(19)
(11)EP 3 528 493 A1

(12)EUROPEAN PATENT APPLICATION
published in accordance with Art. 153(4) EPC

(43)Date of publication:
21.08.2019 Bulletin 2019/34

(21)Application number: 16918906.5

(22)Date of filing:  27.10.2016
(51)International Patent Classification (IPC): 
H04N 5/367(2011.01)
H04N 19/00(2014.01)
H04N 5/369(2011.01)
(86)International application number:
PCT/KR2016/012155
(87)International publication number:
WO 2018/070585 (19.04.2018 Gazette  2018/16)
(84)Designated Contracting States:
AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR
Designated Extension States:
BA ME
Designated Validation States:
MA MD

(30)Priority: 14.10.2016 KR 20160133599

(71)Applicant: Hanwha Techwin Co., Ltd.
Seongnam-si, Gyeonggi-do 13488 (KR)

(72)Inventor:
  • LEE, Chang Min
    Changwon-si Gyeongsangnam-do 51542 (KR)

(74)Representative: Gulde & Partner 
Patent- und Rechtsanwaltskanzlei mbB Wallstraße 58/59
10179 Berlin
10179 Berlin (DE)

  


(54)DEFECTIVE PIXEL INFORMATION RECORDING DEVICE, AND DEFECTIVE PIXEL CORRECTION DEVICE AND METHOD


(57) An apparatus for correcting defective pixels includes an image sensor converting light provided from a subject into an electrical signal and outputting image data, a data memory storing encoded location information, a location information decoder restoring defective pixel location information from the encoded location information stored in the data memory, and a pixel corrector identifying one or more defective pixels from among pixels included in the image data using the defective pixel location information and interpolating the defective pixels using one or more neighboring pixels adjacent to each of the defective pixels.




Description

[Technical Field]



[0001] The present disclosure relates to an apparatus for recording defective pixel information and an apparatus and method for correcting defective pixels.

[Background Art]



[0002] Solid-state image pickup devices such as a charge coupled device (CCD) or a complementary metal-oxide semiconductor (CMOS) device are generally used in image pickup devices such as digital cameras or video cameras. Defective pixels that may be generated in the process of fabricating a solid-state image pickup device may lower image quality or yield. However, it is technically difficult to completely eliminate defective pixels, leading to an increase in the manufacturing cost of an image sensor. Thus, various interpolation techniques for improving image quality using pixels around defective pixels have been suggested.

[0003] Conventionally, defective pixels are detected from a solid-state image pickup device based on standard charge accumulation time as measured under predetermined conditions at the time of shipping of the solid-state image pickup device and based on output values obtained by performing exposure on the solid-state image pickup device. Location information and output level information of the defective pixels are acquired and stored, and the output of the defective pixels is interpolated using the location information and the output level information of the defective pixel and the output levels of pixels adjacent to the defective pixels. That is, conventionally, the defective pixels are detected in advance at the time of shipping of the solid-state image pickup device, and the location information of the defective pixels is stored in a memory to identify the locations of the defective pixels.

[0004] Recently, high-resolution surveillance cameras of 8 million pixels or more have been used, and even cameras supporting 4K (ultra-high definition (UHD)) videos, which have four times the number of pixels of full high definition (HD) videos, are expected to be commercialized. However, as the number of pixels of an image sensor increases, the number of defective pixels also increases. Also, given the price competitiveness of cameras, it is difficult to employ expensive image sensors with few defective pixels in cameras.

[0005] According to a conventional defective pixel correction technique, as the number of defective pixels increases, the capacity of a memory such as a static random access memory (SRAM) needs to be increased accordingly, and as a result, the manufacturing cost of an entire product increases. For example, if only 1% of the pixels of an image sensor supporting 8 million pixels become defective, the number of location data of the defective pixels amounts to 80,000, and the bit size of each defective pixel location data increases accordingly. As a result, the amount of data that needs to be stored increases exponentially.

[0006] Referring to FIG. 1, assuming that in an image sensor supporting 8 million pixels, a total of 24 bits including 12 bits for an x coordinate value and 12 bits for an y coordinate value are needed to record location information 15 of a single defective pixel, about 2 million bits of an SRAM are needed to record such defective pixel location information.

[0007] Specifically, in a case where the gain of an image sensor needs to be increased to capture a dark scene or in an unfavorable image-capturing environment with, for example, a high temperature condition or a long shutter time, the number of defective pixels may considerably increase, and the amount of defective pixel location information may also increase accordingly.

[Disclosure]


[Technical Problem]



[0008] Exemplary embodiments of the present disclosure provide an apparatus for recording defective pixel information and an apparatus for correcting defective pixels, which are for efficiently restoring defective pixels included in a high-resolution image sensor of a camera device.

[0009] Exemplary embodiments of the present disclosure also provide reducing the amount of data stored in a memory of the apparatus for correcting defective pixels by minimizing the size of defective pixel information.

[0010] Exemplary embodiments of the present disclosure also provide losslessly compressing defective pixel information adaptively according to the distribution of defective pixels included in an image sensor.

[0011] However, exemplary embodiments of the present disclosure are not restricted to those set forth herein. The above and other exemplary embodiments of the present disclosure will become more apparent to one of ordinary skill in the art to which the present disclosure pertains by referencing the detailed description of the present disclosure given below.

[Technical Solution]



[0012] According to an exemplary embodiment of the present disclosure, an apparatus for correcting defective pixels, comprising: an image sensor configured to convert light provided from a subject into an electrical signal and outputting image data; a data memory configured to store encoded location information; a location information decoder configured to defective pixel location information from the encoded location information stored in the data memory; and a pixel corrector configured to identify one or more defective pixels from among pixels included in the image data using the defective pixel location information and interpolating the defective pixels using one or more neighboring pixels adjacent to each of the defective pixels, wherein the location information decoder is configured to select one of a first decoding method and a second decoding method, which is different from the first decoding method, based on a distance between the defective pixels and restores the defective pixel location information using the selected decoding method.

[0013] An apparatus for recording defective pixel information includes: an image sensor; a pixel inspector identifying one or more defective pixels from among pixels included in the image sensor and providing location information of the defective pixels; a location information encoder generating encoded location information by encoding the location information provided by the pixel inspector; and a data memory storing the encoded location information, wherein the location information encoder selects one of a first encoding method and a second encoding method, which is different from the first encoding method, based on a distance between the defective pixels and encodes the location information provided by the pixel inspector using the selected encoding method.

[Advantageous Effects]



[0014] According to the aforementioned and other exemplary embodiments of the present disclosure, defective pixels included in a high-resolution image sensor can be efficiently restored without using a memory with large storage capacity.

[0015] In addition, the size of pixel information for use in correcting defective pixels can be minimized.

[Description of Drawings]



[0016] 

FIG. 1 is a schematic view showing the data format of conventional defective pixel location information;

FIG. 2 is a block diagram of an apparatus for recording defective pixel information according to an exemplary embodiment of the present disclosure;

FIG. 3 is a block diagram of a location information encoder of the apparatus of FIG. 2;

FIG. 4 is a schematic view showing a sensor panel or image data consisting of a plurality of pixels;

FIG. 5 is a schematic view showing location information encoded in accordance with a first encoding (or decoding) method;

FIG. 6 is a schematic view showing location information encoded in accordance with a second encoding (or decoding) method;

FIG. 7 is a schematic view showing the correlations between defective pixels when using horizontal raster scanning according to an exemplary embodiment of the present disclosure;

FIG. 8 is a schematic view showing the correlations between defective pixels when using vertical raster scanning according to an exemplary embodiment of the present disclosure;

FIG. 9 is a schematic view showing location information of multiple defective pixels, stored in a data memory;

FIG. 10 is a block diagram of an apparatus for correcting defective pixels according to an exemplary embodiment of the present disclosure;

FIG. 11 is a schematic view showing an example of how to interpolate a defective pixel using one or more valid pixels around the defective pixel T according to an exemplary embodiment of the present disclosure;

FIG. 12 is a block diagram of a location information decoder of the apparatus of FIG. 10; and

FIG. 13 is a schematic view illustrating an example of how to encode and decode location information according to an exemplary embodiment of the present disclosure.


[Modes of the Invention]



[0017] Advantages and features of the present disclosure and methods of accomplishing the same may be understood more readily by reference to the following detailed description of exemplary embodiments and the accompanying drawings. The present disclosure may, however, be embodied in many different forms and should not be construed as being limited to the exemplary embodiments set forth herein. Rather, these exemplary embodiments are provided so that this disclosure will be thorough and complete and will fully convey the concept of the present disclosure to those skilled in the art, and the present disclosure will only be defined within the scope of the appended claims. Like reference numerals indicate like elements throughout the present disclosure.

[0018] Exemplary embodiments of the present disclosure will hereinafter be described with reference to the accompanying drawings.

[0019] FIG. 2 is a block diagram of an apparatus 100 for recording defective pixel information according to an exemplary embodiment of the present disclosure. Referring to FIG. 2, the apparatus 100 may include a test signal applicator 10, an image sensor 20, a pixel inspector 30, a location information encoder 40, a data memory 50, a system processor 90, and a system memory 95.

[0020] The image sensor 20, which is a solid-state image pickup device converting light emitted from a subject into image data (or an electrical signal), may be implemented as a charge coupled device (CCD) or a complementary metal-oxide semiconductor (CMOS) device.

[0021] The test signal applicator 10 applies a test signal to the image sensor 20 to test the image sensor 20. The test signal may be a simple white image or a red (R)-green (G)-blue (B) image for testing for each channel. The test signal may be applied to the image sensor 20 according to various temperature standards in consideration that defective pixels may be generated depending on temperature. Alternatively, the test signal may be applied to the image sensor 20 as light from an actual subject, rather than as a physical signal.

[0022] The pixel inspector 30 tests the image sensor 20 for any defective pixel by comparing a target pixel value that is supposed to be generated in response to the test signal applied by the test signal applicator 10 with pixel values that are actually output by the image sensor 20. Defective pixels detected by the pixel inspector 30 may be dead pixels or may simply be pixels that return pixel values that are different from the target pixel value by more than a predetermined reference level. Alternatively, pixels that do not satisfy a predetermined standard depending on the temperature of the image sensor 20 may also be determined as defective pixels.

[0023] The pixel inspector 30 detects defective pixels from the image sensor 20 and outputs location information of the defective pixels. The location information may be, for example, horizontal and vertical coordinates of each of the defective pixels.

[0024] The location information encoder 40 encodes (or losslessly compresses) the location information output by the pixel inspector 30 and records the encoded location information in the data memory 50. The location information encoder 40 will be described later in detail with reference to FIG. 3.

[0025] The data memory 50 may be implemented as various storage devices such as a random access memory (RAM), a flash memory, or a solid state disk (SSD). Particularly, the data memory 50 may be provided as a static RAM (SRAM).

[0026] The system processor 90 is a controller controlling the entire apparatus 100, and programs that can be executed by the system processor 90 are stored in the system memory 95 and are sequentially read to be executed. A region for storing system information or user settings information is provided in the system memory 46 so that various information or settings can be read and restored later as necessary. The system memory 95 may be implemented as a RAM, a flash memory, an SSD, or the like.

[0027] FIG. 3 is a block diagram of the location information encoder 40. Referring to FIG. 3, the location information encoder 40 generates encoded location information by encoding the defective pixel location information provided by the pixel inspector 30 and stores the encoded location information in the data memory 50. Specifically, the location information encoder 40 may select one of a plurality of encoding methods based on the distance between the defective pixels and may encode the defective pixel location information using the selected encoding method. Here, the distance between the defective pixels may be the distance between each pair of adjacent defective pixels in scan order (e.g., the distance, in a vertical direction, between the defective pixels).

[0028] The defective pixel location information provided by the pixel inspector 30 is temporarily stored in a data buffer 42 and is then provided to a subtractor 43 and a data format recorder 48.

[0029] The subtractor 43 subtracts the vertical coordinate of a previous defective pixel that is followed by a current defective pixel in scan order from the vertical coordinate of the current defective pixel and the result of the subtraction to a comparer 44 as a difference value DIFF. The term "scan order," as used herein, refers to the order of sequentially scanning a plurality of pixels that form a single sensor panel or image data.

[0030] A sub-area size provider 41 generates sub-area size V_comp into which an entire area 55 is divided, and provides the sub-area size V_comp to the comparer 44 and to a sub-area index calculator 45. FIG. 4 illustrates an image sensor panel or image data that consists of a plurality of pixels. Referring to FIG. 4, the entire area 55 consists of a total of 40962048 pixels (i.e., 4096 pixels in a horizontal direction and 2048 pixels in a vertical direction). The entire area 55 may be divided into a plurality of sub-areas in the vertical direction, and the sub-area size V_comp refers to the size of the plurality of sub-areas. FIG. 4 shows three defective pixels, i.e., defective pixels P(n-1), P(n), and P(n+1). The defective pixels P(n-1) and P(n) belong to the same sub-area (REF=0), and the defective pixel P(n+1) belong to a different sub-area (REF=k-1) from the other two defective pixels P(n-1) and P(n). The relative position of the defective pixel P(n+1) in the sub-area where the defective pixel P(n+1) belongs is represented as "IND".

[0031] The sub-area size V_comp may be fixed or may be (proportionally) variable depending on the size, in the vertical direction, of the entire area 55, i.e., V_axis. Alternatively, the sub-area size V_comp may be generated by user input. For example, when V_axis is 2048, as illustrated in FIG. 4, the sub-area size V_comp may be set to 2n (where 2n=32, 64, ...).

[0032] The comparer 44 compares the sub-area size V_comp provided by the sub-area size provider 41 and the difference value DIFF provided by the subtractor 43. Then, if the difference value DIFF is smaller than, or the same as, the sub-area size V_comp, the comparer 48 notifies the data format recorder 48 to generate a data format 52 as illustrated in FIG. 5. On the other hand, if the difference value DIFF is greater than the sub-area size V_comp, the comparer 48 notifies the area index calculator 45 to generate a data format 54 as illustrated in FIG. 6.

[0033] Upon being notified by the comparer 44, the data format recorder 48 generates a data format 52 as illustrated in FIG. 5 (in accordance with a first encoding method). The data format 52 consists of a "DIFF_V" field, which records the difference value DIFF obtained from the vertical coordinates of the previous and current defective pixels, and a "H_POSITION" field, which records non-compressed horizontal coordinate information of the current defective pixel. The difference value DIFF may be recorded in the "DIFF_V" field as, for example, 5-bit information. In this case, the difference value DIFF may have a maximum of 25, i.e., 32. Since the maximum of the difference value DIFF is the same as the sub-area size V_comp, the sub-area size V_comp may be limited up to a value of 32. However, the size of the "DIFF_V" field and the sub-area size V_comp are not particularly limited.

[0034] The bit quantity of the difference value DIFF is limited by the sub-area size V_comp, regardless of the size of the entire area in the vertical direction, i.e., V_axis. For example, when the sub-area size V_comp is 32, as illustrated in FIG. 32, the difference value DIFF may be a value not exceeding 32. Thus, the number of bits allocated to the difference value DIFF can also be reduced to five, as illustrated in FIG. 5.

[0035] Upon being notified by the comparer 44, the sub-area index calculator 45 calculates an index REF of a vertical sub-area where a vertical coordinate Yn of the current defective pixel belongs. For example, as illustrated in FIG. 4, the index REF of the vertical sub-area where the defective pixels P(n-1) and P(n) both belong is 0, and the index REF of the vertical sub-area where the defective pixel P(n+1) belongs is k-1. The index REF of the vertical sub-area of the current defective pixel may be calculated by dividing the vertical coordinate Yn of the defective pixel P(n) by the sub-area size V_comp and rounding down the result of the division.

[0036] A local position calculator 46 calculates a local position IND of the vertical coordinate Yn of the defective pixel P(n) in the sub-area where the defective pixel P(n) belongs based on the sub-area size V_comp provided by the sub-area size provider 41 and the index REF provided by the sub-area index calculator 45. Referring to FIG. 4, if the current defective pixel is the defective pixel P(n+1), a local position IND of the defective pixel P(n+1) is determined by the vertical distance (in the number of pixels) from an upper boundary 51 of the sub-area (REF=k-1) where the defective pixel P(n+1) belongs to the defective pixel P(n+1). The local position IND of the defective pixel P(n+1) may be calculated by subtracting the value obtained by multiplying the index REF of the defective pixel P(n+1) by the sub-area size V_comp from a vertical coordinate Yn+1 of the defective pixel P(n+1).

[0037] The data format recorder 48 generates a data format 54 as illustrated in FIG. 6, using the index REF provided by the sub-area index calculator 45 and the local position IND provided by the local position calculator 46. The data format 54 consists of a first field (i.e., a "CODEWORD" field) in which a codeword is recorded, a second field (i.e., an "REF_LINE_INDEX" field) in which the index REF of the vertical sub-area of the current defective pixel is recorded, a third field (i.e., an "IND_DIFF_V" field) in which the local position IND of the vertical coordinate of the current defective pixel is recorded, and a fourth field (i.e., an "H_POSITION" field) in which non-compressed horizontal coordinate information of the current defective pixel is recorded, in that order. In the data format 54 of FIG. 6, the first and third fields may both have seven bits, and the second and fourth fields may both have ten bits.

[0038] The codeword recorded in the first field is an identifier bit of a predetermined bit size for distinguishing the data format 54, which is generated when the difference value DIFF is greater than the sub-area size V_comp and is encoded in accordance with a second encoding method, from the data format 52, which is generated when the difference value DIFF is smaller than, and the same as, the sub-area size V_comp and is encoded in accordance with the first encoding method. The codeword may preferably consist of bits that do not coincide with those of the "DIFF_V" field. Examples of the codeword consisting of bits that do not coincide with those of the "DIFF_V" field will be described later with reference to FIG. 13. Alternatively, a 1-bit identifier field may be added to the front of each of the data formats 52 and 54 of FIGS. 5 and 6, instead of using the codeword, so that the data formats 52 and 54 are properly distinguishable at a decoder's end.

[0039] The data format recorder 48 may generate either the data format 52 of FIG. 5, which is encoded in accordance with the first encoding method, or the data format 54 of FIG. 6, which is encoded in accordance with the second encoding method, through the comparison of the difference value DIFF and the sub-area size V_comp. In this process, vertical coordinate information of defective pixels is compressed, but horizontal coordinate information of the defective pixels is not compressed. The reason is as follows. In horizontal raster scanning, which is a commonly-used scanning method, there are gradual increases in vertical coordinate information between defective pixels, and thus, the correlations in vertical coordinate information between defective pixels are high. On the other hand, there are frequent repeated increases and decreases in horizontal coordinate information between defective pixels, and thus, the correlations in horizontal coordinate information between defective pixels are low. Accordingly, the application of a compression algorithm to horizontal coordinate information of defective pixels may undesirably cause the amount of data of location information to increase.

[0040] FIG. 7 is a schematic view showing the correlations between defective pixels when using horizontal raster scanning according to an exemplary embodiment of the present disclosure. Referring to FIG. 7, when there are three defective pixels, i.e., P(n-1), P(n), and P(n+1), the horizontal coordinates of the defective pixels P(n-1), P(n), and P(n+1) vastly differ from one another even though the defective pixels P(n-1), P(n), and P(n+1) are located adjacent to one another. This phenomenon becomes apparent when considering typical aspect ratios at which images are longer in a horizontal direction than in a vertical direction. On the other hand, the vertical coordinates of the defective pixels P(n-1), P(n), and P(n+1) do not differ much from one another only by zero or one.

[0041] On the contrary, referring to FIG. 8, in the case of using vertical raster scanning, the vertical coordinates of the defective pixels P(n-1), P(n), and P(n+1) vastly differ from one another, but the horizontal coordinates of the defective pixels P(n-1), P(n), and P(n+1) do not differ much from one another. Thus, vertical coordinate information of the defective pixels P(n-1), P(n), and P(n+1) may not be compressed, and only horizontal coordinate information of the defective pixels P(n-1), P(n), and P(n+1) may preferably be compressed. In the description that follows, it is assumed that typical horizontal raster scanning is used, as illustrated in FIG. 7.

[0042] Referring again to FIG. 3, the data format provided by the data format recorder 48 may be further compressed losslessly by an entropy encoder 49. Various encoding algorithms such as Huffman encoding, variable length encoding, binary arithmetic encoding, or the like may be used by the entropy encoder 49. An encoded bitstream produced by the entropy encoder 49 is stored in the data memory 50. As illustrated in FIG. 9, the encoded bitstream includes a plurality of units, regardless of the data format provided by the data format recorder 48 is the data format 52 or the data format 54, and each of the plurality of units sequentially includes vertical location information EYn and horizontal location information EXn of a defective pixel Pn. The horizontal location information EXn has the same format regardless of the type of encoding method used, but the vertical location information EYn has completely different formats depending on its respective difference value DIFF, as described above with reference to FIGS. 5 and 6. Thus, the vertical horizontal information EYn comes ahead of the horizontal location information EXn in each of the plurality of units of the encoded bitstream to indicate the type of encoding method used.

[0043] Encoding location information provided by the location information encoder 40 and stored in the data memory 50 is transmitted to an apparatus 200 for correcting defective pixels.

[0044] FIG. 10 is a block diagram of an apparatus 200 for correcting defective pixels according to an exemplary embodiment of the present disclosure. Referring to FIG. 10, the apparatus 200 may include a photographing lens 110, an image sensor 120, an analog-to-digital converter (ADC) 125, a pixel corrector 130, a location information decoder 140, a data memory 150, and an image processor 160 and may further include a system processor 190, a system memory 195, a sensor controller 112, a lens controller 114, an image memory 165, and an image display 170.

[0045] Light reflected from a subject is received by the image sensor 120 via the photographing lens 110. The image sensor 120 is a solid-state image pickup device converting the reflected light into image data (or an electrical signal) and may be implemented as a CCD or a CMOS device. The image sensor 120 may be a color sensor in which color filters of three colors, i.e., red (R), green (G), and blue (B) colors are arranged in a Bayer pattern. In one exemplary embodiment, pixel data may be output in such a manner that a pixel array output in the order of R→G→R→G→··· and a pixel array output in the order of β→G→β→G→··· can be repeatedly alternated with each other. However, the present disclosure is not limited to this exemplary embodiment, and the order of colors output this output order is an example, and the order in which to output colors may vary depending on the arrangement of the color filters in the image sensor 120.

[0046] An RGB analog signal output by the image sensor 120 is converted into a digital signal by the ADC 120 through a variety of analog processing, and the digital signal is output to the pixel corrector 130 as digital pixel data.

[0047] A data format stored in the data memory 150 regarding defective pixels is provided to the location information decoder 140, and the location information decoder 140 restores defective pixel location information by decoding the data format provided by the location information encoder 40 of the apparatus 100. The data format stored in the data memory 150 is encoded location information stored in the data memory 50 of the apparatus 100.

[0048] The location information decoder 140 will be described later in detail with reference to FIG. 12.

[0049] The pixel corrector 130 identifies a defective pixel included in the pixel data provided by the ADC 125 using the defective pixel location information provided by the location information decoder 140. Specifically, the pixel corrector 130 detects a defective pixel from among the pixels of the pixel data using the defective pixel location information and interpolates the detected defective pixel using at least one pixel around the detected defective pixel.

[0050] FIG. 11 is a schematic view showing an example of how to interpolate a defective pixel T using one or more valid pixels around the defective pixel T according to an exemplary embodiment of the present disclosure. Referring to FIG. 11, the defective pixel T can be interpolated using pixels a through h around the defective pixel T. For example, the defective pixel T may be interpolated using the mean or median of the pixels a through h. Alternatively, the defective pixel T may be interpolated using a weighted mean of the pixels a through h, calculated in consideration of the correlations between the defective pixel T and the pixels a through h.

[0051] Any defective pixels among the pixels a through h are not used to interpolate the defective pixel T. The exemplary embodiment of FIG. 11 is merely exemplary, and the number of pixels used to interpolate the defective pixel T is not particularly limited.

[0052] The image processor 160 performs a variety of digital processing, such as automatic white balance (AWB), contrast adjustment, gamma adjustment, and the like, on pixel data input thereto. Then, the image processor 160 allows the system processor 190 to control the sensor controller 112 and the lens controller 114 based on the result of the digital processing.

[0053] The sensor controller 112 controls a mechanical or electronic shutter provided in the image sensor 120 to adjust the degree of exposure. Also, the sensor controller 112 controls a flash dimming function by interworking with a flash (not illustrated). Also, the lens controller 114 controls the focus or zoom magnification of the photographing lens 110.

[0054] A still or moving image generated by the image processor 160 is stored in the image memory 165. The image display 170 is implemented as a liquid crystal display (LCD), an organic light-emitting diode (OLED) display, or a cathode ray tube (CRT) and displays the image stored in the image memory 165.

[0055] The system processor 190 is a controller controlling the entire apparatus 200, and programs that can be executed by the system processor 190 are stored in the system memory 195 and are sequentially read to be executed. A region for storing system information or user settings information is provided in the system memory 195 so that various information or settings can be read and restored later as necessary. The system memory 195 may be implemented as a RAM, a flash memory, an SSD, or the like.

[0056] FIG. 12 is a block diagram of the location information decoder 140. Referring to FIG. 12, the location information decoder 140 restores defective pixel location information from the encoded location information stored in the data memory 150 of FIG. 10. Specifically, the location information decoder 140 selects one of first and second decoding methods based on the distance between defective pixels and restores the defective pixel location information using the selected decoding method. Here, the distance between the defective pixels refers to the distance between defective pixels adjacent to each other in scan order (or the distance between the vertical coordinates of the defective pixels).

[0057] An entropy decoder 142 losslessly decodes the data format (i.e., the encoded location information) stored in the data memory 150. Here, a decoding algorithm corresponding to the encoding algorithm used in the entropy encoder 49 of the location information encoder 40 of FIG. 3, such as Huffman decoding, variable length decoding, or binary arithmetic decoding, may be used. If the entropy encoder 49 is not used in the location information encoder 40, the entropy decoder 142 may not be provided in the location information decoder 140.

[0058] A data format determiner 143 reads a number of bits (i.e., as many bits as the size of the "CODEWORD" field of FIG. 6) from the front of the data format to determine whether the data format is to be decoded by the first decoding method or the second decoding method. Since the front part of the data format 52 of FIG. 5 does not coincide with the front part the data format 54 of FIG. 6, a determination can be made as to whether the data format is encoded by the first encoding method or the second encoding method, at an encoder's end based on the bits read from the front of the data format.

[0059] The method in which the data format is decoded by the location information decoder 140 corresponds to the method in which the data format is encoded by the location information encoder 40. Since the method in which the data format is encoded by the location information encoder 40 is determined based on the distance between a current defective pixel and a defective pixel previous to the current defective pixel in scan order, the method in which the data format is decoded by the location information decoder 140 may be understood as being determined based on the distance between the previous and current defective pixels.

[0060] The data format reader 144 reads each of the fields of the data format according to the definition of the data format, as determined by the data format determiner 143. If the data format is determined to be decoded by the first decoding method, the data format reader 144 reads a field (i.e., a "DIFF_V" field) in which a difference value DIFF between a vertical coordinate Yn of the current defective pixel and a vertical coordinate Yn-1 of the previous defective previous is recorded and a field (i.e., an "H_POSITION" field) in which non-compressed horizontal coordinate Xn of the current defective pixel is recorded from the data format (see FIG. 5). The difference value DIFF read by the data format reader 144 is provided to an adder 148, and the horizontal coordinate Xn is provided to a data buffer 147.

[0061] The adder 148 restores the vertical coordinate Yn of the current defective pixel by adding up the difference value DIFF and the vertical coordinate Yn-1 of the previous defective pixel, temporarily stored in the data buffer 147. The restored vertical coordinate Yn is temporarily stored again in the data buffer 147. In this manner, the restored horizontal and vertical coordinates Xn and Yn are both stored in the data buffer 147 and are provided later to the pixel corrector 130 as location information.

[0062] On the other hand, if the data format is determined to be decoded by the second decoding method, the data format reader 144 reads a field (i.e., an "REF_LINE_INDEX" field) in which an index REX of a vertical sub-area where the vertical coordinate Yn of the current defective pixel belongs is recorded, a field (i.e., an "IND_DIFF_V" field) in which a local position IND of the vertical coordinate Yn of the current defective pixel in the sub-area where the vertical coordinate of the current defective pixel belongs is recorded, and a field (i.e., an "H_POSITION" field) in which the non-compressed horizontal coordinate Xn of the current defective pixel is recorded from the data format (see FIG. 6).

[0063] The index REF read by the data format reader 144, is provided to an initial position calculator 145, the local position IND of the vertical coordinate of the current defective pixel in the sub-area of the current defective pixel is provided to a vertical coordinate calculator 146, and the horizontal coordinate Xn is provided to the data buffer 147.

[0064] The initial position calculator 145 calculates the initial position (51 of FIG. 4) of the sub-area where the current defective pixel (e.g., the defective pixel P(n+1)) belongs (REF=k-1). Specifically, the initial position of the sub-area where the current defective pixel belongs may be calculated by multiplying the index REF of the vertical sub-area of the current defective pixel by vertical sub-area size V_comp. The vertical sub-area size V_comp may be provided by a sub-area size provider 141. The vertical sub-area size V_comp may be agreed upon in advance between the apparatus 100 and the apparatus 200, may be defined according to a predetermined set of rules, or may be provided by the apparatus 100.

[0065] The vertical coordinate calculator 146 restores the vertical coordinate Yn of the current defective pixel by adding up the initial position calculated by the initial position calculator 145 and the local position IND of the vertical coordinate Yn of the current defective pixel, and temporarily stores the restored vertical coordinate Yn in the data buffer 147. In this manner, the restored horizontal and vertical coordinates Xn and Yn are both stored in the data buffer 147 and are provided later to the pixel corrector 130 as location information.

[0066] Exemplary embodiments of the present disclosure in which different encoding (or decoding) methods are used depending on the distance between adjacent defective pixels in scan order have been described. The data format 54 of FIG. 6, unlike the data format 52 of FIG. 5, requires a larger number of bits than a conventional data format in which horizontal coordinate information and vertical coordinate information are simply juxtaposed with each other and may thus incur overhead of a large bit quantity in theory. However, when the data format 54 is selected, it means that there are not many defective pixels, in which case, such overhead may not be considerably problematic to the storage capacity of a data memory such as an SRAM. On the other hand, when there exists an issue with the storage capacity of the data memory, i.e., when the number of defective pixels rapidly increases, the data format 52 may be selected over the data format 54, in which case, the amount of data that needs to be stored can be considerably reduced.

[0067] Each of the elements illustrated in, or described above with reference to, FIGS. 1, 3, 10, and 12 may be implemented as a software component, such as a task, a class, a subroutine, a process, an object, an execution thread, or a program, as a hardware component, such as a field-programmable gate array (FPGA) or an application-specific integrated circuit (ASIC), or as the combination of the software component and the hardware component. These elements may be included in a computer-readable storage medium, or may be distributed over a plurality of computers.

[0068] A block of a flowchart may correspond to a module, a segment or a code containing one or more executable instructions implementing one or more logical functions, or to a part thereof. In some cases, functions described by blocks may be executed in an order different from the listed order. For example, two blocks listed in sequence may be executed at the same time or executed in reverse order.

[0069] FIG. 13 is a schematic view illustrating an example of how to encode and decode location information according to an exemplary embodiment of the present disclosure. Referring to FIG. 13, an image sensor panel 57 includes a total of 13 defective pixels, sub-area size V_comp is set to 32, and each coordinate pair contains the y and x coordinates of a defective pixel, in that order. Location information of the 13 defective pixels may be as shown in Table 1 below. Referring to Table 1 below, the coordinate pair of a fourth defective pixel is (50, 4090) in decimal. Since the y coordinate of the fourth defective pixel, i.e., 50, is "032" in hexadecimal and the x coordinate of the fourth defective pixel, i.e., 4090, is "ffa" in hexadecimal, location information of the fourth defective pixel may be represented as "0×032ffa".
[Table 1]
No.Coordinates (y, x)Data Format (Hexadecimal)
1 (0,1) 0×000001
2 (3,1200) 0×0034b0
3 (30, 2400) 0×01e960
4 (50, 4090) 0×032ffa
5 (90, 50) 0×05a032
6 (95, 3000) 0×05fbb8
7 (110, 3500) 0×06edac
8 (130, 1300) 0×082514
9 (150, 1000) 0×0963e8
10 (159, 1500) 0×09f5dc
11 (165, 900) 0×0a5384
12 (170, 1700) 0×0aa6a4
13 (190, 2700) 0×0bea8c


[0070] Meanwhile, the data format of location information according to an exemplary embodiment of the present disclosure for the image sensor panel 57 of FIG. 13 is as shown in Table 2 below.
[Table 2]
No.Coordinates (y, x)Data Format (Hexadecimal)
1 (0,1) 0×00001
2 (3,1200) 0×034b0
3 (30, 2400) 0×1b960
4 (50, 4090) 0×14ffa
5 (90, 50) 0×00C20-0×1A032
6 (95, 3000) 0×05bb8
7 (110, 3500) 0×0fdac
8 (130, 1300) 0×14514
9 (150, 1000) 0×143e8
10 (159, 1500) 0×095dc
11 (165, 900) 0×06384
12 (170, 1700) 0×056a4
13 (190, 2700) 0×14a8c


[0071] Referring to Table 2, the coordinate pair of the fourth defective pixel is (50, 4090), and the coordinate pair of a third defective pixel, which is previous to the fourth defective pixel, is (30, 2400). Accordingly, a difference value DIFF between the vertical coordinates of the third and fourth defective pixels is 20. Since the sub-area size V_comp of FIG. 13 is 32 and the difference value DIFF between the vertical coordinates of the third and fourth defective pixels is smaller than the sub-area size V_comp, the data format as illustrated in FIG. 5 is used for the fourth defective pixel. Specifically, the difference value DIFF of 20 is recorded in a "DIFF_V" field as a hexadecimal value of "14" and the x coordinate of the fourth defective pixel, i.e., 4090, is recorded in an "H_POSITION" field as a hexadecimal value of "ffa". Accordingly, the location information of the fourth defective pixel becomes "0×14ffa", which is the combination of the hexadecimal values of "14" and "ffa".

[0072] Referring again to Table 2, the coordinate pair of a fifth defective pixel is (90, 50), and the coordinate pair of the fourth defective pixel, which is previous to the fifth defective pixel, is (50, 4090), and a difference value DIFF between the vertical coordinates of the fourth and fifth defective pixels is 40. Since the difference value DIFF between the vertical coordinates of the fourth and fifth defective pixels is greater than the sub-area size V_comp, the data format as illustrated in FIG. 6 is used for the fifth defective pixel. "0×00C201A032" may be recorded in a data field. Here, "0×00C20" indicates both a codeword (recorded in a "CODEWORD" field) and an index REF (recorded in an "REF_LINE_INDEX" field), and of the rest of "0×00C201A032", "01A" indicates that an initial position IND (recorded in an "IND_DIFF_V" field) is 26, and "032" denotes an x coordinate value of 50 (recorded in an "H_POSITION" field). As already mentioned above, the index REF refers to the index REF of a sub-area where the y coordinate of a current defective pixel belongs, and the initial position IND refers to the local position of the y coordinate of the current defective pixel in the sub-area of the current defective pixel.

[0073] "0×00C20" can be represented as a (binary) bit string "0_0000_1100_0010_0000". Here, the total number of bits is not 20 but 17 and the zero at the forefront of the hexadecimal value is a bit that can only represent zero or one. All the data formats shown in Table 2 above have a bit size of 17, and the zero at the forefront of each of the data formats (i.e., the zero that follows "0×") is a (1-bit) binary number. In the above bit string, "0000100000" is a codeword, a 7-bit fixed value, and "0000100000" is a 10-bit index REF.

[0074] Here, when a maximum x coordinate value, i.e., the horizontal size of the image sensor panel 57 is 2952, as supported by existing high-resolution image sensors, the data format 52 of FIG. 5 and the data format 54 of FIG. 6 can be easily distinguished from each other, even without the need of signaling a flag, because a difference value DIFF and a codeword do not coincide with each other. The data format identifier 143 of FIG. 12 does not require any particular information to distinguish the two data formats, and sequentially reads the bit string input thereto.

[0075] First, the data format identifier 143 determines whether sixth and seventh bits of the input bit string are "11". If the sixth and seventh bits of the input bit string are "11", the data format identifier 143 can identify the input bit string as being a data format 54 as illustrated in FIG. 6. Specifically, if the input bit string is a data format 52 as illustrated in FIG. 5, the sixth through last bits of the input bit string correspond to an x coordinate, and the sixth and seventh bits of the input bit string can never be "11" because the maximum x coordinate value is 2592. That is, if the first two digits of a 12-digit bit string are "11", the 12-digit bit string is a minimum of 3072 and can never be an x coordinate value, in which case, the data format identifier 143 can easily identify the input bit string as being a data format 54 as illustrated in FIG. 6. Obviously, if at least one of the sixth and seventh bits of the input bit string is zero, the data format identifier 143 may identify the input bit string as being a data format 52 as illustrated in FIG. 5.

[0076] Meanwhile, the fifth defective pixel has an index REF of "0000100000", which corresponds to a value of 32 and represents a full coordinate of the initial position of the sub-area where the fifth defective pixel belongs, rather than the index REF of the sub-area where the fifth defective pixel belongs. Referring to FIG. 13, the full coordinate of the sub-area where the fifth defective pixel belongs is 64, but the coordinate of the sub-area where the fifth defective pixel belongs is recorded as 32, which is obtained by dividing 64 by 2 because the coordinate of each sub-area is always an even number. However, the initial position calculator 145 may calculate the initial position of the sub-area where the fifth defective pixel belongs as 64 by multiplying 32 by 2. In this modified exemplary embodiment, the initial position calculator 145 can calculate the initial position of the sub-area where the current defective pixel belongs simply based on the index REF provided by the data format reader 144 without the need to identify the sub-area size V_comp.

[0077] This modified exemplary embodiment is useful because the location information encoder 40 of FIG. 3 can temporarily store data formats for various sub-area sizes V_comp (e.g., 8, 16, 32, ...) and can encode defective pixel location information using one of the data formats that can produce a smallest data size. In this case, there is no need to signal the sub-area size V_comp because the sub-area size V_comp is set to vary and an appropriate data format corresponding to the sub-area size V_comp is simply transmitted to the location information decoder 140.

[0078] While exemplary embodiments are described above, it is not intended that these embodiments describe all possible forms of the inventive concept of the present disclosure. Rather, the words used in the specification are words of description rather than limitation, and it is understood that various changes may be made without departing from the spirit and scope of the inventive concept of the present disclosure. Additionally, the features of various implementing embodiments may be combined to form further exemplary embodiments of the present disclosure.


Claims

1. An apparatus for correcting defective pixels, comprising:

an image sensor converting light provided from a subject into an electrical signal and outputting image data;

a data memory storing encoded location information;

a location information decoder restoring defective pixel location information from the encoded location information stored in the data memory; and

a pixel corrector identifying one or more defective pixels from among pixels included in the image data using the defective pixel location information and interpolating the defective pixels using one or more neighboring pixels adjacent to each of the defective pixels,

wherein the location information decoder selects one of a first decoding method and a second decoding method, which is different from the first decoding method, based on a distance between the defective pixels and restores the defective pixel location information using the selected decoding method.


 
2. The apparatus of claim 1, wherein the distance between the defective pixels is a distance between defective pixels adjacent to each other in scan order.
 
3. The apparatus of claim 2, wherein the scan order is horizontal direction raster scan order.
 
4. The apparatus of claim 3, wherein the encoded location information includes, as a unit, non-compressed horizontal coordinate information and compressed vertical coordinate information.
 
5. The apparatus of claim 4, wherein the encoded location information includes an ordered pair of the compressed vertical coordinate information and the non-compressed horizontal coordinate information.
 
6.  The apparatus of claim 5, wherein the location information decoder selects one of the first and second decoding methods by reading a predetermined number of bits from the front of the encoded location information.
 
7. The apparatus of claim 6, wherein the location information decoder reads, in accordance with the first decoding method, a field in which a difference value between a vertical coordinate of a current defective pixel and a vertical coordinate of a defective pixel previous to the current defective pixel in the scan order is recorded and a field in which the non-compressed horizontal coordinate information is recorded.
 
8. The apparatus of claim 7, wherein the location information decoder restores the vertical coordinate of the current defective pixel by adding up the vertical coordinate of the previous defective pixel and the difference value.
 
9. The apparatus of claim 6, wherein the location information decoder reads a field in which an index of a vertical sub-area where the vertical coordinate of a current defective pixel belongs is recorded, a field in which a local position of the vertical coordinate of the current defective pixel in the vertical sub-area of the current defective pixel is recorded, and a field in which the non-compressed horizontal coordinate information is recorded.
 
10. The apparatus of claim 9, wherein the location information decoder restores the vertical coordinate of the current defective pixel by multiplying an index of the vertical sub-area of the current defective pixel by vertical sub-area size and adding up the result of the multiplication and the local position of the vertical coordinate of the current defective pixel.
 
11. The apparatus of claim 1, wherein the pixel corrector interpolates each of the defective pixels by calculating a mean, a weighted mean, or a median of the neighboring pixels.
 
12. The apparatus of claim 11, wherein the pixel corrector interpolates each of the defective pixels using all the neighboring pixels except for any defective pixels.
 
13. An apparatus for recording defective pixel information, comprising:

an image sensor;

a pixel inspector identifying one or more defective pixels from among pixels included in the image sensor and providing location information of the defective pixels;

a location information encoder generating encoded location information by encoding the location information provided by the pixel inspector; and

a data memory storing the encoded location information,

wherein the location information encoder selects one of a first encoding method and a second encoding method, which is different from the first encoding method, based on a distance between the defective pixels and encodes the location information provided by the pixel inspector using the selected encoding method.


 
14. The apparatus of claim 13, wherein the distance between the defective pixels is a distance between defective pixels adjacent to each other in scan order.
 
15. The apparatus of claim 14, wherein the scan order is horizontal direction raster scan order.
 
16. The apparatus of claim 15, wherein the encoded location information includes, as a unit, non-compressed horizontal coordinate information and compressed vertical coordinate information in that order.
 
17. The apparatus of claim 16, the location information encoder selects one of the first and second encoding methods by comparing a difference value between a vertical coordinate of a current defective pixel and a defective pixel previous to the current defective pixel in the scan order and vertical sub-area size.
 
18. The apparatus of claim 17, wherein if the difference value is smaller than the vertical sub-area size, the location information encoder generates a field in which the difference value is recorded and a field in which the non-compressed horizontal coordinate information is recorded as the encoded location information in accordance with the first encoding method.
 
19.  The apparatus of claim 17, wherein if the difference value is greater than the vertical sub-area size, the location information encoder generates a field in which an index of a vertical sub-area where the vertical coordinate of the current defective pixel belongs is recorded, a field in which a local position of the vertical coordinate of the current defective pixel in the vertical sub-area of the current defective pixel is recorded, and a field in which the non-compressed horizontal coordinate information is recorded as the encoded location information in accordance with the second encoding method.
 
20. A method of correcting defective pixels, comprising:

converting light provided from a subject into an electrical signal with the use of an image sensor and outputting image data;

reading encoded location information from a data memory;

restoring defective pixel location information from the encoded location information;

identifying one or more defective pixels from among pixels included in the image data using the defective pixel location information; and

interpolating the defective pixels using one or more neighboring pixels adjacent to each of the defective pixels,

wherein the restoring the defective pixel location information, comprises selecting one of a first decoding method and a second decoding method, which is different from the first decoding method, based on a distance between the defective pixels and restoring the defective pixel location information using the selected decoding method.


 




Drawing








































Search report