​​of defective pixels even when two or more defective pixels are adjacent to each other.

[0005] According to embodiments of the present disclosure, an image signal processor may include: a kernel generator configured to generate a target kernel comprising a pair of adjacent target pixels; a texture determiner configured to determine whether the target kernel corresponds to a flat region or a dark region by analyzing the texture of the target kernel; a threshold setting circuit configured to determine a threshold based on the determination result of the texture determiner; a defective pixel determiner configured to determine whether each target pixel is a defective pixel based on pixel data of a reference pixel having the same attributes as the pair of target pixels; and a defective pixel corrector configured to correct one of the target pixels when the target pixel is determined to be a defective pixel.

[0006] According to another embodiment of this disclosure, an image signal processor may include: a texture determiner configured to determine kernel type information by analyzing the texture of a target kernel comprising a pair of adjacent target pixels; a threshold setting circuit configured to determine a threshold based on the kernel type information; a defective pixel determiner configured to determine whether each of a pair of target pixels is a defective pixel based on the threshold; and a defective pixel corrector configured to correct one of the target pixels when the target pixel is determined to be a defective pixel.

[0007] It will be understood that the foregoing general description and the following detailed description of this disclosure are both illustrative and descriptive, and are intended to provide a further description of the claimed embodiments. Attached Figure Description

[0008] The above and other features and advantages of this disclosure will become readily apparent when considered in conjunction with the accompanying drawings, with reference to the following detailed description.

[0009] Figure 1 This is a block diagram illustrating an image signal processor based on some embodiments of the present disclosure.

[0010] Figure 2 This illustrates some embodiments based on this disclosure. Figure 1 The block diagram shown is of a defect pixel detector.

[0011] Figure 3 This illustrates some embodiments based on the present disclosure. Figure 2 The diagram shows the kernels generated by the kernel generator.

[0012] Figure 4 This illustrates some embodiments based on the present disclosure. Figure 2 The diagram shows the operation of the texture determiner in determining the texture.

[0013] Figure 5 This illustrates some embodiments based on the present disclosure. Figure 2 The diagram shows the operation of the defect pixel determiner in determining defect pixels.

[0014] Figure 6 This illustrates some embodiments based on the present disclosure. Figure 1 The diagram shows the operation of the defect pixel corrector in correcting defect pixels.

[0015] Figure 7 This illustrates some embodiments based on the present disclosure. Figure 1 The diagram shows the operation of the defect pixel corrector in correcting defect pixels.

[0016] Figure 8 This illustrates some embodiments based on this disclosure. Figure 1 The flowchart shown illustrates the operation of the image signal processor.

[0017] Figure 9 This illustrates some embodiments based on this disclosure. Figure 1 A block diagram of the computing device corresponding to the image signal processor. Detailed Implementation

[0018] This disclosure provides embodiments and examples of image signal processors capable of performing image conversions, which can be used in configurations that substantially solve one or more technical or engineering problems and mitigate limitations or drawbacks encountered in some other image signal processors. Some embodiments of this disclosure relate to an image signal processor capable of effectively correcting the pixel values ​​of defective pixels even when two or more defective pixels are present adjacent to each other. Recognizing the aforementioned problems, image signal processors based on some embodiments of this disclosure can more accurately detect and correct defective pixels even when two or more defective pixels are present adjacent to each other.

[0019] Reference will now be made in detail to some embodiments of the present disclosure as illustrated in the accompanying drawings. Wherever possible, the same reference numerals will be used throughout the drawings to denote the same or similar parts. Although the present disclosure is readily adaptable to various modifications and alternatives, specific embodiments are shown in the drawings. However, these embodiments should not be construed as limiting oneself to the embodiments set forth herein.

[0020] Various embodiments will now be described with reference to the accompanying drawings. However, it should be understood that the invention is not limited to the specific embodiments, but includes various modifications, equivalents, and / or substitutions of the embodiments. The embodiments of this disclosure can provide various effects that can be directly or indirectly identified.

[0021] Figure 1 This is a block diagram illustrating an image signal processor (ISP) 100 based on some embodiments of the present disclosure.

[0022] Reference Figure 1 The image signal processor (ISP) 100 can perform at least one image signal processing on the image data (IDATA) to generate corrected image data (IDATA_P).

[0023] The image signal processor 100 can reduce noise in image data (IDATA) and perform various types of image signal processing for image quality improvement (e.g., demosaic, defective pixel correction, gamma correction, color filter array interpolation, color matrix, color correction, color enhancement, lens distortion correction, etc.).

[0024] Furthermore, the image signal processor 100 can compress image data created by performing image signal processing for image quality improvement, enabling the image signal processor 100 to create image files using the compressed image data. Alternatively, the image signal processor 100 can recover image data from an image file. In this case, the scheme used to compress such image data can be a reversible or irreversible format. As representative examples of such compression formats, when using still images, the Joint Picture Experts Group (JPEG) format, JPEG 2000 format, etc., can be used. Additionally, when using moving images, multiple frames can be compressed according to the Moving Picture Experts Group (MPEG) standard, enabling the creation of moving image files.

[0025] Image data (IDATA) can be generated by an image sensing device that captures an optical image of a scene, but implementations are not limited thereto. The image sensing device may include: a pixel array comprising a plurality of pixels configured to sense incident light received from the scene; control circuitry configured to control the pixel array; and readout circuitry configured to output digital image data (IDATA) by converting analog pixel signals received from the pixel array into digital image data (IDATA). In some embodiments of this disclosure, image data (IDATA) is generated by the image sensing device.

[0026] The pixel array may include a color filter array (CFA), wherein the color filters are arranged according to a predetermined pattern (e.g., a Bayer pattern, a quad Bayer pattern, a nine Bayer pattern, an RGBW pattern, etc.) such that each color filter can sense light of a predetermined wavelength band. The pattern of the image data (IDATA) may be determined based on the type of pattern of the CFA. The term "predetermined" (e.g., predetermined pattern, threshold, size, distance, condition, algorithm, and wavelength band) used herein with respect to parameters means that the value of the parameter is determined before it is used in the processing or algorithm. In some embodiments, the value of the parameter is determined before the processing or algorithm begins. In other embodiments, the value of the parameter is determined during processing or during algorithm execution, but before it is used in the processing or algorithm.

[0027] The image signal processor (ISP) 100 according to embodiments of the present disclosure may include a defective pixel detector 200 and a defective pixel corrector 300.

[0028] The defective pixel detector 200 can use image data (IDATA) to detect defective pixels. The defective pixel detector 200 can detect defective pixels and output defective pixel data (DPD) to the defective pixel corrector 300.

[0029] A defective pixel refers to a pixel that does not generate pixel data corresponding to the intensity of the incident light. A defective pixel can be a pre-defined, fixed defective pixel based on pixel attributes (e.g., a phase difference detection autofocus (PDAF) pixel, a defective pixel with defects due to manufacturing process limitations, etc.), or a defective pixel that is temporarily unable to generate normal pixel data due to environmental or structural reasons. Here, a PDAF pixel can be a pixel used to obtain phase difference information to achieve autofocus functionality, and from an image data processing perspective, it can be classified as a defective pixel.

[0030] The defect pixel detector 200 can detect the location information of defect pixels from image data (IDATA).

[0031] In some implementations, for ease of description, the digital data corresponding to the pixel signals of individual pixels will hereinafter be defined as pixel data, and the collection (aggregate) of pixel data corresponding to a predetermined unit (e.g., a frame or a kernel) will hereinafter be defined as image data (IDATA). Here, a frame may correspond to the entire pixel array, and a kernel may refer to a unit used for image signal processing. In this implementation, if a pixel is included in a kernel, this may mean that the corresponding pixel is arranged to correspond to an example case of a kernel corresponding to a specific operating unit.

[0032] The defective pixel detector 200 can receive pre-stored location information of defective pixels from an image sensing device that generates image data (IDATA), and can determine whether a target pixel is a defective pixel based on the location information of the defective pixels. The image sensing device can store the location information of fixed-position defective pixels generated during the manufacturing process in an internal storage device (e.g., one-time programmable (OTP) memory), and can provide the location information of the defective pixels to the image signal processor 100. More detailed operation of the defective pixel detector 200 will be described later. Figure 2 describe.

[0033] When the defective pixel detector 200 determines that a target pixel is a defective pixel, the defective pixel corrector 300 can correct the pixel data of the target pixel based on image data including the kernel of the target pixel. In this case, the pixel data of the target pixel may refer to normal color pixel data that would be available if the target pixel were not a defective pixel.

[0034] In one implementation, the defective pixel corrector 300 can correct the pixel data of the target pixel using pixel data of pixels that have the same properties as the target pixel among the pixels included in the kernel. In another implementation, the defective pixel corrector 300 can perform defective pixel correction on a unit of a mask of a predetermined size. In this case, defective pixel correction may be an operation of calculating (e.g., linear interpolation) pixel data of at least one pixel of the same type (homogeneous) (and / or different type (heterogeneous)) as the target pixel within a mask centered on the mask, such that the target pixel to be corrected is located, and then obtaining the pixel data corresponding to the target pixel. More detailed operation of the defective pixel corrector 300 will be described later. Figure 6 and Figure 7 describe.

[0035] Figure 2 This illustrates some embodiments based on this disclosure. Figure 1 The block diagram shown is of the defect pixel detector 200.

[0036] Reference Figure 2 The defect pixel detector 200 may include a kernel generator 210, a texture determiner 220, a threshold setting circuit 230, and a defect pixel determiner 240.

[0037] Kernel generator 210 can generate a kernel for identifying defective pixels from pixel data included in image data (IDATA). The kernel generated by kernel generator 210 can be moved within the input image.

[0038] For example, kernel generator 210 can generate an operational kernel to detect defective pixels while moving pixels in two-pixel units from the beginning to the end of a pixel. An example of a kernel that can be generated by kernel generator 210 will be described below. Figure 3 To describe in more detail.

[0039] Texture determiner 220 can analyze the texture of each kernel based on image data (IDATA). Image data (IDATA) corresponding to a frame can include textures of various sizes and shapes. Texture refers to a set (or aggregation) of pixels with similarity; for example, a target object with a uniform color in a scene can be identified as a texture.

[0040] Texture can be one of the characteristics indicating whether a target kernel is a flat region, an edge (or corner) region, or a patterned region more complex than an edge region. Here, the target kernel may include the target pixels to be corrected and may refer to a unit used for image signal processing. A flat region may refer to a region in the target kernel that has generally very similar pixel data without a specific orientation, so that a flat region can be considered a textured region that is simpler than an edge region. Additionally, texture can be one of the characteristics indicating whether the target kernel is a dark region. A dark region may refer to a textured region in the target kernel that is darker than a preset brightness.

[0041] Texture determiner 220 analyzes the texture to determine whether a target kernel including the target pixel is a flat region or a dark region. In one embodiment, texture determiner 220 can also be used as a flat kernel determiner or a dark kernel determiner. Here, the target kernel may correspond to a target kernel used to determine a flat region or a dark region from a set of pixel data of a predetermined unit including the pixel data of the target pixel.

[0042] When the target kernel does not correspond to a predetermined pattern shape, the texture determiner 220 can determine the target kernel as a flat region. Here, the predetermined pattern shape may represent a corner pattern, an edge pattern, etc. For example, when the standard deviation of the pixel values ​​of each pixel included in the target kernel is less than a set value, the texture determiner 220 can determine the target kernel as a flat region. The set value may correspond to a value pre-stored in the image signal processor (ISP) 100 to determine the type of kernel. In another embodiment, the texture determiner 220 can determine whether the target kernel belongs to a flat region by setting a threshold based on the median of the pixels included in the target kernel.

[0043] Texture determiner 220 can determine the brightness of the target kernel based on individual pixel data in the target kernel. For example, texture determiner 220 can determine whether the target kernel is a dark area by calculating the median using the pixel values ​​of green pixels located around the target pixel. In this disclosure, the pixel value of a pixel can be regarded as the value of the pixel data of the pixel.

[0044] Texture determiner 220 can send kernel type information of the target kernel to threshold setting circuit 230. For example, texture determiner 220 can send kernel type information indicating whether the target kernel is a flat region or a dark region to threshold setting circuit 230. Implementation of texture determination by texture determiner 220 will be described below. Figure 4 To describe in more detail.

[0045] The threshold setting circuit 230 can receive kernel type information from the texture determiner 220 and can determine (or set) a threshold value used as a reference value for detecting defective pixels. In one embodiment, the threshold setting circuit 230 can determine (or set) the threshold in different ways based on the kernel type information of the target kernel.

[0046] When the target kernel is determined to be a flat region, the threshold setting circuit 230 can determine (or set) a first threshold. For example, when the target kernel is a flat region, the first threshold can be set based on the brightness of the target pixel. That is, the threshold setting circuit 230 can also determine a specific ratio of the brightness values ​​of the current kernel (e.g., the average pixel value of green pixels). In another embodiment, the threshold setting circuit 230 can determine (or set) the first threshold based on the standard deviation of the pixel values ​​of individual pixels located within the target kernel. In another embodiment, the threshold setting circuit 230 can determine (or set) the first threshold by comparing the standard deviation of the pixel values ​​of individual pixels located in the same channel within the target kernel with the standard deviation of the pixel values ​​of individual pixels located within the target kernel. In another embodiment, the threshold setting circuit 230 can determine (or set) the first threshold based on the median obtained by the texture determiner 220. In another embodiment, the threshold setting circuit 230 can determine (or set) the first threshold based on the pixel values ​​of the same pixels (i.e., homogeneous pixels) that have the same attributes as the target pixel.

[0047] When the target kernel is determined to be a dark region, the threshold setting circuit 230 can determine (or set) a second threshold. For example, when the target kernel is determined to be a dark region, the threshold setting circuit 230 can set the second threshold to a fixed constant.

[0048] The threshold setting circuit 230 can transmit threshold information, including the set threshold, to the defective pixel determiner 240. For example, the threshold setting circuit 230 can transmit threshold information (i.e., a first threshold) corresponding to a target kernel used as a flat region or threshold information (i.e., a second threshold) corresponding to a target kernel used as a dark region to the defective pixel determiner 240.

[0049] The defect pixel determiner 240 can receive threshold information and determine whether the target pixel is a "pair of defect pixels". Here, a pair of defect pixels can refer to a pair of defect pixels located adjacent to each other.

[0050] In one implementation, the defective pixel determiner 240 can determine a defective pixel by using the brightness difference between a target pixel and a reference pixel with the same color as the target pixel. Here, the target pixel may indicate a target object to be determined as a defective pixel or a normal pixel. Additionally, the reference pixel may correspond to multiple neighboring pixels outside the target pixel channel used as the target for defect correction.

[0051] The defective pixel determiner 240 can compare the pixel data of a target pixel with the pixel data of a reference pixel. The defective pixel determiner 240 can compare a first threshold or a second threshold with the difference obtained when comparing the pixel data of the target pixel with the pixel data of the reference pixel, and can determine whether the target pixel is a pair of defective pixels.

[0052] When a target pixel is determined to be a pair of defective pixels, the defective pixel determiner 240 can generate defective pixel data (DPD) that includes both the coordinate information and pixel data of the target pixel. In one embodiment, the defective pixel data (DPD) may include information indicating whether the defective pixel is included in the target kernel as a flat area or as a dark area. The method for determining defective pixels will be described later. Figure 5 Detailed description.

[0053] Figure 3 This illustrates some embodiments based on the present disclosure. Figure 2 The diagram shows the kernel generated by kernel generator 210.

[0054] Reference Figure 3 The kernel generated by the kernel generator 210 according to this disclosure may include (M×N) pixels arranged in a matrix structure. Here, M and N are different natural numbers, and M may be a natural number greater than N. As the kernel size increases, the resources required for detecting and correcting defective pixels may increase. As a result, kernels with a horizontal size and a vertical size smaller than the horizontal size can be used to reduce resources. That is, the kernels according to embodiments of this disclosure may correspond to asymmetric kernels, each having a horizontal size and a vertical size smaller than the horizontal size.

[0055] As an example, the kernel generated by kernel generator 210 is a (10×5) kernel unit with 10 rows and 5 columns. In the target kernel, the green (G) filter, red (R) filter, and blue (B) filter can be arranged in a four-Bayer pattern. The value of one of the colors green (G), red (R), and blue (B) can be matched with each of the four unit pixel groups (UPGs) arranged in a (2×2) matrix, resulting in the formation of a (10×5) kernel.

[0056] Although the embodiments of this disclosure present a kernel arrangement in a four-Bayer pattern for ease of description, the technical concept of this disclosure can also be applied to other kernels with color pixels arranged in other patterns such as a nine-Bayer pattern, a six-Bayer pattern, an RGBW pattern, or a single pattern. The type of image pattern is not limited to this and can be appropriately changed as needed. In addition, depending on the performance of the image signal processor (ISP) 100, the required correction accuracy, the arrangement method of the color pixels, etc., a kernel with a size other than (10×5) can also be used, but the kernel unit is not limited.

[0057] Figure 3 (A) and Figure 3 The implementation shown in (B) represents an example case where the target kernel includes a target pixel (T) corresponding to a green (G) color filter. Additionally, Figure 3 (C) and Figure 3 The implementation shown in (D) can represent an example case where the target kernel includes a target pixel (T) corresponding to a specific color filter. For example, the specific color filter could be a red (R) color filter, a blue (B) color filter, etc., but for ease of description, the red (R) color filter will be described below as a representative example.

[0058] exist Figure 3 (A) to Figure 3 In each of (D), pixels (P00–P49) can form a (10×5) kernel. Pixels (P00–P49) included in the target kernel can be grouped into multiple unit pixel groups. For example, each of the multiple unit pixel groups can include pixels corresponding to the same color filter and adjacent to each other. For example, pixels (P30, P31, P40, P41) corresponding to the green (G) color filter and adjacent to each other can be grouped into one unit pixel group. Pixels (P32, P33, P42, P43) corresponding to the red (R) color filter and adjacent to each other can be grouped into another unit pixel group. Pixels (P10, P11, P20, P21) corresponding to the blue (B) color filter and adjacent to each other can be grouped into another unit pixel group. It is understood that the remaining pixels included in the target kernel are also grouped in the same manner as described above.

[0059] In the target kernel, a target pixel (T) can correspond to a pair of pixels (P04, P05) located at the center of the bottom row. Target pixels (T) can be a pair of pixels (P04, P05) arranged adjacent to each other. In a pair of pixels (P04, P05), the target pixel arranged on the left can be defined as the left pixel (CL), and the target pixel arranged on the right can be defined as the right pixel (CR).

[0060] Reference Figure 3As shown in kernel (A), the target pixel (T) can be located in the (2×1) unit pixel group (UPG1) corresponding to the green (G) color filter. (See reference...) Figure 3 As shown in kernel (B), the target pixel (T) can be located in the pixels included in the down row of the (4×4) unit pixel group (UPG2) corresponding to the green (G) filter. (See reference...) Figure 3 The kernel shown in (C) allows the target pixel (T) to be located within a (2×1) unit pixel group (UPG3) corresponding to the red (R) color filter. (See reference...) Figure 3 The kernel shown in (D) indicates that the target pixel (T) can be located in the pixels included in the down row of the (4×4) unit pixel group (UPG4) corresponding to the red (R) color filter.

[0061] Figure 4 This illustrates some embodiments based on the present disclosure. Figure 2 The diagram shows the operation of the texture determiner 220 in determining the texture.

[0062] Figure 4 (A) and Figure 4 (B) can represent that the target kernel corresponds to the green (G) filter and includes, for example, Figure 3 The example shown in (A) is the target pixel (T) arranged in the top row. Figure 4 (C) and Figure 4 The (D) can represent the target kernel corresponding to the green (G) filter and include, for example, Figure 3 The example shown in (B) is the target pixel (T) arranged in the bottom row.

[0063] Figure 4 (E) and Figure 4 The (F) can represent the target kernel corresponding to the red (R) color filter and include, for example, Figure 3 The example shown in (C) is the target pixel (T) arranged in the top row. Figure 4 (G) and Figure 4 The (H) can represent that the target kernel corresponds to the red (R) color filter and includes, for example, Figure 3 The example shown is the target pixel (T) arranged in the bottom row line. (D)

[0064] In this disclosure, texture determiner 220 can determine whether the target kernel corresponds to a flat area or a dark area among the pixels included in the target kernel by using the median of a plurality of green pixels that include the target pixel (T) and are arranged adjacent to the target pixel (T).

[0065] For example, texture determiner 220 can calculate the median by grouping multiple green pixels into a predetermined number of adjacent pixels arranged in a specific direction. Here, the method for calculating the median can be performed by sorting the pixel values ​​included in the target kernel in ascending order and then extracting the median based on the sorting result, but the scope or spirit of the calculation method is not limited to this. For example, if there are a total of four pixels to be calculated, the median can be obtained by calculating the average of the second and third largest pixel values ​​among the four pixel values.

[0066] Since there can be two defective pixels in a group of pixels, the texture determiner 220 can group pixels in at least two different directions such that there are no two defective pixels in a group of pixels. That is, the texture determiner 220 can group pixels in each of the horizontal (hor) and vertical (ver) directions, and can perform calculation operations on each pixel group.

[0067] According to one embodiment of this disclosure, a total of 16 green pixels among the green pixels arranged in rows (R1 to R4) are used for calculation operations, but the embodiment is not limited to this, and the number of pixels used for calculation operations may also be appropriately changed to another number.

[0068] Figure 4 (A) Figure 4 (C) Figure 4 (E) and Figure 4 (G) can represent an example of the grouping of individual green pixels in the horizontal direction. That is, the texture determiner 220 can perform this calculation using the green pixels in rows (R1 to R4) arranged in multiple rows (R1 to R5).

[0069] Texture determiner 220 calculates the median by grouping four green pixels (G00, G01, G02, G03) arranged in the central region of the green pixels in row (R1) of the target kernel. The median (med1) is calculated based on the horizontally grouped green pixels (G00, G01, G02, G03) within row (R1). hor It can be represented by the following formula 1.

[0070] [Formula 1]

[0071] med1 hor =med(G00,G01,G02,G03)

[0072] Texture determiner 220 calculates the median by grouping four green pixels (G10, G11, G12, G13) arranged in the central region of the green pixels in row (R2) of the target kernel. The median (med2) is calculated based on the horizontally grouped green pixels (G10, G11, G12, G13) within row (R2). hor It can be represented by the following formula 2.

[0073] [Equation 2]

[0074] med2 hor =med(G10,G11,G12,G13)

[0075] Texture determiner 220 calculates the median by grouping four green pixels (G20, G21, G22, G23) arranged in the central region of the green pixels in row (R3) of the target kernel. The median (med3) is calculated based on the horizontally grouped green pixels (G20, G21, G22, G23) within row (R3). hor It can be represented by the following formula 3.

[0076] [Formula 3]

[0077] med3 hor =med(G20,G21,G22,G23)

[0078] Texture determiner 220 calculates the median by grouping four green pixels (G30, G31, G32, G33) arranged in the central region of the green pixels in row (R4) of the target kernel. The median (med4) is calculated based on the horizontally grouped green pixels (G30, G31, G32, G33) in row (R4). hor It can be represented by the following formula 4.

[0079] [Formula 4]

[0080] med4 hor =med(G30,G31,G32,G33)

[0081] Texture determiner 220 can calculate the horizontal median (med1) using Equations 1 to 4. hor ,med2 hor ,med3 hor ,med4 hor The maximum value among them is determined as the representative level median (med). hor ).

[0082] Figure 4 (B) Figure 4 (D) Figure 4 (F) and Figure 4 (H) can represent an example of the green pixels being grouped in the vertical direction. That is, the texture determiner 220 can perform this calculation using the green pixels in columns (C2 to C9) arranged in multiple columns (C1 to C10).

[0083] Texture determiner 220 calculates the median by grouping four green pixels (G00, G10, G20, G30) from the green pixels in columns (C2, C3) of the target kernel. The median (med1) is calculated based on the vertically grouped green pixels (G00, G10, G20, G30) within columns (C2, C3). ver It can be represented by the following formula 5.

[0084] [Formula 5]

[0085] med1 hor =med(G00,G10,G20,G30)

[0086] Texture determiner 220 calculates the median by grouping four green pixels (G00, G10, G20, G30) from the green pixels in columns (C4, C5) of the target kernel. The median (med2) is calculated based on the vertically grouped green pixels (G01, G11, G21, G31) within columns (C4, C5). ver It can be represented by the following formula 6.

[0087] [Formula 6]

[0088] med2 hor =med(G01,G11,G21,G31)

[0089] Texture determiner 220 calculates the median by grouping four green pixels (G00, G10, G20, G30) from the green pixels in columns (C6, C7) of the target kernel. The median (med3) is calculated based on the vertically grouped green pixels (G02, G12, G22, G32) within columns (C6, C7). ver It can be represented by the following formula 7.

[0090] [Formula 7]

[0091] med3 ver =med(G02,G12,G22,G32)

[0092] Texture determiner 220 calculates the median by grouping four green pixels (G00, G10, G20, G30) from the green pixels in columns (C8, C9) of the target kernel. The median (med4) is calculated based on the vertically grouped green pixels (G03, G13, G23, G33) within columns (C8, C9). ver It can be represented by the following formula 8.

[0093] [Formula 8]

[0094] med4 ver =med(G03,G13,G23,G33)

[0095] Texture determiner 220 can calculate the horizontal median (med1) using Equations 5 to 8. ver ,med2 ver ,med3 ver ,med4 ver The maximum value among them is determined as the representative level median (med). ver ).

[0096] Texture determiner 220 can select a representative level median (med). hor ) and representative vertical median (med) ver The median value with the smaller value in the target kernel is used as the representative median (med) to prevent two defective pixels from appearing in a group of pixels in the target kernel. The texture determiner 220 can determine whether the target kernel corresponds to a dark area based on the representative median (med).

[0097] Texture determiner 220 can use the aforementioned representative median to determine whether the target kernel corresponds to a flat region. The method by which texture determiner 220 determines a flat region is represented by Equation 9 below.

[0098] [Formula 9]

[0099] flat flag = |abs(G i -med) <th dyn |≤2

[0100] As can be seen from Equation 9, the texture determiner 220 can determine (or set) the threshold (th) based on the representative median (med). dyn When the representative median (med) is compared with the pixel value of the green pixel in the target kernel (G) i The difference between (i.e., the absolute value "abs") is equal to or less than the threshold (th). dyn When the number of pixels is 2 or less, the texture determiner 220 can determine the corresponding target kernel as a flat region.

[0101] When using dynamic range (i.e., the difference between the maximum and minimum values) to calculate the deviation of pixel values ​​between pixels when determining textures, sorting may be required. However, when one or more textures are determined comparatively as in this disclosure, separate sorting can be omitted, resulting in a reduction in the number of computational operations.

[0102] Figure 5 This illustrates some embodiments based on the present disclosure. Figure 2 The diagram shows the operation of the defect pixel determiner in determining defect pixels.

[0103] Figure 5 (A) and Figure 5 (B) shows that the target kernel corresponds to the green (G) filter and includes the features described above. Figure 3 The example shown in (A) is the target pixel (T) arranged in the top row. Figure 5 (C) and Figure 5 The (D) indicates that the target kernel corresponds to the green (G) filter and includes the features described above. Figure 3 The example shown in (B) is the target pixel (T) arranged in the bottom row.

[0104] Figure 5 (E) and Figure 5 The (F) indicates that the target kernel corresponds to the red (R) color filter and includes the features described above. Figure 3 The example shown in (C) is the target pixel (T) arranged in the top row. Figure 5 (G) and Figure 5 The (H) indicates that the target kernel corresponds to the red (R) color filter and includes the features described above. Figure 3 The example shown is the target pixel (T) arranged in the bottom row line. (D)

[0105] The defective pixel determiner 240 can determine whether a target pixel is a defective pixel based on a first threshold and a second threshold received from the threshold setting circuit 230. That is, when the target kernel is a flat region, the defective pixel determiner 240 can determine whether the target pixel is a defective pixel by comparing the difference between the pixel data of the target pixel and the average pixel value of the reference pixel with the first threshold. On the other hand, when the target kernel is a dark region, the defective pixel determiner 240 can determine whether the target pixel is a defective pixel by comparing the difference between the pixel data of the target pixel and the average pixel value of the reference pixel with the second threshold. In the following description, the first threshold and the second threshold will be collectively referred to as "thresholds".

[0106] The defective pixel determiner 240 compares the pixel data of a target pixel (which serves as the target object for determining whether a target pixel is a defective pixel) with the average pixel value of a reference pixel within the target kernel. When the difference between the comparison results is equal to or greater than a threshold, the defective pixel determiner 240 determines that the target pixel is a pair of defective pixels that do not have normal pixel data.

[0107] like Figure 3 As shown, in a pair of target pixels (P04, P05), the target pixel on the left can be defined as the left pixel (CL) (i.e., the first target pixel), and the target pixel on the right can be defined as the right pixel (CR) (i.e., the second target pixel). When both the left pixel (CL) and the right pixel (CR) are determined to be defective pixels, the defective pixel determiner 240 can determine the left pixel (CL) and the right pixel (CR) as "paired defective pixels".

[0108] For example, such as Figure 5 As shown in (A), the defective pixel determiner 240 compares the pixel data of the left pixel (CL) with the average pixel value of reference pixels (P00, P01, P08, P12, P13, P16, P17, P22, P23, P26, P27, P30, P31, P34, P35, P38, P40, P41, P44, P45) that correspond to green and are located adjacent to the left pixel (CL). Then, when the difference based on the comparison result is greater than or equal to a threshold, the defective pixel determiner 240 determines that the left pixel (CL) is a defective pixel.

[0109] like Figure 5 As shown in (B), the defective pixel determiner 240 compares the pixel data of the right pixel (CR) with the average pixel value of reference pixels (P01, P08, P09, P12, P13, P16, P17, P22, P23, P26, P27, P31, P34, P35, P38, P39, P44, P45, P48, P49) that are located adjacent to the right pixel (CR) and correspond to green. Then, when the difference based on the comparison result is greater than or equal to a threshold, the defective pixel determiner 240 determines that the right pixel (CR) is a defective pixel.

[0110] As is understandable, use Figure 5 (C) to Figure 5 The remaining target kernels of (H) determine the defective pixels in the same way as described above, so their redundant descriptions will be omitted in this paper for convenience.

[0111] Figure 5The implementation discloses that the number of green reference pixels to be compared with the target pixel is 20 and the number of red reference pixels is 12, but the implementation is not limited to this, and the number of reference pixels can be appropriately changed.

[0112] Figure 6 This illustrates some embodiments based on the present disclosure. Figure 1 The diagram shows the operation of the defect pixel corrector in correcting defect pixels.

[0113] Figure 6 (A) and Figure 6 The implementation shown in (B) represents an example case where the target kernel includes a target pixel (T) corresponding to a green (G) color filter. Additionally, Figure 6 (C) and Figure 6 The implementation shown in (D) can represent an example case where the target kernel includes a target pixel (T) corresponding to a red (R) color filter.

[0114] When a pair of defective pixels is detected by the defective pixel determiner 240, the defective pixel corrector 300 can correct the pair of defective pixels. For example, the defective pixel corrector 300 can correct only the pixel value of one defective pixel (e.g., the right pixel CR) in the pair of defective pixels, excluding the pixel values ​​of the remaining defective pixels (e.g., the left pixel CL).

[0115] For example, such as Figure 6 As shown in (A), the left pixel (CL) (P04) can be corrected by averaging the pixel values ​​of pixel (P05) and reference pixel (P13). Here, reference pixel (P13) can be a pixel included in a different unit pixel group than the left pixel (CL), a pixel homogeneous with the left pixel (CL), and located at the shortest distance from the left pixel (CL). That is, the correction value of the left pixel (CL) (P04) can be calculated using the following formula 10.

[0116] [Formula 10]

[0117] P04 = (P13 + P15) / 2

[0118] like Figure 6 As shown in (B), the left pixel (CL) (P04) can be corrected by averaging the pixel values ​​of pixel (P05) and reference pixel (P14). Here, reference pixel (P14) can be a pixel included in the same unit pixel group as the left pixel (CL), a pixel homogeneous with the left pixel (CL), and positioned as the closest to (or shortest in distance to) the left pixel (CL). That is, the correction value of the left pixel (CL) (P04) can be calculated using Equation 11 below.

[0119] [Equation 11]

[0120] P04 = (P14 + P05) / 2

[0121] like Figure 6 As shown in (C), the left pixel (CL) (P04) can be corrected by averaging the pixel values ​​of pixel (P05) and reference pixels (P13, P34). Here, the reference pixels (P01, P34) are pixels included in a different unit pixel group than the left pixel (CL), are homogeneous with the left pixel (CL), and are positioned as the pixels closest to the left pixel (CL) (with the shortest distance). That is, the correction value of the left pixel (CL) (P04) can be calculated using Equation 12 below.

[0122] [Equation 12]

[0123] P04 = (2 × P05 + P01 + P34) / 4

[0124] like Figure 6 As shown in (D), the left pixel (CL) (P04) can be corrected by averaging the pixel values ​​of pixel (P05) and reference pixel (P14). Here, reference pixel (P14) can be a pixel included in the same unit pixel group as the left pixel (CL), a pixel homogeneous with the left pixel (CL), and located at the shortest distance from the left pixel (CL). That is, the correction value of the left pixel (CL) (P04) can be calculated using Equation 13 below.

[0125] [Equation 13]

[0126] P04 = (P14 + P05) / 2

[0127] exist Figure 6 The implementation describes an example of correcting defective pixels based on the average pixel value of reference pixels of the same type as the target pixel, but the implementation is not limited thereto. If needed, the median value described above can be used to correct defective pixels, and the scope or spirit of the method for correcting defective pixels is not limited thereto.

[0128] For example, the defective pixel corrector 300 can replace the pixel value of the defective pixel with the median (i.e., the median of the pixel values ​​of the reference pixels). For example, the reference pixels can be enumerated sequentially according to the size of their pixel values, and the pixel value of the defective pixel can be replaced with the pixel value of the reference pixels that has the median value.

[0129] Figure 7 This illustrates some embodiments based on the present disclosure. Figure 1 The diagram shows the operation of the defect pixel corrector in correcting defect pixels. Figure 7 This illustrates an implementation of a defective pixel corrector 300 correcting a defective pixel when there is another defective pixel in the target kernel that is not the target pixel to be corrected.

[0130] Figure 7 (A) to Figure 7 The embodiment shown in (D) represents an example case where the target kernel includes a target pixel (T) corresponding to a green (G) color filter. Additionally, Figure 7 (E) to Figure 7 The embodiment shown in (I) represents an example case where the target kernel includes a target pixel (T) corresponding to a red (R) color filter. Figure 7 In the image, defective pixels will be indicated by an "×".

[0131] For example, such as Figure 7 As shown in (A), the left pixel (CL) (P04) can be corrected by averaging the pixel values ​​of pixel (P05) and reference pixel (P12). Here, reference pixel (P12) is a homogeneous pixel included in a different unit pixel group than the left pixel (CL). The homogeneous pixel located closest to pixel (P04) is pixel (P13). However, pixel (P13) is a defective pixel so that the reference pixel (P12) located close to the left pixel (CL) can be used. That is, the correction value of the left pixel (CL) (P04) can be calculated by Equation 14 below.

[0132] [Formula 14]

[0133] P04 = (P12 + P05) / 2

[0134] like Figure 7 As shown in (B), the pixel value of pixel (P13) can be used to correct the left pixel (CL) (P04). Here, the reference pixel (P13) can be a homogeneous pixel included in a different unit pixel group than the left pixel (CL), and can be located at the shortest distance from the left pixel (CL). The homogeneous pixel closest to pixel (P04) is pixel (P05). However, pixel (P05) is a defective pixel, so that only the reference pixel (P13) can be used. That is, the correction value of the left pixel (CL) (P04) can be calculated by the following formula 15.

[0135] [Formula 15]

[0136] P04 = P13

[0137] like Figure 7As shown in (C), the left pixel (CL) (P04) can be corrected by averaging the pixel values ​​of pixel (P05) and reference pixel (P15). Here, reference pixel (P15) is a homogeneous pixel included in the same unit pixel group as the left pixel (CL). The homogeneous pixel located closest to pixel (P04) is pixel (P14). However, pixel (P14) is a defective pixel so that reference pixel (P12) located close to the left pixel (CL) can be used. That is, the correction value of the left pixel (CL) (P04) can be calculated by Equation 16 below.

[0138] [Formula 16]

[0139] P04 = (P15 + P05) / 2

[0140] like Figure 7 As shown in (D), the left pixel (CL) (P04) can be corrected by averaging the pixel values ​​of the reference pixels (P14, P15). Here, the reference pixel (P14) can be a homogeneous pixel located at the shortest distance from the left pixel (CL), and also included in the same unit pixel group as the left pixel (CL). Alternatively, the reference pixel (P15) can be a pixel located at the second shortest distance from the left pixel (CL), and also included in the same unit pixel group as the left pixel (CL). Although the homogeneous pixel closest to the pixel (P04) is pixel (P05), pixel (P05) is a defective pixel, so that the reference pixels (P14, P15) located close to the left pixel (CL) can be used. That is, the correction value of the left pixel (CL) (P04) can be calculated using Equation 17 below.

[0141] [Equation 17]

[0142] P04 = (P14 + P15) / 2

[0143] like Figure 7 As shown in (E), the left pixel (CL) (P04) can be corrected by averaging the pixel values ​​of pixel (P05) and reference pixel (P34). Here, reference pixel (P34) can be a pixel that is a homogeneous pixel included in a different unit pixel group than the left pixel (CL). Although the homogeneous pixels closest to pixel (P04) are pixels (P01) and (P34), pixel (P01) is a defective pixel so that reference pixel (P34) can be used. That is, the correction value of the left pixel (CL) (P04) can be calculated by Equation 18 below.

[0144] [Formula 18]

[0145] P04 = (3 × P05 + P34) / 4

[0146] like Figure 7As shown in (F), the left pixel (CL) (P04) can be corrected by averaging the pixel values ​​of the reference pixels (P01, P34). Here, the reference pixels (P01, P34) are homogeneous pixels included in a different unit pixel group than the left pixel (CL). Although the homogeneous pixel closest to the pixel (P04) is pixel (P05), pixel (P05) is a defective pixel so that the reference pixels (P01, P34) located close to the left pixel (CL) can be used. That is, the correction value of the left pixel (CL) (P04) can be calculated by Equation 19 below.

[0147] [Formula 19]

[0148] P04 = (P01 + P34) / 2

[0149] like Figure 7 As shown in (G), the left pixel (CL) (P04) can be corrected by averaging the pixel values ​​of pixel (P05) and reference pixel (P01). Here, reference pixel (P01) is a homogeneous pixel included in a different unit pixel group than the left pixel (CL). Although the homogeneous pixels closest to pixel (P04) are pixels (P01) and (P34), pixel (P34) is a defective pixel so that reference pixel (P01) can be used. That is, the correction value of the left pixel (CL) (P04) can be calculated by the following formula 20.

[0150] [Formula 20]

[0151] P04 = (3 × P05 + P01) / 4

[0152] As is understandable, use Figure 7 (H) and Figure 7 The target kernel of (I) corrects defective pixels in the same way as described above, so its redundant description will be omitted in this paper for convenience.

[0153] Figure 8 This illustrates some embodiments based on this disclosure. Figure 1 The flowchart shown illustrates the operation of the image signal processor (ISP).

[0154] Reference Figure 8 Kernel generator 210 can generate a kernel for determining defective pixels (operation S1). The kernel generated by kernel generator 210 can move within the input image. For example, kernel generator 210 can generate a kernel while scanning the image data (IDATA) formed as a matrix array from the upper left to the lower right and moving to the right of the image data (IDATA) by a distance corresponding to two pixels for each scan action.

[0155] Subsequently, the texture determiner 220 can determine whether the target kernel corresponds to a flat region or a dark region by aiming at the kernel generated by the kernel generator 210 (operation S2).

[0156] Subsequently, texture determiner 220 can determine whether the target kernel corresponds to a flat region (operation S3). Here, the target kernel used as a flat region can mean that the target kernel is a kernel included in the flat region.

[0157] When the target kernel is determined to be a flat region, the threshold setting circuit 230 can determine (or set) a first threshold (operation S5). For example, when the target kernel is a flat region, the threshold setting circuit 230 can determine (or set) the first threshold based on the brightness of the target pixel.

[0158] When the target pixel is a green pixel, the threshold setting circuit 230 can determine the representative median (med) and threshold (th) obtained by the texture determiner 220. dyn The first threshold is determined (or set) by summing the offset value and the first threshold.

[0159] Additionally, when the target pixel is a color pixel (e.g., a red or blue pixel), the threshold setting circuit 230 can determine a first threshold using pixel data of homogeneous (similar characteristic) pixels that have the same attributes as the target pixel among the pixels included in the target kernel. Here, pixels with the same attributes as the target pixel may correspond to pixels that correspond to the same color filter or the same channel as the target pixel. Furthermore, the same channel may refer to the position of pixels with the same relative position from the center point of the microlens. For example, the threshold setting circuit 230 can determine (or set) the first threshold by calculating the minimum value among a predetermined number (e.g., 18) of pixel values ​​of the same pixels and then summing the offset values.

[0160] Texture determiner 220 can determine whether the target kernel corresponds to a dark region (operation S4). Here, the target kernel used as a dark region can mean that the target kernel is a kernel included in the dark region.

[0161] When the target kernel is determined to be a dark area, the threshold setting circuit 230 can determine (or set) a second threshold (operation S6). For example, the threshold setting circuit 230 can set the second threshold to a predetermined constant value by taking into account analog gain or digital gain, because the threshold setting circuit 230 is only enabled in dark texture mode when the target kernel is a dark area.

[0162] Subsequently, the defective pixel determiner 240 can determine a reference pixel group including at least one reference pixel within the target kernel to be compared with a unit pixel group including the target pixel. The defective pixel determiner 240 can compare the pixel data of the target pixel with the pixel data of the reference pixel. The defective pixel determiner 240 can determine whether the target pixel is a pair of defective pixels by comparing the difference obtained by comparing the pixel data of the target pixel with the pixel data of the reference pixel with the aforementioned threshold (operation S7).

[0163] For example, when the difference between the threshold included in the threshold information and the difference value is greater than or equal to a preset difference value, the defective pixel determiner 240 can determine that the target pixel is a defective pixel. For example, when the target kernel corresponds to a flat region and the difference between the first threshold and the difference value is greater than or equal to a preset difference value, the defective pixel determiner 240 can determine that the target pixel is a defective pixel. For example, when the target kernel corresponds to a dark region and the difference between the second threshold and the difference value is greater than or equal to a preset difference value, the defective pixel determiner 240 can determine that the target pixel is a defective pixel.

[0164] When the defective pixel determiner 240 determines that the target pixel is not a defective pixel (S7 is no), the defective pixel determiner 240 can proceed to operation S9. Then, the kernel generator 210 can generate an operation kernel for detecting defective pixels by moving a distance corresponding to two unit pixels (operation S9).

[0165] Subsequently, when a pair of defective pixels is detected as a determination result of the defective pixel determiner 240, the defective pixel corrector 300 can interpolate the pixel data of the pair of defective pixels using pixel data of pixels included in the target kernel (operation S8). In this disclosure, the pixel value of one defective pixel in the pair of defective pixels may not be corrected and may be excluded without correction, so that the pixel value of this defective pixel may not be included in the image data (IDATA_P). Then, only the pixel value of the remaining defective pixel can be corrected to generate corrected image data (IDATA_P) that includes the data of the interpolated target pixel.

[0166] Figure 9 It is shown that... Figure 1 Block diagram of the computing device 900 corresponding to the image signal processor.

[0167] Reference Figure 9 The computing device 900 can represent a device for performing... Figure 1 An implementation of the hardware configuration for the operation of the image signal processor 100.

[0168] The computing device 900 can be mounted on a chip independent of the chip on which the image sensing device is mounted. According to one embodiment, the chip on which the image sensing device is mounted and the chip on which the computing device 900 is mounted can be implemented in a single package (e.g., a multi-chip package (MCP)), but the embodiment is not limited thereto.

[0169] in addition, Figure 1 The internal configuration or arrangement of the image sensing device and image signal processor 100 described herein may vary depending on the implementation. For example, at least a portion of the image sensing device may be included in the image signal processor 100. Alternatively, at least a portion of the computing device 900 may be included in the image sensing device. In this case, at least a portion of the computing device 900 may be mounted together on a chip on which the image sensing device is mounted.

[0170] The computing device 900 may include a processor 910, a memory 920, an input / output interface 930, and a communication interface 940.

[0171] Processor 910 can handle execution Figure 1 The processor 910 refers to the instructions and / or data required for the operation of the components (200, 300) of the image signal processor 100 described herein. That is, processor 910 may refer to image signal processor 100, but the implementation is not limited thereto.

[0172] The memory 920 may store instructions and / or data required to perform the operation of the components (200, 300) of the image signal processor 100, and may be accessed by the processor 910. For example, the memory 920 may be volatile memory (e.g., dynamic random access memory (DRAM), static random access memory (SRAM), etc.) or non-volatile memory (e.g., programmable read-only memory (PROM), erasable PROM (EPROM), etc.), EEPROM (electrically erasable PROM), flash memory, etc.).

[0173] That is, a computer program for performing the operation of the image signal processor 100 disclosed herein is recorded in the memory 920 and executed and processed by the processor 910, thereby realizing the operation of the image signal processor 100.

[0174] The input / output interface 930 is an interface for connecting external input devices (e.g., keyboard, mouse, touch panel, etc.) and / or external output devices (e.g., display) to the processor 910 to allow sending and receiving data.

[0175] The communication interface 940 is a component that can send and receive various data with external devices (e.g., application processors, external memory, etc.) and can be a device that supports wired or wireless communication.

[0176] It is evident from the above description that even when two or more defective pixels exist adjacent to each other, the image signal processor based on some embodiments of this disclosure can detect and correct defective pixels more accurately.

[0177] The embodiments of this disclosure can provide various effects that can be directly or indirectly identified through the above-described techniques.

[0178] Although several 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 disclosure. Furthermore, these embodiments can be combined to form additional embodiments.

[0179] Cross-references to related applications

[0180] This application claims priority and benefit to Korean Patent Application No. 10-2024-0176381, filed on December 2, 2024, the entirety of which is incorporated herein by reference.

Claims

1. An image signal processor, the image signal processor comprising: A kernel generator that generates a target kernel consisting of a pair of adjacent target pixels; A texture determiner that analyzes the texture of the target kernel to determine whether the target kernel corresponds to a flat region or a dark region; A threshold setting circuit that determines a threshold based on the determination result of the texture determiner; A defective pixel determiner determines whether each of the target pixels is a defective pixel based on pixel data of a reference pixel having the same attributes as the pair of target pixels; as well as A defective pixel corrector corrects one of the target pixels when the target pixel is determined to be a defective pixel.

2. The image signal processor according to claim 1, wherein, The target kernel is an asymmetric kernel comprising multiple pixels arranged in an M×N matrix structure with M rows and N columns, where M and N are different natural numbers. The number of rows M is greater than the number of columns N.

3. The image signal processor according to claim 1, wherein, The kernel generator: The target kernel is generated by moving pixels in units of two pixels within the input image.

4. The image signal processor according to claim 1, wherein, The target kernel includes multiple pixels corresponding to the same color filter. The plurality of pixels includes the pair of target pixels, and Adjacent pixels are grouped within the plurality of pixels.

5. The image signal processor according to claim 1, wherein, The target pixels correspond to the same color filter and are positioned adjacent to each other within the same unit pixel group.

6. The image signal processor according to claim 1, wherein, The texture determiner: The target kernel is determined as either the flat region or the dark region based on the median of the target pixel and a predetermined number of reference pixels.

7. The image signal processor according to claim 6, wherein, The texture determiner: The first representative median is determined by grouping a specific number of pixels arranged adjacently in a first direction among a plurality of pixels included in the target kernel; The second representative median is determined by grouping a specific number of pixels arranged adjacently in a second direction different from the first direction among the plurality of pixels; and The dark region is determined based on the smaller of the first and second representative medians.

8. The image signal processor according to claim 7, wherein, The texture determiner: The threshold is determined based on the representative median; and When the difference between the representative median and the pixel in the target kernel corresponding to the green filter is less than or equal to 2, the corresponding target kernel is determined as the flat region.

9. The image signal processor according to claim 1, wherein, The threshold setting circuit: When the target kernel is the flat region, the threshold is determined as a first threshold based on the brightness of the pair of target pixels; and When the target kernel is the dark region, the threshold is determined to be a second threshold with a fixed constant value.

10. The image signal processor according to claim 9, in, The texture determiner also determines a threshold and an offset value, and The threshold setting circuit includes: When the pair of target pixels are pixels corresponding to the green filter, the first threshold is determined based on the median of the reference pixel and the threshold and offset value determined by the texture determiner.

11. The image signal processor according to claim 1, in, The reference pixel includes at least one color filter having the same color as the color filter corresponding to at least one of the target pixels, and The reference pixel is located in a unit pixel group that is different from the pair of target pixels.

12. The image signal processor according to claim 1, in, The pair of target pixels includes a first target pixel and a second target pixel. Wherein, the defect pixel determiner: When the difference between the pixel data of the first target pixel in the pair of target pixels and the average pixel data of the first reference pixels surrounding the first target pixel is equal to or greater than the threshold, the first target pixel is determined to be the defective pixel; and When the difference between the pixel data of the second target pixel in the pair of target pixels and the average pixel data of the second reference pixels surrounding the second target pixel is equal to or greater than the threshold, the second target pixel is determined to be the defective pixel, and The first reference pixel and the second reference pixel are included in the reference pixel.

13. The image signal processor according to claim 12, wherein, When both the first target pixel and the second target pixel are determined to be defective pixels, the defective pixel determiner will also determine the first target pixel and the second target pixel as a pair of defective pixels.

14. The image signal processor according to claim 13, wherein, The defective pixel corrector: When the pair of defective pixels are determined, the pixel values ​​of the first target pixel, excluding the pixel value of the second target pixel, are corrected.

15. The image signal processor according to claim 13, wherein, The defective pixel corrector: The first target pixel is corrected by averaging the pixel values ​​of the reference pixel.

16. The image signal processor according to claim 13, wherein, The defective pixel corrector: The first target pixel is corrected using the median of the reference pixel.

17. An image signal processor, the image signal processor comprising: A texture determiner that determines kernel type information by analyzing the texture of a target kernel, which includes a pair of adjacent target pixels; A threshold setting circuit that determines a threshold based on the kernel type information; A defective pixel determiner that determines whether each of the pair of target pixels is a defective pixel based on the threshold. as well as A defective pixel corrector corrects one of the target pixels when the target pixel is determined to be a defective pixel.

18. The image signal processor according to claim 17, wherein, The texture determiner determines the kernel type information based on whether the target kernel corresponds to a flat or dark region.

19. The image signal processor according to claim 17, wherein, The target pixels correspond to the same color filter and are positioned adjacent to each other within the same unit pixel group.

20. The image signal processor according to claim 17, in, The pair of target pixels includes a first target pixel and a second target pixel. Wherein, the defect pixel determiner: When the difference between the pixel data of the first target pixel in the pair of target pixels and the average pixel data of the first reference pixels surrounding the first target pixel is equal to or greater than the threshold, the first target pixel is determined to be the defective pixel; and When the difference between the pixel data of the second target pixel in the pair of target pixels and the average pixel data of the second reference pixels surrounding the second target pixel is equal to or greater than the threshold, the second target pixel is determined to be the defective pixel.

Images