Coding device

Abstract

To provide a coding device, a decoding device, a coding method, a decoding method, a coding program, and a decoding program, capable of suppressing a dark part coding distortion by a gray level correction.SOLUTION: A coding device 200 includes: a gray scale correction part 201 that corrects a gray scale of a RAW image data from an imaging element having an optical black on the basis of a gamma coefficient and a value of the optical black; and a coding part 202 that codes a gray scale correction RAW image data where a gray scale is corrected by the gray scale correction part. A decoding device includes: an acquisition part that acquires a coded RAW image data obtained by coding the gray scale correction RAW image data obtained by correcting the gray scale on the basis of the gamma coefficient and the value of the optical black; a decoding part that decodes the coded RAW image data acquired by the acquisition part to the gray scale correction RAW image data and a gray scale reverse correction part that reversely corrects the gray scale of the gray scale correction RAW image data decoded by the decoding part on the basis of the gamma coefficient and the value of the optical black, and outputs the RAW image data before the correction of the gray scale.SELECTED DRAWING: Figure 2

Description

Incorporation by Reference

[0001] This application is a Japanese application filed on January 16, 2018. This application claims priority from US Pat. No. 6,313,323, filed on Nov. 18, 2002, the contents of which are incorporated herein by reference. [Technical field]

[0002] The present invention relates to an encoding device, a decoding device, an encoding method, a decoding method, an encoding program, and Regarding the decryption program. [Background technology]

[0003] A technology for performing gradation correction on RAW image data has been disclosed. When the optical black of the output from the filter is not zero, the above-mentioned √γ etc. are used to suppress the dark coding distortion. When the image is encoded with gradation correction by The degree of suppression of dark coding distortion is reduced. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2006-270478 A Summary of the Invention

[0005] The encoding device according to one aspect of the invention disclosed in the present application includes an image sensor having optical black. RAW image data from the camera is corrected based on a gamma coefficient and the optical black value. A code for encoding the gradation-corrected RAW image data that has been gradation-corrected by the correction unit. and a chemical portion.

[0006] The decoding device according to one aspect of the invention disclosed in this application is a device for decoding a gamma coefficient, an optical black value, and The gradation-corrected RAW image data, which has been gradation-corrected based on the an acquisition unit for acquiring the encoded RAW image data; a decoding unit for decoding the gradation-corrected raw image data; The corrected RAW image data is subjected to inverse gradation correction based on the gamma coefficient and the optical black value. and an inverse correction unit that outputs RAW image data before gradation correction.

[0007] The encoding method according to one aspect of the invention disclosed in the present application is directed to an image sensor having optical black. RAW image data from the camera is corrected based on a gamma coefficient and the optical black value. The gradation-corrected RAW image data that has been gradation-corrected by the correction process is encoded. The encoding process is then performed.

[0008] The decoding method according to one aspect of the invention disclosed in this application is a method for decoding a gamma coefficient and an optical black value. The gradation-corrected RAW image data, which has been gradation-corrected based on the and a process of acquiring the encoded raw image data. a decoding process for decoding the gradation corrected RAW image data; The gradation corrected RAW image data is gradated based on the gamma coefficient and the optical black value. Then, an inverse correction process is performed to output the RAW image data before the tone correction. [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 is a block diagram illustrating an example of a hardware configuration of an information processing device. [Diagram 2] FIG. 2 is a block diagram of an example of a functional configuration of the encoding device according to the first embodiment. [Diagram 3] FIG. 3 is an explanatory diagram showing an example of the data structure of encoded tone-corrected RAW image data. [Figure 4] FIG. 4 is a block diagram illustrating an example of the configuration of the encoding unit. [Diagram 5] FIG. 5 is a flowchart illustrating an example of a procedure of the encoding process performed by the encoding device. [Figure 6] FIG. 6 is a block diagram illustrating an example of a functional configuration of a decoding device. [Figure 7] FIG. 7 is a block diagram illustrating an example of the configuration of the decoding unit. [Figure 8] FIG. 8 is a flowchart illustrating an example of a decoding process procedure performed by the decoding device. [Figure 9] FIG. 9 is a graph showing the input / output characteristics of tone correction when no optical black value is used. [Figure 10] FIG. 10 is a graph showing the gain characteristics of tone correction when no optical black value is used. [Figure 11] FIG. 11 is a graph showing the input / output characteristics of the inverse tone correction when the optical black value is not used. [Figure 12] FIG. 12 is a graph showing input / output characteristics 1 of tone correction when the optical black value is used. [Figure 13] FIG. 13 is a graph showing the gain characteristic in input / output characteristic 1 of tone correction when the optical black value is used. [Figure 14] FIG. 14 is a graph showing input / output characteristics 1 of the gradation inverse correction when the optical black value is used. [Figure 15] FIG. 15 is a graph showing input / output characteristics 2 of tone correction when the optical black value is used. [Figure 16] FIG. 16 is a graph showing gain characteristics in input / output characteristics 2 of tone correction when an optical black value is used. [Figure 17] FIG. 17 is a graph showing input / output characteristics 2 of the gradation inverse correction when the optical black value is used. [Figure 18] FIG. 18 is a graph showing input / output characteristics 3 of tone correction when the optical black value is used. [Figure 19]FIG. 19 is a graph showing the gain characteristic in the input / output characteristic 3 of the tone correction when the optical black value is used. [Figure 20] FIG. 20 is a graph showing input / output characteristics 3 of the inverse tone correction when the optical black value is used. [Figure 21] FIG. 21 is a graph showing input / output characteristics 4 of tone correction when the optical black value is used. [Figure 22] FIG. 22 is a graph showing the gain characteristic in the input / output characteristic 4 of the tone correction when the optical black value is used. [Figure 23] FIG. 23 is a graph showing input / output characteristics 4 of the inverse tone correction when the optical black value is used. [Figure 24] FIG. 24 is a graph showing input / output characteristics 5 of tone correction when the optical black value is used. [Diagram 25] FIG. 25 is a graph showing the gain characteristic in the input / output characteristic 5 of the tone correction when the optical black value is used. [Figure 26] FIG. 26 is a graph showing input / output characteristics 5 of the gradation inverse correction when the optical black value is used. [Figure 27] FIG. 27 is a graph showing input / output characteristics 6 of tone correction when the optical black value is used. [Figure 28] FIG. 28 is a graph showing the gain characteristic in the input / output characteristic 6 of the tone correction when the optical black value is used. [Figure 29] FIG. 29 is a graph showing input / output characteristics 6 of the inverse tone correction when the optical black value is used. [Diagram 30] FIG. 30 is a graph showing input / output characteristics 7 of tone correction when the optical black value is used. [Diagram 31] FIG. 31 is a graph showing the gain characteristic in the input / output characteristic 7 of the tone correction when the optical black value is used. [Diagram 32] FIG. 32 is a graph showing input / output characteristics 7 of the inverse tone correction when the optical black value is used. [Diagram 33] FIG. 33 is a graph showing input / output characteristics 8 of tone correction when the optical black value is used. [Diagram 34] FIG. 34 is a graph showing the gain characteristic in the input / output characteristic 8 of the tone correction when the optical black value is used. [Diagram 35] FIG. 35 is a graph showing input / output characteristics 8 of the inverse tone correction when the optical black value is used. [Diagram 36] FIG. 36 is a graph showing input / output characteristics 9 of tone correction when the optical black value is used. [Figure 37] FIG. 37 is a graph showing the gain characteristic in the input / output characteristic 9 of the tone correction when the optical black value is used. [Figure 38] FIG. 38 is a graph showing input / output characteristics 9 of the inverse tone correction when the optical black value is used. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS EXAMPLES

[0010] <Example of hardware configuration of information processing device> FIG. 1 is a block diagram showing an example of a hardware configuration of an information processing device. 00 is a device that includes either an encoding device or a decoding device, or both. The processing device 100 is, for example, an imaging device such as a digital camera or a digital video camera. It can be a personal computer, tablet, smartphone, or game console. stomach.

[0011] The information processing device 100 includes a processor 101, a storage device 102, and an operation device 104. 03, LSI (Large Scale Integration) 104, and imaging unit The device 105 and the communication IF (Interface) 106 are connected to the bus. The processor 101 controls the information processing device 100. The storage device 102 serves as a working area for the processor 101 .

[0012] The storage device 102 is a non-transitory or temporary storage device for storing various programs and data. The storage device 102 is, for example, a ROM (Read Only Memory), RAM (Random Access Memory), HDD (H The operation device 103 is a hard disk drive (HDD) and a flash memory. The operation device 103 may be, for example, a button, a switch, a touch panel, or the like. There is a flannel.

[0013] The LSI104 supports color interpolation, white balance adjustment, edge emphasis, gamma correction, and gradation conversion. An integrated circuit that performs specific processes such as image processing, encoding, decoding, compression and decompression It is a road.

[0014] The imaging unit 105 captures an image of a subject and generates RAW image data. 105 includes an imaging optical system 151, an imaging element 153 having a color filter 152, and a signal and a processing circuit 154.

[0015] The imaging optical system 151 includes, for example, a plurality of lenses including a zoom lens and a focus lens. For simplicity, the imaging optical system 151 is illustrated as a single lens in FIG. do.

[0016] The image sensor 153 captures (photographs) an image of a subject formed by a light beam passing through the imaging optical system 151. The image sensor 153 is a solid-state image sensor of a progressive scanning type (for example, C CD (Charge Coupled Device) image sensor) , XY address type solid-state image sensor (for example, CMOS (Complementary Metal Oxide Semiconductor) Metal Oxide Semiconductor (MIS) image sensors good.

[0017] A pixel group 160 having a photoelectric conversion unit is arranged in a matrix on the light receiving surface of the image sensor 153. Each pixel of the image sensor 153 transmits light of a different color component. A plurality of types of color filters 152 are arranged according to a predetermined color arrangement. Each pixel of the image sensor 153 corresponds to each color component by color separation in the color filter 152. The sensor outputs an electrical signal.

[0018] In the first embodiment, for example, red (R), green (G), and blue (B) color filters 15 2 are periodically arranged on the light receiving surface according to a Bayer array of 2 rows and 2 columns. The odd-numbered rows of the color array of the image sensor 153 have G and B pixels arranged alternately, and the even-numbered rows of the color array have The rows alternate between R and G pixels, and the green pixels form a checkerboard pattern across the entire color array. As a result, the image sensor 153 captures color RAW images during shooting. Image data can be acquired.

[0019] In addition, the pixel group 160 arranged in a two-dimensional array in the image sensor 153 is called effective pixels. The effective pixel area 161 is made up of an optical black pixel area 162. The signal charge generated by actually receiving light and photoelectric conversion is amplified and sent to the signal processing circuit 154. This is the pixel area to be output.

[0020] The optical black pixel area 162 is a pixel area for outputting optical black, which is a reference for the black level. The optical black pixel area 162 is provided, for example, on the outer periphery of the effective pixel area 161. The optical black pixel area 162 is a pixel area that is obtained by subtracting, for example, thermal noise generated in the image sensor 153. It is provided to correct.

[0021] The signal processing circuit 154 converts the image signal input from the image sensor 153 into an analog signal. signal processing (correlated double sampling, black level correction, etc.), A / D conversion processing, and digital signal The RA output from the signal processing circuit 154 is sequentially processed. The W image data is input to the LSI 104 or the storage device 102. connects to an external device via a network to transmit and receive data.

[0022] <Example of functional configuration of encoding device> FIG. 2 is a block diagram of an example of a functional configuration of the encoding device according to the first embodiment. The device 200 includes a tone correction unit 201, an encoding unit 202, a recording unit 203, and a setting unit 204. The image processing apparatus includes a tone correction unit 201, an encoding unit 202, a recording unit 203, and a setting unit 204. Specifically, for example, a program stored in the storage device 102 is executed by the processor 1 01 or by the LSI 104.

[0023] The gradation correction unit 201 performs gradation correction on the RAW image data. Light from an image sensor 153 that photoelectrically converts light from a subject obtained through a filter 152 In other words, RAW image data is gamma-corrected. Image processing such as correction, demosaicing, white balance adjustment, and color conversion processing are performed. The amount of data in RAW image data is determined by the image processing and encoding described above. The image data is larger than the data that has been subjected to the image processing (also called compression processing). The data may be a still image or one frame of a moving image.

[0024] The gradation correction unit 201 corrects a RAW image from an image sensor 153 having an optical black pixel area 162. The image data is gradation corrected based on the gamma coefficient value (gamma value) and the optical black value. For example, the gradation correction unit 201 may correct the RAW image data using the optical black value as a reference. The tone correction unit 201 performs tone correction using noise reduction. The details of the gradation correction by the gradation correction unit 201 will be described later. State.

[0025] The encoding unit 202 encodes the RAW image data (hereinafter referred to as , tone-corrected RAW image data) is encoded, and the encoded tone-corrected RAW image data (hereinafter The encoding unit 202 outputs the tone-corrected RAW image data. If the image data is a still image, the gradation-corrected RAW image data is encoded in the frame. Encode as an I picture.

[0026] Furthermore, when the tone-corrected RAW image data is one frame of a moving image, the encoding unit 202 By using intraframe predictive coding, the gradation-corrected RAW image data is converted into an I-picture, or The gradation-corrected RAW image is generated by interframe predictive coding using the gradation-corrected RAW image data. The data is encoded into a P picture or a B picture.

[0027] When the optical black value output from the image sensor 153 is not zero, the dark coding distortion suppression Therefore, if gradation correction is simply performed using √γ etc., the gradation correction will be The degree of suppression of the dark coding distortion caused by the optical black value is reduced. Gradation correction such as gamma is performed, and the header information of the encoded gradation-corrected RAW image data is In addition to the gamma value used during production, an optical black value is added. This allows the dark area code to be corrected by gradation correction. This eliminates the decrease in the degree of suppression of chemical distortion.

[0028] The recording unit 203 records the encoded tone-corrected RAW image data in the storage device 102 . The setting unit 204 sets the gamma coefficient value, the calculated temporary gain, and the exposure value, and sends the gamma coefficient value, the calculated temporary gain value, and the exposure value to the tone correction unit 201. The details of the setting unit 204 will be described later in the third and fourth embodiments.

[0029] <Example of data structure of encoded tone-corrected RAW image data> FIG. 3 is an explanatory diagram showing an example of the data structure of encoded tone-corrected RAW image data. The tone-corrected RAW image data 300 includes header information 301 and an encoded data string 302. The header information 301 is information added by the encoding unit 202. 1 includes image format information 311 and control information 312. The encoded data sequence 302 includes gradation correction This is a data string obtained by encoding the correct RAW image data.

[0030] Each element of the header information 301 will be specifically described. The image format information 311 is Size of previous tone-corrected RAW image data, size of encoded tone-corrected RAW image data, color Identification information that specifies the array pattern, including the number of pixels in the gradation-corrected RAW image data before encoding nothing.

[0031] The control information 312 is the type of encoded tone-corrected RAW image data (I picture, P picture, etc. , B picture), identification information of the reference frame, and the gamut when gradation correction is applied. The black level reference is the optical black level 322, the calculated temporary gain 323, and the first position. The first porch width 324, the second porch width 325, and the tone correction identifier 326.

[0032] The calculated provisional gain 323 will be described later in the third and fourth embodiments. The second porch width 325 and the second porch width 24 will be described later in the fourth embodiment. is an identifier for identifying an algorithm for performing gradation correction. Therefore, the decoding device can determine by which algorithm the encoded tone-corrected RAW image data 300 is encoded. It is possible to specify whether tone correction has been performed.

[0033] <Example of configuration of the encoding unit 202> 4 is a block diagram showing an example of the configuration of the encoding unit 202. The encoding unit 202 A multiplication unit 401, a subtraction unit 402, an orthogonal transformation unit 403, a quantization unit 404, and a variable-length coding unit a second storage unit 405, a second quantization unit 406, a second orthogonal transformation unit 407, an addition unit 408, and a second storage unit The image processing unit 409 includes a motion detection unit 410 and a motion compensation unit 411.

[0034] The first storage unit 401 stores the tone-corrected RAW image data output from the tone correction unit 201. The tone-corrected RAW image data stored in the first storage unit 401 is stored as an image to be encoded. The data is output to the subtraction unit 402 in the order of input. The W image data is erased from the first storage unit 401 in sequence.

[0035] When generating a P picture or a B picture, the subtraction unit 402 subtracts the input original image (tone-corrected RAW image data) component frame and The subtraction unit 402 outputs a difference signal (prediction error value) between the predicted value and the input signal. When generating a frame, the component frames of the input original image are output as is.

[0036] When generating an I-picture, the orthogonal transformation unit 403 bypasses the subtraction unit 402 and The orthogonal transform unit 403 performs an orthogonal transform on the original image. When generating a picture, the above-mentioned difference signal is subjected to an orthogonal transformation.

[0037] The quantization unit 404 quantizes the frequency coefficients (orthogonal coefficients) input from the orthogonal transformation unit 403 in units of blocks. The output of the quantization unit 404 is fed to a variable length coding unit 405 and and the inverse quantization unit 406, respectively.

[0038] The variable length coding unit 405 variably encodes the quantization coefficients and the motion vectors from the motion detection unit 410. Length-encoded and tone-corrected RAW image data (I-picture, P-picture, B-picture) Output.

[0039] The inverse quantization unit 406 inversely quantizes the quantized coefficients in units of blocks, which are the units of coding, to obtain a circumferential The inverse orthogonal transform unit 407 decodes the frequency coefficients decoded by the inverse quantization unit 406. is then inversely orthogonally transformed to decode the prediction error value (or the original image).

[0040] The adder 408 adds the decoded prediction error value and a prediction Then, the decoded value of the picture (reference frame ) is stored in the second storage unit 409. Note that the reference frames that are not referenced in the subsequent motion compensation prediction are stored in the second storage unit 409. The programs are sequentially erased from the second storage unit 409 .

[0041] The motion detection unit 410 uses the reference frame in the second storage unit 409 to perform gradation correction of the encoding target. A motion vector for predicting the correct RAW image data is detected. The motion vector is a motion vector. The signal is output to the compensation unit 411 and the variable-length coding unit 405 .

[0042] The motion compensation unit 411 performs the motion compensation for the gradation of the encoding target based on the motion vector and the reference frame. The predicted value is output by predicting the corrected RAW image data on a block-by-block basis. This predicted value is subtracted. The signal is output to the signal processing unit 402 and the adding unit 408.

[0043] When motion compensation prediction is performed for a certain block, the gradation-corrected RAW image to be coded is If the image data exactly matches the predicted value, only the motion vector is coded. If the RAW image data partially matches the predicted value, the motion vector and the difference image are In addition, the gradation-corrected raw image data to be encoded is completely different from the predicted value. If a block is coded, the entire image is coded.

[0044] <Example of encoding process> FIG. 5 is a flowchart showing an example of the encoding process procedure performed by the encoding device 200. The imager 200 converts a RAW image output from an image sensor 153 having an optical black pixel region 162 into Input of image data is accepted (step S501), and the gradation correction unit 201 performs gradation correction. The RAW image data is subjected to tone correction using an algorithm (step S502).

[0045] Next, the encoding device 200 encodes the tone-corrected RAW image data by the encoding unit 202. Then, the encoding device 200 encodes the encoded data by the recording unit 203 (step S503). The encoded tone-corrected RAW image data 300 is stored in the storage device 102 (step S50 4).

[0046] <Example of functional configuration of a decoding device> FIG. 6 is a block diagram showing an example of a functional configuration of a decoding device. The decoding device 600 includes an acquisition unit The image processing device includes an acquisition unit 601, a decoding unit 602, and a gradation inverse correction unit 603. 02, and the gradation inverse correction unit 603 specifically stores, for example, The processor 101 executes the program, or the LSI 104 executes the program. This function is realized by

[0047] The acquisition unit 601 acquires the encoded tone-corrected RAW image data encoded by the encoding device 200. If the decoding device 600 can communicate with the encoding device 200, The unit 601 receives the encoded tone-corrected RAW image data 30 transmitted from the encoding device 200. 0 is received. If the decoding device 600 is implemented in the same device as the encoding device 200, The acquisition unit 601 acquires the encoded tone-corrected RAW image data 30 stored in the storage device 102. Reads 0.

[0048] The decoding unit 602 decodes the encoded tone-corrected RAW image data 300 using the control information 312. Specifically, for example, the decoding unit decodes the encoded tone correction data into tone correction RAW image data. Depending on the type of RAW image data 300 (I picture, P picture, B picture), A frame is identified, and the encoded tone-corrected RAW image data 300 is converted into tone-corrected RAW image data. Decrypt it to:

[0049] The gradation inverse correction unit 603 converts the gradation-corrected RAW image data decoded by the decoding unit 602 into The gradation is inversely corrected and the image data is restored to the RAW image data before gradation correction. The inverse correction unit 603 refers to the gradation correction identifier 326 of the control information 312 and converts the RAW image data The gradation correction algorithm applied to the data is identified, and a corresponding gradation correction algorithm is The gradation inverse correction algorithm is executed. The gradation inverse correction unit 603 performs the gradation inverse correction. A reduction process (NR process) may be performed.

[0050] <Example of configuration of the decoding unit 602> 7 is a block diagram showing an example of the configuration of the decoding unit 602. The decoding unit 602 is a variable-length code A decoding unit 701, an inverse quantization unit 702, an inverse orthogonal transformation unit 703, an addition unit 704, and a third accumulation unit It has an accumulation unit 705 and a motion compensation unit 706 .

[0051] The variable-length code decoding unit 701 decodes the input coded tone-corrected RAW image data 300. The decoded quantization coefficients are then sent to the inverse quantization unit, where they are then output as quantization coefficients and motion vectors. The decoded motion vector input to the motion compensation unit 702 is input to the motion compensation unit 706 .

[0052] The inverse quantization unit 702 inversely quantizes the quantized coefficients on a block-by-block basis to decode the frequency coefficients. The inverse orthogonal transform unit 703 performs inverse orthogonal transform on the frequency coefficients decoded by the inverse quantization unit 702 to obtain a prediction coefficient. The measured error value (or the original image signal) is decoded.

[0053] The adder 704 adds the decoded prediction error value and the prediction value generated by the motion compensation unit 706. The adder 704 outputs the decoded image data in units of blocks. The image data output from the is output as gradation-corrected RAW image data, and The data is input to the storage unit 705 .

[0054] The third storage unit 705 stores the decoded values ​​of the image as a reference frame. Image data that is not referenced in the compensation prediction is sequentially deleted from the third storage unit 705. 706, a block of a decoding target image is decoded based on a motion vector and a reference frame. The predicted value is output to the adder 704 .

[0055] <Example of decryption process> 8 is a flowchart showing an example of a procedure of a decoding process performed by the decoding device 600. 600 acquires the encoded tone-corrected RAW image data 300 by an acquisition unit 601 (step Step S801), the decoding unit 602 converts the encoded tone-corrected RAW image data 300 into tone-corrected RAW image data 300. The decoding device 600 then decodes the data into the correct RAW image data (step S2102). The decoded gradation-corrected RAW image data is inversely gradated by the inverse gradation correction unit 603 to obtain RA The W image data is restored (step S803).

[0056] <Specific examples of tone correction and inverse tone correction> Next, the gradation correction by the gradation correction unit 201 and the gradation inverse correction unit 603 according to the first embodiment A specific example of the gradation inverse correction by the above will be described with reference to FIGS. 9 to 17. For comparison, the gradation correction and gradation inverse correction without using the optical black value are shown in the graphs of Figs. Gradation correction 1 and gradation inverse correction 1 of the first embodiment are shown in FIG. Correction 2 and gradation inverse correction 2 are shown in FIGS.

[0057] (Characteristics of tone correction when optical black value is not used) FIG. 9 is a graph showing the input / output characteristics of tone correction when no optical black value is used. is the signal level of the RAW image data from the image sensor 153 (input signal level E), The vertical axis represents the signal level (output signal level E') of the tone-corrected RAW image data. The same applies to Figs. 12, 15, 18, 21, and 24. However, the input signal level in Fig. 9 The signal E is a signal of RAW image data from an image sensor 153 that does not have an optical black pixel area 162. This is at the issue level.

[0058] The signal level is, for example, a voltage value indicating the luminance of a pixel in the effective pixel area 161. The signal level is normalized in the range of 0.0 to 1.0. A pixel with a low signal level is a dark pixel, and a pixel with a low signal level is a A pixel with a high level indicates a bright pixel. Also, OB (Optical Black) is the optical black. This is the target value.

[0059] The gradation correction algorithm shown in the input / output characteristic graph 900 of FIG. 9 is expressed by the following formula (1): can be.

[0060] E´=OETF[E]=E 1 / γ ...(1)

[0061] where E is the signal level of the RAW image data (input signal level), and E' is the gradation correction RA W is the signal level of the image data (output signal level), OETF[] is the gradation correction function, γ is the Gunn It is the value.

[0062] In the input / output characteristic graph 900, when the input signal level is from 0.0 to the optical black value OB, Gradation correction is performed for the section SC9. In order to encode the dark and light areas equally, gamma correction is applied in the development stage, which emphasizes the dark areas and reduces the coding. The higher the compression ratio of the encoding, the more noticeable the encoding distortion becomes.

[0063] FIG. 10 is a graph showing the gain characteristics of gradation correction when no optical black value is used. The gain of gradation correction when the optical black value is not used is the enhancement degree of the input signal level E. The input characteristic graph 1000 is expressed by the following formula (2) by differentiating formula (1).

[0064] G = {OETF[E]}' = (1 / γ) × E {(1 / γ)-1} ...(2)

[0065] G is the gain of the input signal level E. In the gain characteristic graph 1000, The gain G value in the section SC9 where the signal level E is from 0.0 to the optical black value OB is set to the optical black value O. The input signal level E, which is invalid after development processing, is larger than the gain G of B. The gain of the pixels in the section SC9 from 0.0 to the optical black value OB is large. After processing, the gain in the area above the optical black value OB that is effective as an image becomes smaller, and the coding The distortion suppression effect is reduced.

[0066] FIG. 11 is a graph showing the input / output characteristics of the inverse tone correction when the optical black value is not used. The horizontal axis is the signal level of the gradation-corrected RAW image data (input signal level E'), and the vertical axis is , the signal level of the RAW image data that has been inversely corrected from the gradation corrected RAW image data (output signal level E).

[0067] The same applies to Figs. 14, 17, 20, 23, and 26. However, the output of Fig. 11 The output signal level E is the level of the image without the optical black pixel area 162, restored by the gradation inverse correction. The signal level is the signal level of the RAW image data from the image sensor 153. Similarly, for example, the voltage value indicating the luminance of the pixel in the effective pixel area 161 is 0.0 to 1 0.0. Pixels with low signal levels are dark pixels, and pixels with high signal levels are dark pixels. A pixel indicates a bright pixel.

[0068] The gradation inverse correction algorithm shown in the input / output characteristic graph 1100 of FIG. 10 is expressed by the following formula (3): be expressed.

[0069] E=EOTF[E´]=E γ ...(3)

[0070] where E' is the signal level of the gradation-corrected RAW image data (input signal level), and E is the restored The signal level (output signal level) of the RAW image data that is output. EOTF[] is the inverse gradation correction function. where γ is the gamma value.

[0071] (Example 1 of tone correction characteristics when using optical black value) Next, referring to Figs. 12 to 14, a characteristic example 1 of gradation correction when an optical black value is used will be described. I will explain this in more detail.

[0072] FIG. 12 is a graph showing input / output characteristics 1 of gradation correction when an optical black value is used. In the output characteristic graph 1200, the gradation correction unit 201 detects an input signal lower than the optical black value OB. No gradation correction is performed for level E. In other words, the input / output characteristic graph 120 in FIG. In the case of 0, the output signal level is in the section SC9 where the input signal level E is from 0.0 to the optical black value OB. The input signal level E' is 0.0, the input signal level E rises from the optical black value OB, and the input signal level When the level E is 1.0, the output signal level E' is at its maximum value.

[0073] The gradation correction algorithm shown in the input / output characteristic graph 1200 of FIG. 12 is expressed by the following formula (4): It will be revealed.

[0074]

number

[0075] where E is the signal level of the RAW image data (input signal level), and E' is the gradation correction RA W is the signal level of the image data (output signal level), OETF[] is the tone correction function, and OB is the optical The gradation correction unit 201 uses the gradation correction algorithm of the following equation (4): The encoding unit 202 outputs a gradation correction identifier 326 indicating the gradation correction unit The gradation correction identifier 326 from 201 is added as the control information 312 in the header information 301. do.

[0076] FIG. 13 shows the gain characteristic in the input / output characteristic 1 of the gradation correction when the optical black value is used. This is a graph showing the gain of the input signal level E in tone correction when the optical black value OB is used. The in characteristic graph 1300 is expressed by the following formula (5) by differentiating formula (4).

[0077]

number

[0078] G is the gain of the input signal level E. In the gain characteristic graph 1300, Since the gain G value in the range from 0.0 to the optical black value OB is 0.0, The input signal level E in the section SC9 from 0.0, which is the dark area, to the optical black value OB is not emphasized. Also, since the gain G value reaches its maximum value at the optical black value OB, the image will be By increasing the gain in the section above the effective optical black value OB, the coding distortion suppression effect is improved. It can be made bigger.

[0079] When developing or adjusting an image, it is necessary to make the dark areas brighter in order to make the image easier to see. In this case, the distortion of the dark area occurring between the encoding unit 202 and the decoding unit 602 becomes noticeable. By increasing the gain G of the dark area by tone correction, the encoding unit 202 and the decoding unit 602 This suppresses distortion in the dark areas that occurs between the Even if the dark areas are made brighter to make the pattern easier to see, distortion in the dark areas becomes less noticeable.

[0080] FIG. 14 is a graph showing input / output characteristics 1 of the gradation inverse correction when the optical black value is used. The gradation inverse correction algorithm shown in the input / output characteristic graph 1400 of FIG. 14 is expressed by the following formula (6): It will be revealed.

[0081] E = EOTF [E'] = (E') γ +OB····(6)

[0082] where E' is the signal level of the gradation-corrected RAW image data (input signal level), and E is the restored The signal level (output signal level) of the RAW image data that is output. EOTF[] is the inverse gradation correction function. The gradation inverse correction unit 603 adds a gradation correction identifier 32 indicating the formula (4) to the control information 312. If 6 is detected, the gradation inverse correction of equation (6) is performed.

[0083] In the input / output characteristic graph 1400, the output signal level in the restored RAW image data is Since the minimum value of the black level E is the optical black value OB, the effective optical black value O By increasing the gain in the section above B, the coding distortion suppression effect can be increased. Cut.

[0084] (Example 2 of tone correction characteristics when using optical black value) Next, using FIG. 15 to FIG. 17, a second example of the characteristics of the gradation correction when the optical black value is used will be described. In the characteristic example 1 of the gradation correction when the optical black value OB is used, the input / output of FIG. In the characteristic graph 1200, the output signal level E′ is restored in the bright section SC12. In the characteristic example 2 of the gradation correction when the optical black value OB is used, As shown in Example 1 of the gradation correction characteristics when B is used, the reproduction of the original black is improved and the black floating is suppressed. While controlling this, we aim to improve the reproducibility of bright areas.

[0085] FIG. 15 is a graph showing input / output characteristics 2 of gradation correction when an optical black value is used. In Example 2 of the input / output characteristics of gradation correction when the optical black value OB is used, As in the case of the input / output characteristic example 1 of the gradation correction, the gradation correction unit 201 outputs a black value lower than the optical black value OB. No tone correction is performed for input signal level E.

[0086] In other words, in the input / output characteristic graph 1500 of FIG. 15, when the input signal level E is from 0.0 In the section SC9 up to the optical black value, the output signal level E' is 0.0, and the input signal level E is Rising from the optical black value OB, input signal level E is 1.0 and output signal level E' is the upper limit Take 1.0.

[0087] The gradation correction algorithm shown in the input / output characteristic graph 1500 of FIG. 15 is expressed by the following formula (7): It will be revealed.

[0088]

number

[0089] where E is the signal level of the RAW image data (input signal level), and E' is the gradation correction RA W is the signal level of the image data (output signal level), OETF[] is the tone correction function, and OB is the optical The gradation correction unit 201 uses the gradation correction algorithm of the following equation (7): The encoding unit 202 outputs a gradation correction identifier 326 indicating the gradation correction unit The gradation correction identifier 326 from 201 is added as the control information 312 in the header information 301. do.

[0090] In this way, by setting the maximum value of the output signal level E' to 1.0, the input / output characteristic graph Compared to the F1200, it is possible to improve color reproduction in bright areas.

[0091] FIG. 16 shows the gain characteristic in the input / output characteristic 2 of the gradation correction when the optical black value is used. This is a graph showing the gain of the input signal level E in tone correction when the optical black value OB is used. The in characteristic graph 1600 is expressed by the following formula (8) by differentiating formula (7).

[0092]

number

[0093] G is the gain of the input signal level E. When the input signal level E goes from 0.0 to the optical black value O Since the gain G value in the section SC9 from B to C is 0.0, the dark area changes from 0.0 to optical black. The input signal level E in the section SC9 up to the optical black value OB is not emphasized. Since the value of InG is the maximum value (infinity), the effective optical black value O By increasing the gain in the section above B, the coding distortion suppression effect can be increased. Cut.

[0094] In addition, as in Example 1 of input / output characteristics of gradation correction when optical black value is used, By increasing the gain G, the distortion of the dark areas occurring between the encoding unit 202 and the decoding unit 602 is reduced. This makes it easier to see the patterns in the dark areas during development and image adjustment. Even if dark areas are made brighter, distortion in the dark areas becomes less noticeable.

[0095] FIG. 17 is a graph showing input / output characteristics 2 of the gradation inverse correction when the optical black value is used. The gradation inverse correction algorithm shown in the input / output characteristic graph 1700 of FIG. 17 is expressed by the following formula (9): It will be revealed.

[0096] E = EOTF[E'] = (1-OB) × (E') γ +OB····(9)

[0097] where E' is the signal level of the gradation-corrected RAW image data (input signal level), and E is the restored The signal level (output signal level) of the RAW image data that is output. EOTF[] is the inverse gradation correction function. The gradation inverse correction unit 603 adds a gradation correction identifier 32 indicating the formula (7) to the control information 312. If 6 is detected, the inverse tone correction of equation (9) is performed.

[0098] Input / output characteristics of gradation correction when using optical black value OB As with Example 1, the restored RAW image In the image data, the minimum value of the output signal level E is the optical black value OB, so the output signal level The color of the pixels in the section SC11 where E is between 0.0 and the optical black value OB is not reproduced, and after development The gain is increased in the area above the optical black value OB, which is effective as an image. The distortion suppression effect can be increased.

[0099] In addition, in a region where the input signal level E′ does not exist as in the input / output characteristic graph 1400 of FIG. The gap SC14 does not exist in the input / output characteristic graph 1700 of FIG. In the example 2 of the input / output characteristics of the gradation correction when the value OB is used, the color reproducibility of the pixels in the bright area is improved. It is possible. EXAMPLES

[0100] The second embodiment reproduces an input signal level E below the optical black value OB while maintaining the optical black value of the first embodiment. As shown in Example 2 of the input / output characteristics of gradation correction when using black value, the gain G of the optical black value OB is set to maximum. In the second embodiment, the differences from the first embodiment will be mainly described, and the same structure as the first embodiment will be described. The configuration and the same processing contents are also used in the second embodiment, and the description thereof will be omitted.

[0101] FIG. 18 is a graph showing input / output characteristics 3 of gradation correction when an optical black value is used. In Example 3 of the input / output characteristics of gradation correction when the optical black value is used, Unlike the correction input / output characteristic examples 1 and 2, the gradation correction unit 201 detects an input lower than the optical black value OB. A negative gradation correction, i.e., a negative 1 / γ correction, is performed on the input signal level E to obtain the optical black value. Positive gradation correction, i.e., positive 1 / γ correction, is performed for input signal levels E that are equal to or greater than OB. do.

[0102] Specifically, for example, in the input / output characteristic graph 1800, a waveform 1801 in a section SC9 is a waveform obtained by negative 1 / γ correction, and the waveform 180 in the section above the optical black value OB 2 is the waveform obtained by positive 1 / γ correction.

[0103] The gradation correction algorithm shown in the input / output characteristic graph 1800 of FIG. 18 is expressed by the following formula (10): be expressed.

[0104]

number

[0105] However, E is the signal level of the RAW image data (input signal level), E' is the signal level of the tone-corrected RA W image data (output signal level), OETF[] is the tone correction function, and OB is the optical black value, and γ is the gamma value.

[0106] Among the expressions in Equation (10), when the condition is E < OB, the expression is a negative 1 / γ correction expression, and when the condition is OB ≤ E, the expression is a positive 1 / γ correction expression. The tone correction unit 201 outputs a tone correction identifier 326 indicating the tone correction algorithm of Equation (10) to the encoding unit 202, and the encoding unit 202 assigns the tone correction identifier 326 from the tone correction unit 201 as control information 312 in the header information 301.

[0107] By performing negative 1 / γ correction in the section SC9, it is possible to improve the color reproducibility near the optical black value OB in the tone-corrected RAW image data. That is, near the optical black value OB of the input / output characteristic graph 1800, the noise on the side where the input signal level E is smaller (negative direction) than the optical black value OB and the noise on the side where the input signal level E is larger (positive direction) than the optical black value OB are canceled out by the noise reduction process, so more efficiently than in the first embodiment. Therefore, it is possible to suppress the black floating near the optical black value OB of the tone-corrected RAW image data and improve the color reproducibility.

[0108] FIG. 19 is a graph showing the gain characteristics in the input / output characteristics 3 of tone correction when using the optical black value. The gain characteristic graph 1900 of the input signal level E in the tone correction when using the optical black value OB is expressed by the following Equation (11) by differentiating Equation (10).​​​​​​​​​​​​​​​​

[0109]

number

[0110] G is the gain of the input signal level E. Specifically, for example, the gain characteristic graph 1 In 900, the gain G reaches its minimum value when the input signal level E is 0.0 in section SC9. It increases from 0.0 and reaches a maximum value (infinity) at the optical black value OB.

[0111] The section SC19 is set to a value between the optical black value OB and the input signal level E that is twice the optical black value OB. The gain G is the gain at which the input signal level E becomes optically It reaches a maximum value (infinity) at the optical black value OB, decreases from the optical black value OB, and reaches a minimum value at 2OB. That is, a waveform 1901 showing the gain characteristic of section SC9 and a waveform 1902 showing the gain characteristic of section SC19 are A waveform 1902 showing the characteristic is linearly symmetrical with respect to the optical black value OB.

[0112] As a result, the optical black value OB is increased in the dark sections SC9 and SC19. In addition, in sections SC9 and SC19, the input signal level The further E is from the optical black value OB, the more the gain G is attenuated. Coding distortion is suppressed by increasing the gain in the area above the optical black value OB, which is effective as an image. It is possible to increase the suppression effect and reduce noise occurring in the dark areas. can.

[0113] In addition, since the waveforms 1901 and 1902 are linearly symmetric with respect to the optical black value OB, In the vicinity of the black value OB, the side where the input signal level E is smaller than the optical black value OB (negative side) noise on the side where the input signal level E is greater than the optical black value OB (positive direction) The noise reduction process cancels out the noise generated by the image and suppresses the black level. It is possible.

[0114] In addition, as in the first embodiment, the gain G of the dark area is increased by the gradation correction. 2 and the decoder 602. When adjusting the image, even if you brighten the dark areas to make them easier to see, the distortion in the dark areas will be noticeable. It becomes difficult.

[0115] FIG. 20 is a graph showing input / output characteristics 3 of the inverse tone correction when the optical black value is used. The gradation inverse correction unit 603 detects a specific input signal level E' of the gradation corrected RAW image data with a value of β or less. A positive gradation inverse correction, i.e., a positive gamma correction, is performed for the input signal level lower than β. Then, negative gradation inverse correction, that is, negative gamma correction, is performed on the pixel.

[0116] Section SC201 is a section in which the input signal level E' is from 0.0 to β shown in equation (10). The section SC202 is the section in which the input signal level E' ranges from β to 1.0. For example, in the input / output characteristic graph 2000, a waveform 2001 in a section SC201 is as follows: The waveform 2002 in the section SC202 is obtained by negative gamma correction. This is the waveform obtained.

[0117] The gradation inverse correction algorithm shown in the input / output characteristic graph 2000 of FIG. 20 is expressed by the following formula (12): It is expressed as:

[0118]

number

[0119] where E' is the signal level of the gradation-corrected RAW image data (input signal level), and E is the restored The signal level (output signal level) of the RAW image data that is output. EOTF[] is the inverse gradation correction function. Among formulas (12), the formula when the condition E'<β is the formula for negative gamma correction, The formula when the condition is β≦E′ is a positive gamma correction formula. When the gradation correction identifier 326 indicating the formula (10) is detected in the information 312, the gradation inversion of the formula (12) is performed. Perform the correction.

[0120] By performing negative gamma correction in section SC201, a restored RAW image with inverse gradation correction is obtained. This can increase the suppression effect of coding distortion near the optical black value OB in the image data, and also reduces the effect of the dark area Since the noise generated in the image forming apparatus can be reduced, color reproducibility can be improved. EXAMPLES

[0121] The third embodiment is an example in which the gain of the bright area is increased in the second embodiment. The differences from the second embodiment will be mainly explained, and the same configuration and processing contents as the second embodiment will be explained. This is also used in the third embodiment, and the explanation thereof will be omitted.

[0122] FIG. 21 is a graph showing input / output characteristics 4 of gradation correction when an optical black value is used. In the input / output characteristic example 4 of the gradation correction when the optical black value is used, the gradation correction unit 201 Negative gradation correction for input signal levels E lower than the value OB, i.e., negative offset 1 / γ correction is performed, and positive gradation correction is performed for input signal levels E that are equal to or greater than the optical black value OB. That is, a positive offset 1 / gamma correction is performed.

[0123] Specifically, for example, in the input / output characteristic graph 2100, a waveform 2101 in a section SC9 is a waveform obtained by negative offset 1 / γ correction, and the area above the optical black value OB Waveform 2102 is the waveform obtained with positive offset 1 / gamma correction.

[0124] The gradation correction algorithm shown in the input / output characteristic graph 2100 of FIG. 21 is expressed by the following formula (13): be expressed.

[0125]

number

[0126] where E is the signal level of the RAW image data (input signal level), and E' is the gradation correction RA W is the signal level of the image data (output signal level), OETF[] is the tone correction function, and OB is the optical is the optical black value, γ is the gamma value, and g is the calculated temporary gain. <OB The formula for the negative offset 1 / gamma correction is the formula for the negative offset 1 / gamma correction, and the formula for the condition OB≦E is is a positive offset 1 / γ correction formula. The gradation correction unit 201 calculates the gradation correction amount of the formula (13). The gradation correction identifier 326 indicating the algorithm is output to the encoding unit 202, and the encoding unit 202 The gradation correction identifier 326 from the gradation correction unit 201 is combined with the control information 312 in the header information 301. and grant it.

[0127] FIG. 22 shows the gain characteristic in the input / output characteristic 4 of the gradation correction when the optical black value is used. This is a graph showing the gain of the input signal level E in tone correction when the optical black value OB is used. The in characteristic graph 2200 is expressed by the following equation (14) by differentiating equation (13).

[0128]

number

[0129] G is the gain of the input signal level E. Specifically, for example, the gain characteristic graph 2 In 200, the gain G reaches its minimum value when the input signal level E is 0.0 in section SC9. The calculated provisional gain g increases from 0.0 and reaches a maximum value at the optical black value OB. IN is a value set by the setting unit 204.

[0130] The gain G is maximum when the input signal level E is at the optical black value OB in the section SC19. The calculated temporary gain g is taken, and it decreases from the optical black value OB and reaches a minimum value at 2OB. , a waveform 2201 showing the gain characteristics of section SC9 and a waveform 2202 showing the gain characteristics of section SC19. 202 is linearly symmetric with respect to the optical black value OB.

[0131] In addition, the gain G at the optical black value OB is changed from infinity in the second embodiment to the calculated provisional gain g. By setting it, the gain characteristic graph 2200 is different from the gain characteristic graph 1900. Therefore, the input signal level of the bright area becomes higher. It is possible to emphasize the pixels in the bright area and the pixels in the bright area.

[0132] FIG. 23 is a graph showing input / output characteristics 4 of the inverse tone correction when the optical black value is used. The gradation inverse correction unit 603 calculates a specific input signal level E' value OU of the gradation corrected RAW image data. For T[1] and above, positive gradation inverse correction, i.e., positive offset gamma correction, is performed. Negative inverse gradation correction for input signal levels lower than T[1], i.e., negative offset Perform gamma correction.

[0133] Define the interval SC231 as the interval where the input signal level E´ ranges from 0.0 to OUT[1] shown in Equation (13). Define the interval SC232 as the interval where the input signal level E´ ranges from OUT[1] to 1.0 Specifically, for example, among the input-output characteristic graphs 2300, the waveform 2301 in the interval SC23 1 is a waveform obtained by negative offset γ correction, and the waveform 2302 in the interval SC232 is a waveform obtained by positive offset γ correction.

[0134] The tone inversion correction algorithm that shows the input-output characteristic graph 2300 in Figure 23 is represented by the following Equation (15). It is expressed as follows.

[0135]

Equation

[0136] However, E´ is the signal level (input signal level) of the tone-corrected RAW image data, E is the signal level (output signal level) of the restored RAW image data, and EOTF[] is the tone inversion correction function. Among Equation (15), the equation when the condition is E´ < OUT[1] is the equation for negative offset γ correction, and the equation when the condition is OUT[1] ≤ E´ is the equation for positive offset γ correction. When the tone inversion correction unit 603 detects the tone correction identifier 326 shown in Equation (13) in the control information 312, it executes the tone inversion correction of Equation (15). By performing negative offset γ correction in the interval SC231, it is possible to improve the color reproducibility near the optical black value OB in the tone-inverted RAW image data. Also, since the gain in the bright part also increases, it is possible to emphasize the pixels on the bright part side by tone correction.

[0137]

[0138] ​​​​​Further, the setting unit 204 may set the calculated gain G instead of the calculated temporary gain g OB in the tone inverse correction unit 6 03 (the following formula (16)). By setting the calculated gain G OB in the tone inverse correction unit 603, it is possible to improve the reproducibility of the optical black and bright parts, which are the dark parts in the original RAW image data, in tone inverse correction.

[0139] G OB = C × g ····· (16)

Example

[0140] Example 4 is an example in which an input signal level E with a certain width including the optical black value is provided for the calculated temporary gain g in Example 3. In Example 4, the description will be centered on the differences from Example 3, and for the same configuration and the same processing content, they will also be reused in Example 4 and the description thereof will be omitted.

[0141] FIG. 24 is a graph showing the input / output characteristics 5 of tone correction when using the optical black value. In the example 5 of the input / output characteristics of tone correction when using the optical black value, a certain port width D including the optical black value OB is set by the setting unit 204. At the input signal level E of the port width D, the same gain G is obtained.

[0142] Of the port width D, the range where the value of the input signal level E is from P1 (0.0 < P1 < OB) to the optical black value OB is defined as the first port width D1, and the range where the value of the input signal level E of the port width D is from the optical black value OB to P2 (OB ≤ P2) is defined as the second port width D2. The port width D is the range where the gain G of the input signal level E is constant.

[0143] For the section SC24 where the value of the input signal level E is from 0.0 to P1, the tone correction unit 201 Then, perform negative tone correction, i.e., negative offset 1 / γ correction, on the input signal level E For the interval of the value of the input signal level E within the port width D, perform tone correction where the gain G is constant, and for the interval where the value of the input signal level E is equal to or greater than P2, perform positive tone correction, i.e., positive offset 1 / γ correction.

[0144] Specifically, for example, among the input-output characteristic graphs 2400, the waveform 240 1 in the interval SC24 is a waveform obtained by negative offset 1 / γ correction, and the waveform 2 402 in the interval of the port width D is a waveform obtained by tone correction where the gain G is constant. The waveform 2403 in the interval where the value of the input signal level E is equal to or greater than P2 is a waveform obtained by positive offset 1 / γ correction is.

[0145] The tone correction algorithm showing the input-output characteristic graph 2400 in FIG. 24 is expressed by the following formula (17). is expressed.

[0146]

Equation

[0147] However, E is the signal level of the RAW image data (input signal level), E' is the signal level of the tone-corrected RAW W image data (output signal level), OETF[] is the tone correction function, OB is the optical black value, γ is the gamma value, and g is the calculated provisional gain. Also, IN[1] is the first port width D1, and IN[2] is the second port width D2.

[0148] Among the formula (17), the formula for the case where the condition is E < OB is the formula for negative offset 1 / γ correction, and the formula for the case where the condition is IN[1] ≤ E < IN[2], i.e., the port width D, is the gai n formula, and the formula for the case where the condition is E ≥ IN[2] is the formula for positive offset 1 / γ correction. This is a gradation correction in which the offset G is constant, and the formula when the condition is IN[2]≦E is a positive offset. The gradation correction unit 201 uses the gradation correction algorithm of equation (17). The encoding unit 202 outputs the gradation correction identifier 326 to the gradation correction unit 2 A tone correction identifier 326 starting from 01 is added as control information 312 in the header information 301. .

[0149] FIG. 25 shows the gain characteristic in the input / output characteristic 5 of gradation correction when the optical black value is used. This is a graph showing the gain of the input signal level E in gradation correction when the optical black value OB is used. The in characteristic graph 2500 is expressed by the following equation (18) by differentiating equation (17).

[0150]

number

[0151] G is the gain of the input signal level E. Specifically, for example, the gain characteristic graph 2 In the case of 500, the gain G reaches its minimum value when the input signal level E is 0.0 in section SC24. The calculated provisional gain g increases from 0.0 and reaches a maximum value at P1, which is the right end of the porch width D. The calculated temporary gain g is a value set by the setting unit 204.

[0152] The gain G is constant at the calculated provisional gain g for the porch width D. The gain G is expressed as follows: At or above P2, which is the left end of the porch width D, the optical black value decreases from OB. By setting D1, noise near the optical black value OB can be preserved. By setting the second pouch width D2, the higher the compression ratio, the greater the black expression. can.

[0153] In addition, the second pouch width D2 is variable depending on the amount of exposure set by the setting unit 204. Specifically, the optical black value OB is fixed, and the position of P2 is changed. For example, By setting the exposure amount to underexposure in unit 204, P2 is adjusted to the input signal level E is increased, and the second pouch width P2 is increased.

[0154] On the other hand, by setting the exposure amount to overexposure in the setting unit 204, P2 is The signal level E is changed in a decreasing direction, and the second porch width P2 is narrowed. The second pouch width D2 can be expanded or contracted depending on the amount of exposure, and the optical black value can be adjusted depending on the amount of exposure. Nearby pixels can be emphasized.

[0155] FIG. 26 is a graph showing input / output characteristics 5 of the inverse tone correction when the optical black value is used. The gradation inverse correction unit 603 calculates a specific input signal level E' value OU of the gradation corrected RAW image data. For T[2] and above, positive gradation inverse correction, i.e., positive offset gamma correction, is performed. Inverse tone correction with constant gain G from T[1] to OUT[2], OUT[1] For input signal levels lower than Execute.

[0156] The section SC261 is a section in which the input signal level E' is from 0.0 to OUT[1] The section SC262 is the section where the input signal level E' is from OUT[1] to OUT[2 ], and section SC263 is the section where the input signal level E' is from OUT[2] to 1.0. The section is as follows.

[0157] Specifically, for example, in the input / output characteristic graph 2600, the waveform 23 in the section SC261 01 is a waveform obtained by negative offset γ correction, which is the waveform 260 in the section SC262 2 is a waveform obtained by tone reverse correction with a constant gain G, which is the wave in the section SC263 form 2603 is a waveform obtained by positive offset γ correction.

[0158] The tone reverse correction algorithm showing the input / output characteristic graph 2600 in Fig. 26 is represented by the following formula (19) as follows.

[0159]

Equation

[0160] However, E´ is the signal level (input signal level) of the tone-corrected RAW image data, E is the signal level (output signal level) of the restored RAW image data, and EOTF[] is the tone reverse correction function. Among the formula (19), when the condition is E´ < OUT[1], the formula is the formula for negative offset γ correction, when the condition is In[1] ≦ E < In[2], the formula is the formula for tone reverse correction with a constant gain G, and when the condition is OUT[2] ≦ E´, the formula is the formula for positive offset γ correction. When the tone reverse correction unit 603 detects the tone correction identifier 326 indicating the formula (17) in the control information 312, it executes the tone reverse correction of the formula (19).

[0161] By performing negative offset γ correction in the section SC261, it is possible to improve the color reproducibility near the optical black value OB in the tone-reversed RAW image data. Also, since the gain in the bright part also increases, pixels on the bright part side can be emphasized by tone correction.

Example

[0162] ​​​​​​​​In the fifth embodiment, the input signal level E in the input / output characteristic graph is set to 0 to OB. In the fifth embodiment, the difference from the third embodiment will be mainly described. The same configuration and the same processing contents as those in the third embodiment are also used in the fifth embodiment. The explanation will be omitted.

[0163] FIG. 27 is a graph showing input / output characteristics 6 of gradation correction when an optical black value is used. In the input / output characteristic example 6 of the gradation correction when the optical black value is used, the gradation correction unit 201 For input signal levels E lower than the value OB, tone correction is performed so that the positive slope becomes constant. For an input signal level E equal to or greater than the optical black value OB, a positive gradation correction is performed, i.e., a positive off-state correction is performed. Set 1 / Perform gamma correction.

[0164] Specifically, for example, in the input / output characteristic graph 2700, a waveform 2701 in a section SC9 is a linear waveform in which the output signal level E' increases as the input signal level E increases. That is, in the section SC9, there is a direct proportion between the input signal level E and the output signal level E'. The waveform 2702 in the section SC27 (OB≦E) above the optical black value OB is positive. This is a waveform obtained by offset 1 / γ correction. The gradation correction unit 201 adjusts the input signal level Fitting with gradation correction so that when E is 1.0, the output signal level E' is 1.0 do.

[0165] The gradation correction algorithm shown in the input / output characteristic graph 2700 of FIG. 27 is expressed by the following formula (20): be expressed.

[0166]

number

[0167] However, E is the signal level of the RAW image data (input signal level), and E' is the signal level of the tone-corrected RA W image data (output signal level), OETF[] is the tone correction function, and OB is the optical black value, and γ is the gamma value. S is a fixed value indicating the slope of the waveform 2701 in section SC9 There is. In FIG. 27, for example, S = 4. The slope S may be a positive value. α is a fitting coefficient adjustable by the setting unit 204. Specifically, for example, the setting unit 204 can adjust the fitting coefficient α according to the increase or decrease of the ISO sensitivity in the information processing apparatus 100 is possible.

[0168] Among the equations (20), the equation when the condition is E < OB is a tone correction equation such that the positive slope with respect to the input signal level E is constant, and the equation when the condition is OB ≦ E is a positive offset 1 / γ correction equation. The tone correction unit 201 outputs the tone correction identifier 326 indicating the tone correction algorithm of the equation (20) to the encoding unit 202, and the encoding unit 202 adds the tone correction identifier 326 from the tone correction unit 201 as control information 312 in the header information 301 is. is. shows. 01 to the control information 312 in the header information 301 .

[0169] FIG. 28 is a graph showing the gain characteristics in the input / output characteristics 6 of tone correction when using the optical black value is. The gain characteristic graph 2800 of the input signal level E in the tone correction when using the optical black value OB is expressed by the following equation (21) by differentiating the equation (20). is.

[0170]

Equation

[0171] G is the gain of the input signal level E. Specifically, for example, the gain characteristic graph 2 In 800, the gain G has a constant value in section SC9.

[0172] In addition, as in the first embodiment, the gain G of the dark area is increased by the gradation correction. 2 and the decoder 602. When adjusting the image, even if you brighten the dark areas to make them easier to see, the distortion in the dark areas will be noticeable. It becomes difficult.

[0173] FIG. 29 is a graph showing input / output characteristics 6 of the inverse tone correction when the optical black value is used. The gradation inverse correction unit 603 has a positive slope for an input signal level E' that is lower than the optical black value OB. A gradation inverse correction is performed so that is constant, and a negative gradation is Inverse tone correction, that is, negative gamma correction, is performed.

[0174] Section SC291 is the section where the input signal level E' is from 0.0 to β shown in equation (10). Section SC292 is the section where the input signal level E' ranges from β to 1.0. For example, in the input / output characteristic graph 2900, a waveform 2901 in a section SC291 is as follows: This is a waveform obtained by inverse gradation correction so that the positive slope is constant. Shape 2902 is the waveform obtained with positive gamma correction.

[0175] The gradation inverse correction algorithm shown in the input / output characteristic graph 2900 of FIG. 29 is expressed by the following formula (22): It is expressed as:

[0176]

number

[0177] However, E´ is the signal level of the tone-corrected RAW image data (input signal level), and E is the signal level of the restored RAW image data (output signal level), and EOTF[] is the tone inverse correction function. Among the equations (22), when the condition is E´ < β, the equation for the input signal level E lower than the optical black value OB is a tone inverse correction equation such that the positive slope is constant, and when the condition is β ≦ E´, the equation is a positive γ correction equation. When the tone inverse correction unit 603 detects the tone correction identifier 326 shown in the equation (20) in the control information 312, it

[0178] executes the tone inverse correction of the equation (22). By executing the tone inverse correction such that the positive slope is constant in the interval SC201, the encoding distortion suppression effect near the optical black value OB in the tone-inverse-corrected restored RAW image data can be increased, and the noise generated in the dark part can be reduced, so that the color reproducibility can be

[0179] Example 6 shows an example in which two fitting coefficients are used in Example 5. Specifically, for example, in Example 5, in the input / output characteristic graph 2700, the fitting coefficient α was adjustable in the interval where the input signal level E is equal to or higher than OB (OB ≦ E). In Example 6, different fitting coefficients are used in each of the intervals SC9 where the input signal level E is less than OB (E < OB) and the interval where it is equal to or higher than OB (OB ≦ E). In Example 6, the description will focus on the differences from Example 5, and for the same

[0180] FIG. 30 is a graph showing input / output characteristics 7 of gradation correction when an optical black value is used. In the input / output characteristic example 7 of the gradation correction when the optical black value is used, the gradation correction is performed in the same manner as in the input / output characteristic example 6. The positive portion 201 has a constant positive slope for an input signal level E that is lower than the optical black value OB. For input signal levels E that are equal to or greater than the optical black value OB, positive gradation correction is performed. Positive, i.e., a positive offset 1 / gamma correction is performed.

[0181] Specifically, for example, in the input / output characteristic graph 3000, the waveform 3001 in the section SC9 is a linear waveform in which the output signal level E' increases with increasing input signal level E, The waveform 3002 in the section above the optical black value OB is obtained by positive offset 1 / γ correction. The gradation correction unit 201 outputs a signal level E when the input signal level E is 1.0. Fit using tone correction so that ´ becomes 1.0.

[0182] The gradation correction algorithm shown in the input / output characteristic graph 3000 of FIG. 30 is expressed by the following formula (23): It is expressed.

[0183]

number

[0184] where E is the signal level of the RAW image data (input signal level), and E' is the gradation correction RA W is the signal level of the image data (output signal level), OETF[] is the tone correction function, and OB is the optical is the optical black value, γ is the gamma value, α1 is an adjustable fitting coefficient, and is in the interval S C9 waveform 3001. α1 is, for example, α1=2 in FIG. α1 can be any positive value. α2 is a fitting coefficient different from α1. depends on

[0185] The setting unit 204 can adjust the fitting coefficient α according to the increase or decrease of the ISO sensitivity in the information processing apparatus 100. Thereby, in the section SC9, the input / output characteristic graph 3000 can be adjusted with the fitting coefficient α1, and in the section SC27 where OB ≦ E, the input / output characteristic graph 3000 can be adjusted with a fitting coefficient α2 different from the fitting coefficient α1.

[0186] Among the formulas (23), the formula when the condition is E < OB is a gradation correction formula such that the positive slope with respect to the input signal level E is constant, and the formula when the condition is OB ≦ E is a positive offset 1 / γ correction formula. The gradation correction unit 201 outputs the gradation correction identifier 326 indicating the gradation correction algorithm of the formula (23) to the encoding unit 202, and the encoding unit 202 assigns the gradation correction identifier 326 from the gradation correction unit 201 as control information 312 in the header information 301. .

[0187] FIG. 31 is a graph showing the gain characteristics in the input / output characteristics 7 of gradation correction when using the optical black value. The gain characteristic graph 3100 of the input signal level E in the gradation correction when using the optical black value OB is expressed by the following formula (24) by differentiating the formula (23).

[0188]

Equation

[0189] G is the gain of the input signal level E. Specifically, for example, in the gain characteristic graph 3100, the gain G takes a constant value in the section SC9. ​​​​​​​​​​​

[0190] In addition, as in the first embodiment, the gain G of the dark area is increased by the gradation correction. 2 and the decoder 602. When adjusting the image, even if you brighten the dark areas to make them easier to see, the distortion in the dark areas will be noticeable. It becomes difficult.

[0191] FIG. 32 is a graph showing input / output characteristics 7 of the inverse tone correction when the optical black value is used. The gradation inverse correction unit 603 has a positive slope for an input signal level E that is lower than the optical black value OB. β is a constant, and negative gradation inversion is performed for input signal levels lower than β. correction, i.e., negative gamma correction.

[0192] Section SC321 is a section in which the input signal level E' is from 0.0 to β shown in equation (10). Section SC292 is the section where the input signal level E' ranges from β to 1.0. For example, in the input / output characteristic graph 3200, a waveform 3201 in a section SC321 is as follows: The waveform 3202 in section SC322 is obtained by negative gamma correction. This is the waveform obtained.

[0193] The gradation inverse correction algorithm shown in the input / output characteristic graph 3200 of FIG. 32 is expressed by the following formula (25): It is expressed as:

[0194]

number

[0195] where E' is the signal level of the gradation-corrected RAW image data (input signal level), and E is the restored The signal level (output signal level) of the RAW image data that is output. EOTF[] is the inverse gradation correction function. Among formula (25), the formula when the condition E'<β is lower than the optical black value OB. This is the equation for inverse tone correction so that the positive slope is constant for the input signal level E, and the condition is The equation when β≦E′ is a positive gamma correction equation. 2, if a gradation correction identifier 326 indicating the formula (23) is detected, the gradation inverse correction of the formula (25) is performed. Execute.

[0196] By performing inverse gradation correction so that the positive slope becomes constant in the section SC321, The effect of suppressing coding distortion near the optical black value OB in the restored RAW image data that has been inversely corrected is greatly improved. This can reduce noise in the dark areas, improving color reproducibility. Improvements can be achieved. EXAMPLES

[0197] Example 7 is another example of Example 2, and reproduces an input signal level E below the optical black value OB. While using the optical black value of the first embodiment, the optical black value is used as in the input / output characteristic example 2 of the gradation correction. This is an example of maximizing the gain G of the black value OB. In the seventh embodiment, the differences from the second embodiment are mainly described. The same configuration and the same processing contents as those in the second embodiment are also used in the seventh embodiment. , the explanation of which will be omitted.

[0198] FIG. 33 is a graph showing input / output characteristics 8 of gradation correction when an optical black value is used. In Example 8 of the input / output characteristics of gradation correction when the optical black value is used, Unlike the correction input / output characteristic examples 1 and 2, the gradation correction unit 201 detects an input lower than the optical black value OB. A negative gradation correction, i.e., a negative 1 / γ correction, is performed on the input signal level E to obtain the optical black value. For an input signal level E above OB, perform positive tone correction, i.e., positive 1 / γ correction. Do it.

[0199] Specifically, for example, among the input / output characteristic graphs 3300, the waveform 3301 in the section SC9 is a waveform obtained by negative 1 / γ correction, and the waveform 330 2 in the section above the optical black value OB is a waveform obtained by positive 1 / γ correction.

[0200] The tone correction algorithm showing the input / output characteristic graph 3300 in FIG. 33 is expressed by the following formula (26). Be expressed.

[0201]

Number

[0202] However, E is the signal level (input signal level) of the RAW image data, E´ is the signal level (output signal level) of the tone-corrected RAW W image data, OETF[] is the tone correction function, OB is the optical black value, γ is the gamma value, and α is a fitting coefficient adjustable by the setting unit 204. Specifically, for example, the setting unit 204 can adjust the fitting coefficient α according to the increase or decrease of the ISO sensitivity in the information processing apparatus 100. Be possible.

[0203] Among the formula (26), the formula when the condition is E < OB is the formula for negative 1 / γ correction, and the condition When it is OB ≤ E, the formula is the formula for positive 1 / γ correction. The tone correction unit 201 outputs the tone correction identifier 326 indicating the tone correction algorithm of the formula (26) to the encoding unit 202, and the encoding The unit 202 gives the tone correction identifier 326 from the tone correction unit 201 as control information 312 in the header information 301. Be given. Do it.

[0204] By performing negative 1 / γ correction in section SC9, the gradation-corrected RAW image data is This improves the color reproducibility near the optical black value OB in the input / output characteristics. In the vicinity of the optical black value OB of the graph 3300, the input signal level E The noise level E is greater than the optical black value OB and the noise level E is smaller (negative direction). The noise reduction process cancels out the noise on the positive side (positive direction) and the noise on the positive side (negative direction). Therefore, the noise can be effectively removed from the RAW image data after tone correction. This suppresses the black floating around the optical black value OB of the image data, thereby improving color reproducibility.

[0205] FIG. 34 shows the gain characteristic in the input / output characteristic 8 of gradation correction when the optical black value is used. This is a graph showing the gain of the input signal level E in gradation correction when the optical black value OB is used. The in characteristic graph 3400 is expressed by the following equation (27) by differentiating equation (26).

[0206]

number

[0207] G is the gain of the input signal level E. Specifically, for example, the gain characteristic graph 3 In 400, the gain G reaches its minimum value when the input signal level E is 0.0 in section SC9. It increases from 0.0 and reaches a maximum value (infinity) at the optical black value OB.

[0208] In addition, by increasing the gain G of the dark area in the tone correction, the encoding unit 202 and the decoding unit 602 This suppresses distortion in the dark areas that occurs between the Even if the dark areas are made brighter to make the pattern easier to see, distortion in the dark areas becomes less noticeable.

[0209] FIG. 35 is a graph showing input / output characteristics 8 of the inverse tone correction when the optical black value is used. The gradation inverse correction unit 603 detects a specific input signal level E' of the gradation corrected RAW image data with a value of β or less. A positive gradation inverse correction, i.e., a positive gamma correction, is performed for the input signal level lower than β. Then, negative gradation inverse correction, that is, negative gamma correction, is performed on the pixel.

[0210] Section SC351 is the section where the input signal level E' is from 0.0 to β shown in equation (10). Section SC352 is the section where the input signal level E' ranges from β to 1.0. For example, in the input / output characteristic graph 3500, a waveform 3501 in a section SC351 is as follows: The waveform 3502 in section SC352 is obtained by negative gamma correction. This is the waveform obtained.

[0211] The gradation inverse correction algorithm shown in the input / output characteristic graph 3500 of FIG. 35 is expressed by the following formula (28): It is expressed as:

[0212]

number

[0213] where E' is the signal level of the gradation-corrected RAW image data (input signal level), and E is the restored The signal level (output signal level) of the RAW image data that is output. EOTF[] is the inverse gradation correction function. Among the formulas (28), the formula when the condition E'<β is the formula for negative gamma correction, The formula when the condition is β≦E′ is a positive gamma correction formula. When the gradation correction identifier 326 indicating the formula (26) is detected in the information 312, the gradation inversion of the formula (28) is performed. Perform the correction.

[0214] By performing negative γ correction in the interval SC351, the coding distortion suppression effect near the optical black value OB in the restored RAW image data with tone inverse correction can be enhanced, and the noise generated in the dark part can be reduced, thus improving color reproducibility.

Example

[0215] Example 8 shows an example using two fitting coefficients in Example 7. Specifically, for example, in Example 7, in the interval SC27 where the input signal level E is OB or higher (OB ≦ E) in the input-output characteristic graph 3300, the fitting coefficient α can be adjusted. In Example 8, different fitting coefficients are used in each of the interval SC9 where the input signal level E is less than OB (E < OB) and the interval SC27 where the input signal level E is OB or higher (OB ≦ E). In Example 8, the description will focus on the differences from Example 7, and the same configurations and processing contents as in Example 7 will also be borrowed in Example 8, and the description thereof will be omitted.

[0216] Figure 36 is a graph showing the input-output characteristics 9 of tone correction when using the optical black value. In the input-output characteristics example 9 of tone correction when using the optical black value, similar to the input-output characteristics example 8, different from the input-output characteristics examples 1 and 2 of tone correction when using the optical black value, the tone correction unit 201 performs negative tone correction, that is, negative 1 / γ correction, for the input signal level E lower than the optical black value OB, and performs positive tone correction, that is, positive 1 / γ correction, for the input signal level E equal to or higher than the optical black value OB.

[0217] Specifically, for example, among the input-output characteristic graph 3600, the waveform 3601 in the interval SC9 ​​​​​​​​​​​​is a waveform obtained by negative 1 / γ correction, and the waveform 3602 in the section SC27 above the optical black value OB is a waveform obtained by positive 1 / γ correction.

[0218] The tone correction algorithm showing the input / output characteristic graph 3600 in FIG. 36 is expressed by the following formula (29). expressed.

[0219] [Number]

[0220] However, E is the signal level of the RAW image data (input signal level), E' is the signal level of the tone-corrected RAW W image data (output signal level), OETF[] is the tone correction function, OB is the optical black value, γ is the gamma value, and α1 and α2 are adjustable fitting coefficients. α2 is a fitting coefficient different from α1 and depends on α1.

[0221] The setting unit 204 can adjust the fitting coefficient α according to the increase or decrease of the ISO sensitivity in the information processing apparatus 100. Thereby, in the section SC9, the input / output characteristic graph 3600 can be adjusted with the fitting coefficient α1, and in the section SC27 where OB ≤ E, the input / output characteristic graph can be adjusted with a fitting coefficient α2 having a value different from the fitting coefficient α1. graph 3600.

[0222] Among the formula (29), the formula when the condition is E < OB is the formula for negative 1 / γ correction, and the condition when is OB ≤ E is the formula for positive 1 / γ correction. The tone correction unit 201 outputs a tone correction identifier 326 indicating the tone correction algorithm of the formula (29) to the encoding unit 202, and the encoding unit 202 encodes the tone correction identifier 326 from the tone correction unit 201 into the control in the header information 301. ​​ This will be provided as information 312.

[0223] By performing negative 1 / γ correction in section SC9, the gradation-corrected RAW image data is This improves the color reproducibility near the optical black value OB in the input / output characteristics. In the vicinity of the optical black value OB of graph 3600, the input signal level E The noise level E is greater than the optical black value OB and the noise level E is smaller (negative direction). The noise reduction process cancels out the noise on the positive side (positive direction) and the noise on the positive side (negative direction). Therefore, the noise can be effectively removed from the RAW image data after tone correction. This suppresses the black floating around the optical black value OB of the image data, thereby improving color reproducibility.

[0224] FIG. 37 shows the gain characteristics in the input / output characteristics 9 of gradation correction when the optical black value is used. This is a graph showing the gain of the input signal level E in gradation correction when the optical black value OB is used. The in characteristic graph 3700 is expressed by the following equation (30) by differentiating equation (29).

[0225]

number

[0226] G is the gain of the input signal level E. Specifically, for example, the gain characteristic graph 3 In the 700, the gain G reaches its minimum value when the input signal level E is 0.0 in section SC9. It increases from 0.0 and reaches a maximum value (infinity) at the optical black value OB.

[0227] In addition, by increasing the gain G of the dark area in the tone correction, the encoding unit 202 and the decoding unit 602 This suppresses distortion in the dark areas that occurs between the Even if the dark areas are made brighter to make the pattern easier to see, distortion in the dark areas becomes less noticeable.

[0228] FIG. 38 is a graph showing input / output characteristics 9 of the inverse tone correction when the optical black value is used. The gradation inverse correction unit 603 detects a specific input signal level E' of the gradation corrected RAW image data with a value of β or less. A positive gradation inverse correction, i.e., a positive gamma correction, is performed for the input signal level lower than β. Then, negative gradation inverse correction, that is, negative gamma correction, is performed on the pixel.

[0229] Section SC381 is the section where the input signal level E' is from 0.0 to β shown in equation (10). Section SC382 is the section where the input signal level E' ranges from β to 1.0. For example, in the input / output characteristic graph 3800, a waveform 3801 in a section SC381 is as follows: The waveform 3802 in section SC382 is obtained by negative gamma correction. This is the waveform obtained.

[0230] The gradation inverse correction algorithm shown in the input / output characteristic graph 3800 of FIG. 38 is expressed by the following formula (31): It is expressed as:

[0231]

number

[0232] where E' is the signal level of the gradation-corrected RAW image data (input signal level), and E is the restored The signal level (output signal level) of the RAW image data that is output. EOTF[] is the inverse gradation correction function. Among the formulas (31), the formula when the condition E'<β is the formula for negative gamma correction, The formula when the condition is β≦E′ is a positive gamma correction formula. When the gradation correction identifier 326 indicating the formula (29) is detected in the information 312, the gradation inversion of the formula (31) is performed. Perform the correction.

[0233] By performing negative gamma correction in section SC381, the restored RAW image is inversely toned. This can increase the suppression effect of coding distortion near the optical black value OB in the image data, and also reduces the effect of the dark area Since the noise generated in the image forming apparatus can be reduced, color reproducibility can be improved.

[0234] As described above, according to this embodiment, it is possible to suppress dark coding distortion caused by tone correction. In addition, for example, it is possible to make the pattern in the dark areas easier to see during development processing or image adjustment. When expressing dark areas brighter, the gain G of the dark areas is increased by tone correction. This can suppress the distortion of the dark areas that occurs between the image sensor 202 and the decoder 602. When processing or adjusting an image, even if you brighten the dark areas to make them easier to see, Distortion in the image becomes less noticeable. In addition, the original RAW image data can be restored by inverse gradation correction. can be done. [Explanation of symbols]

[0235] 100 Information processing device, 153 Image sensor, 161 Effective pixel area, 162 Optical black Pixel area, 200 encoding device, 201 gradation correction unit, 202 encoding unit, 203 recording unit, 204 setting unit, 312 control information, 326 gradation correction identifier, 600 decoding device, 601 Acquisition unit, 602 Decoding unit, 603 Tone inverse correction unit

Claims

[Claim 1] a correction unit that performs gradation correction on raw image data from an image pickup element having optical black based on a gamma coefficient and the value of the optical black; an encoding unit that encodes the gradation-corrected RAW image data that has been gradation-corrected by the correction unit; An encoding device having the following construction.

Images