JP2015213276A - Image processing system and image processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technique capable of obtaining a range compressed image whose object color reproducibility using a range near saturation is improved.SOLUTION: In an image processing system 1 that performs range compression of the number of bits of each pixel value constituting an input image from A bits to B bits (where, A>B), a maximum value detection section 41 for detecting the maximum value of pixel values constituting an input image, a margin determination section 42 for determining a margin value on the basis of the maximum value, a coefficient calculation section 44 for calculating a coefficient on the basis of a sum of the maximum value and the margin value, and a compression section 45 for performing range compression of the input image by using the coefficient are provided.

Description

  The present invention relates to an image processing apparatus and an image processing method.

  In recent years, an imaging function called WDR (Wide Dynamic Range) or HDR (High Dynamic Range) for obtaining an image (hereinafter also referred to as “WDR signal”) that captures a dynamic range that exceeds the dynamic range that can be captured by the sensor has increased. ing. Such a photographing function is particularly effective in a scene with a very large contrast ratio such as a composition of backlight.

  As means for obtaining the WDR signal, a short-exposure image (hereinafter also simply referred to as “short-exposure image”) and a long-exposure image (hereinafter also simply referred to as “long-exposure image”) are successively displayed one by one. Means for photographing and compositing, means for photographing short-exposure images and long-exposure images for each region, means for using a sensor that improves imaging ability and obtains a wide dynamic range by one-shot photography, etc. There is. The dynamic range obtained by such means may be a range of 16 bits to 20 bits or more.

  However, even if a WDR signal having such a dynamic range is obtained, the display capability of the display to be finally displayed may be limited to about 10 bits. If various signal processing is performed, a problem that the processing amount becomes enormous will occur. Therefore, for example, it is necessary to significantly compress the range of the WDR signal so that each pixel value is compressed from 20 bits to 10 bits.

  Here, even if a WDR system capable of capturing a 20-bit range is used, the subject range may reach the upper limit of 20 bits, but depending on the scene, it only reaches about 16 bits. Sometimes there isn't. Therefore, when the ratio of the output upper limit range to the input upper limit range is fixedly used as the compression ratio, a dark image may be output using only the low luminance side of the output range, and the compression ratio is small. If it is too high, the compression will be insufficient and the output image will be saturated.

  Therefore, in order to prevent the output image from being saturated while effectively utilizing the dynamic range of the output image, it is necessary to adaptively perform compression according to the dynamic range of the input image. As a technique for achieving such an object, techniques disclosed in the following Patent Documents 1 to 4 are disclosed.

  For example, in Patent Document 1, the maximum value of the input RGB signal is detected, the level ratio of the maximum value that can be taken by the signal after compression of the input RGB signal with respect to the input RGB signal is defined as the compression ratio, and the compression ratio is used as the input RGB signal. A technique for multiplying is disclosed. According to such arithmetic processing, it is possible to prevent the compressed signal from being saturated beyond the output range.

  Patent Document 2 discloses a technique in which a correction value is defined according to the level of an input signal and the correction value is added to or subtracted from the input signal. With this calculation, color saturation can be reduced particularly by lowering the level of the highlight portion of the input signal.

  Patent Document 3 calculates a luminance signal from an input RGB signal, subtracts the luminance signal from the input RGB signal to generate a color difference signal, performs knee processing on the luminance signal, and outputs the luminance signal. A technique for re-adding to a color difference signal is disclosed.

  In Patent Document 4, gain calculation is performed on an input RGB signal, a maximum value is detected from the calculated RGB signal, a suppression coefficient is obtained based on the detected maximum value, and a saturation level is obtained using the suppression coefficient. A technique for compressing the pixel value is disclosed.

JP 2000-115796 A JP 2005-109871 A JP 2009-111894 A JP 2004-180090 A

  However, in the technique described in Patent Document 1, the color reproducibility of an object using a range near saturation is not sufficient.

  In addition, the technique described in Patent Document 2 aims to keep the range of the output signal within a range that does not saturate, and the color reproducibility of an object that uses a range near saturation is not sufficient.

  Further, with the technique described in Patent Document 3, the color reproducibility of an object using a range near the saturation of the output signal is not sufficient. Further, in the technique described in Patent Document 3, knee processing is performed to perform range compression only on a high-luminance portion of a certain level or higher, and it seems difficult to apply this technique to significant range compression. It is.

  Further, in the technique described in Patent Document 4, it is assumed that the number of bits of each pixel value is amplified from about 8 bits to about 9 bits and the saturated pixel value is kept within 8 bits. Compression is not expected. Further, with the technique described in Patent Document 4, the color reproducibility of an object using a range near saturation is not sufficient.

  Therefore, the present invention provides a technique capable of obtaining a range-compressed image with improved color reproducibility of an object that uses a range near saturation.

  According to an embodiment of the present invention, in an image processing apparatus that performs range compression on the number of bits of each pixel value constituting an input image from A bits to B bits (A> B), the pixels constituting the input image A maximum value detecting unit for detecting a maximum value, a margin determining unit for determining a margin value based on the maximum value, and a coefficient calculating unit for calculating a coefficient based on an added value of the maximum value and the margin value And a compression unit that performs range compression on the input image using the coefficient.

  According to such a configuration, it is possible to obtain a range-compressed image in which the color reproducibility of a high-brightness / high-saturation object is improved by performing control so as not to use a certain range from the upper limit of the output range. .

  The margin determination unit may determine the margin value based on a comparison result between a saturation value of a pixel value constituting the input image and the maximum value, or a comparison result between a predetermined threshold and the maximum value The margin value may be determined based on

  For example, the margin determination unit may determine zero as the margin value when the maximum value is greater than or equal to the saturation value. This is because when the input image is saturated, it is natural that the compressed image is also saturated.

  For example, the margin determination unit may determine a specified value as the margin value when the maximum value is lower than the threshold value. For example, when the maximum value is between the saturation value and the threshold value, the margin determination unit may determine a value calculated by linear interpolation as the margin value.

  The saturation value may be set in any way. For example, the input image is a composite image of a short exposure image and a long exposure image, and the saturation value is the number of bits of each pixel value output from a sensor, the short exposure image captured by the sensor, and the It may be a multiplication value with the exposure ratio of the long exposure image.

  Alternatively, the input image is a composite image of a short-exposure image and a long-exposure image, and the margin determination unit includes a saturation value of a pixel value constituting the short-exposure image and a pixel value constituting the short-exposure image. The margin value may be determined based on a comparison result with the maximum value.

According to another embodiment of the present invention, in the image processing method for performing range compression on the number of bits of each pixel value constituting an input image from A bit to B bit (A> B), the input image Detecting a maximum value of constituent pixel values; determining a margin value based on the maximum value; calculating a coefficient based on an added value of the maximum value and the margin value; and the coefficient Range-compressing the input image using
An image processing method is provided.

  According to such a method, it is possible to obtain a range-compressed image in which the color reproducibility of a high-brightness / high-saturation object is improved by performing control so as not to use a certain range from the upper limit of the output range. .

  As described above, according to the present invention, it is possible to obtain a range-compressed image with improved color reproducibility of an object that uses a range near saturation.

It is a figure for demonstrating the color reproducibility at the time of using the range near saturation. It is a figure which shows the function structural example of the image processing apparatus which concerns on embodiment of this invention. It is a block diagram which shows the detailed functional structural example of a range compression part. It is a figure for demonstrating the example of a margin value determination. WDR composite image in which the maximum pixel value has not reached the saturation value, WDR composite image after compression to the 12-bit range using the maximum value, and compression to the 12-bit range using the sum of the maximum value and the margin value It is a figure which shows the histogram of each WDR synthetic | combination image after having performed. WDR composite image in which the maximum pixel value reaches the saturation value, WDR composite image after compression to the 12-bit range using the sum of the maximum value and the margin value, and compression to the 12-bit range using the maximum value It is a figure which shows the histogram of each WDR synthetic | combination image after having performed.

  Exemplary embodiments of the present invention will be described below in detail with reference to the accompanying drawings. In the present specification and drawings, components having substantially the same functional configuration are denoted by the same reference numerals, and redundant description is omitted.

  In the present specification and drawings, a plurality of constituent elements having substantially the same functional configuration may be distinguished by attaching different alphabets after the same reference numeral. However, when it is not necessary to particularly distinguish each of a plurality of constituent elements having substantially the same functional configuration, only the same reference numerals are given.

  First, color reproducibility when using a range near saturation will be described with reference to FIG. FIG. 1 is a diagram for explaining color reproducibility when a range near saturation is used. In FIG. 1, each RGB pixel value representing a blue object is shown as an 8-bit value for simplicity. As shown in FIG. 1, when the “short exposure image” has a value of (R, G, B) = (96, 144, 192), the photographed object is a blue object having very high brightness. is there. Specifically, the clear blue sky is a typical example of a blue object.

  After performing WDR composition on the short-exposure image and long-exposure image, and performing range compression in accordance with the output range on the WDR composite image using conventional technology, the “output image” shown in FIG. Is output. If the “short exposure image” and the “output image” in FIG. 1 are expressed and compared in the HSV color system, only the V (brightness) is different between the “short exposure image” and the “output image”. Since H (hue) and S (saturation) are not different and are not saturated, there is no problem at first glance in the “output image”.

  However, when the color of the “output image” is observed, the sky reflected in the “output image” does not look bright blue but looks bright light blue, and the impression of the actual image is very different. As an example, even if a WDR composite image is generated after capturing a high-brightness / high-saturation object with a “short-exposure image”, deep colors cannot be reproduced in the “output image” unless the range compression is performed properly. There is a problem that is different from the real thing.

  Therefore, in this specification, when performing range compression on an input image, not only does the output image not be saturated beyond the output range, but also color reproducibility near the saturation level is insufficient. Pay attention to. Specifically, in the present specification, a range-compressed image with improved color reproducibility of a high-brightness / high-saturation object is obtained by performing control so as not to use a certain range from the upper limit of the output range. Propose a possible technology. A case where input RGB signals are used as an example of the input image will be described.

  Next, embodiments of the present invention will be described in detail. First, a functional configuration of the image processing apparatus 1 according to the embodiment of the present invention will be described. FIG. 2 is a diagram illustrating a functional configuration example of the image processing apparatus 1 according to the embodiment of the present invention. As shown in FIG. 2, the image processing apparatus 1 includes a sensor 10, a frame memory 20, a synthesis unit 30, and a range compression unit 40. Hereinafter, the function of each functional block included in the image processing apparatus 1 will be described in detail sequentially.

  The image processing apparatus 1 changes the exposure setting of the sensor 10 and continuously shoots two images. Here, it is assumed that short exposure shooting is performed first and then long exposure shooting is performed. However, long exposure photography may be performed first and then short exposure photography. The short-exposure image and the long-exposure image photographed in this way are written in the frame memory 20 as a pair. Shooting of the long exposure image and the short exposure image and writing of the captured long exposure image and the short exposure image to the frame memory 20 are continuously performed.

  Here, in the present specification, it is sufficient that at least the long exposure image at the current time and the short exposure image at the current time remain in the frame memory 20. Therefore, the interval at which the long exposure image and the short exposure image are taken is not particularly limited. Further, each frame of the long exposure image and the short exposure image may be written in the frame memory 20, or may be written in the frame memory 20 once every a plurality of frames.

  For example, when the frame memory 20 has an area in which a pair of a long exposure image and a short exposure image is written, as in the example illustrated in FIG. 2, the image processing apparatus 1 sets the pair of the long exposure image and the short exposure image. Can be written continuously to the area. For example, the short exposure image at the previous time and the long exposure image at the previous time written in the frame memory 20 may be overwritten by the short exposure image at the current time and the long exposure image at the current time.

  In the example shown in FIG. 2, the image processing apparatus 1 has one common system for outputting the long exposure image and the short exposure image, and the sensor 10 sometimes outputs the long exposure image and the short exposure image. Although the output is divided, the long exposure image and the short exposure image may be output simultaneously. In such a case, the image processing apparatus 1 may have two systems, that is, a system for outputting a long exposure image from the sensor 10 and a system for outputting a short exposure image. Each shutter time is determined by, for example, a dynamic range of a subject to be imaged, a sensor specification, and the like.

  In the embodiments of the present invention, the terms short exposure image and long exposure image are used, but these terms do not limit the absolute exposure time of each of the two captured images. Therefore, when two images with different exposure times are taken, an image with a relatively short exposure time corresponds to a short exposure image and an image with a relatively long exposure time is a long exposure. Corresponds to an image. For example, the long exposure time (hereinafter also simply referred to as “exposure ratio”) relative to the short exposure time may be set from several times to several tens of times.

  The combining unit 30 generates a WDR combined image by combining the long exposure image and the short exposure image. Specifically, the compositing unit 30 refers to the use image selection information, uses the short exposure image in the short exposure image use region, and uses the long exposure image in the long exposure image use region to generate the composite image. Generate. For example, even if the number of bits of each of the short exposure image and the long exposure image output from the sensor 10 is 12 bits, the number of bits of the WDR composite image generated by the combining unit 30 may be about 16 bits.

  Note that the use image selection information may be generated in any way. For example, an area that has been saturated in the long exposure image is highly likely not to be saturated in the short exposure image. A short exposure image may be selected as a use image. On the other hand, for example, for a region that is not saturated in the long exposure image, the long exposure image may be selected as the use image of the region.

  The range compression unit 40 performs range compression on the number of bits of each pixel value constituting the WDR composite image from A bits to B bits (A> B). As such compression processing, tone mapping according to a look-up table (LUT) may be used, but any method may be used. For example, even if the number of bits of the WDR synthesized image generated by the synthesis unit 30 is about 16 bits, the range compression unit 40 performs compression so that the number of bits of the WDR synthesized image becomes 12 bits.

  The subsequent stage of the range compression unit 40 is connected to an image processing engine including, for example, a demosaic unit that generates an RGB plane from Bayer data, a contour enhancement unit, and color management. Therefore, it is preferable that the data amount of the output signal from the range compression unit 40 be adjusted so as to match the size of the input data to the image processing engine (for example, about 12 bits). If the data size is simply reduced, the image is converted into a dark image. Therefore, it is preferable that the high luminance side is strongly compressed so as to approximate human visual characteristics.

  In the example described above, an example in which two types of images (short exposure image and long exposure image) having different exposure amounts are captured and combined has been described. However, the type of image used for composition is not particularly limited. For example, the present embodiment can be applied to an example in which three or more types of images having different exposure amounts are captured and combined. When three or more types of images are captured and combined, it is expected that the dynamic range of the obtained WDR combined image can be expanded to 20 bits or more. However, the range after compression by the range compressor 40 should be constant.

  Next, a detailed functional configuration example of the range compression unit 40 will be described with reference to FIG. FIG. 3 is a block diagram illustrating a detailed functional configuration example of the range compression unit 40. As illustrated in FIG. 3, the image processing apparatus 1 includes a maximum value detection unit 41, a margin determination unit 42, an addition unit 43, a coefficient calculation unit 44, and a compression unit 45. In the following description, the case where the input RGB signal to the range compression unit 40 is a WDR synthesized image generated by the synthesis unit 30 will be mainly described. However, the input RGB signal to the range compression unit 40 is converted to a WDR synthesized image. It is not limited.

  The maximum value detection unit 41 detects a maximum value hmax of pixel values constituting the input image. For example, as shown in FIG. 3, when an input RGB signal with an extended dynamic range is input, the maximum value detection unit 41 detects the maximum value hmax from all the pixel values constituting the input RGB signal. It's okay.

  Here, it is desirable that the maximum value detection unit 41 takes a histogram of all the pixel values constituting the input RGB signal, and detects the largest pixel value among the pixel values existing in a predetermined number or more as the maximum value. If even one pixel is detected as the maximum value, a defective pixel may be detected, or a very bright object that is very small and less important than the entire image may be detected. This is because the possibility can be reduced.

  Further, when taking a histogram of the input RGB signal, it is more desirable to take the histogram after quantizing the gradation. This is because, when the input RGB signal is a WDR composite image, if a histogram of all gradations is taken, the memory used for the histogram becomes very large. The maximum value hmax detected by the maximum value detection unit 41 is output to the margin determination unit 42 and the addition unit 43.

  The margin determination unit 42 determines the margin value margin based on the maximum value hmax detected by the maximum value detection unit 41. Details of the determination of the margin value margin by the margin determination unit 42 will be described later. Further, the adding unit 43 calculates an added value hmax ′ between the maximum value hmax detected by the maximum value detecting unit 41 and the margin value margin determined by the margin determining unit 42.

  The coefficient calculation unit 44 calculates a coefficient coeff based on the added value hmax ′. For example, the coefficient calculation unit 44 calculates the coefficient coeff by dividing the upper limit value of the output range by the added value hmax ′. Then, the coefficient coeff for making the added value hmax 'coincide with the upper limit value of the output range can be calculated.

  The compression unit 45 performs range compression on the input RGB signal using the coefficient coeff. The compression unit 45 can compress the input RGB signal by multiplying each pixel value of the input RGB signal by a coefficient coeff.

  In this way, by performing control so as not to use a certain range from the upper limit of the output range, it is possible to obtain a range-compressed image with improved color reproducibility of an object with high luminance and high saturation. Hereinafter, details of the determination of the margin value margin by the margin determination unit 42 will be described. FIG. 4 is a diagram for explaining an example of determining the margin value margin. FIG. 4 shows an example of the correspondence relationship between the maximum value hmax and the margin value margin.

  For example, the margin determination unit 42 may determine the margin value margin based on the comparison result between the saturation value VALu of the pixel values constituting the input RGB signal and the maximum value hmax. For example, as illustrated in FIG. 4, the margin determination unit 42 may determine zero as the margin value margin when the maximum value hmax is equal to or greater than the saturation value VALu.

  Here, when the input RGB signal is a WDR composite image of a short exposure image and a long exposure image, the saturation value VALu may be any value that can be taken in the saturation state of the short exposure image. Specifically, the saturation value VALu may be a theoretical upper limit value of the pixel value of the WDR composite image. For example, as shown in the following formula (1), the saturation value VALu is the number of bits n of each pixel value output from the sensor 10 and the exposure ratio Exp of the long exposure image to the short exposure image captured by the sensor 10. It may be theoretically calculated by the multiplication value.

VALu = 2 ⁿ × Exp (1)

When the bit number n of each pixel value output from the sensor 10 is 12, and the exposure ratio of the long exposure image to the short exposure image is 10 times, the theoretical upper limit value of the pixel value of the WDR composite image is 2 ¹² × 10 = 40960, and it is basically considered that a pixel value larger than the upper limit value is present in the WDR synthesized image.

  Alternatively, the saturation value VALu may be a saturation value of pixel values constituting a short exposure image. For example, in such a case, the margin determination unit 42 may determine the margin value margin based on the comparison result between the saturation value of the pixel value constituting the short exposure image and the maximum value of the pixel value constituting the short exposure image. The comparison between the saturation value of the pixel value constituting the short exposure image and the maximum value of the pixel value constituting the short exposure image may be performed by a comparison unit (not shown).

  For example, the margin determination unit 42 may determine the margin value margin based on a comparison result between a predetermined threshold value TH and the maximum value hmax. For example, as illustrated in FIG. 4, the margin determination unit 42 may determine the specified value as the margin value margin when the maximum value hmax is lower than the threshold value TH.

  Further, as shown in FIG. 4, when the maximum value hmax is between the saturation value VALu and the threshold value TH, the margin determination unit 42 may determine a value calculated by linear interpolation as the margin value margin. Good.

  Here, the description is made using the exposure ratio of the long exposure image to the short exposure image. However, when using three or more types of images taken with different exposure amounts, the longest exposure image with respect to the shortest exposure image is used. The VALu may be calculated using the exposure ratio.

  Subsequently, an example of an effect produced by the embodiment of the present invention will be described. First, a case where the maximum pixel value hmax has not reached the saturation value VALu will be described. FIG. 5 shows a WDR composite image (upper stage) in which the maximum pixel value hmax has not reached the saturation value VALu, a WDR composite image (middle stage) after compression into a 12-bit range using the maximum value hmax, and a maximum value hmax. It is a figure which shows the histogram of each WDR synthetic | combination image (lower stage) after compressing to 12 bit range using added value hmax 'with margin value margin.

  Referring to the WDR composite image after compression using the maximum value hmax (middle of FIG. 5), the output range is fully used up, but as described in the background section, the high luminance and high saturation portion The color reproducibility will be insufficient.

  On the other hand, referring to the WDR composite image (lower part of FIG. 5) after compression into the 12-bit range using the added value of the maximum value hmax and the margin value margin, the vicinity of the saturation of the output range is not used. The color reproducibility of high saturation objects is improved. However, if the margin value margin is too large, the number of unused gradations on the highlight side of the output range increases, and the contrast of the output image may decrease, or the image may appear dark. Therefore, it is preferable to set a margin value that balances the color reproducibility of an object with high luminance and high saturation and the contrast of an image.

  Next, a case where the maximum pixel value hmax reaches the saturation value VALu will be described. FIG. 7 shows a WDR composite image in which the maximum pixel value reaches the saturation value (upper stage), and a WDR composite image after compression into a 12-bit range using the added value hmax ′ of the maximum value hmax and the margin value margin. It is a figure which shows the histogram of each WDR synthetic | combination image (lower stage) after compressing to 12 bit range using (middle stage) and maximum value hmax.

  Referring to the WDR composite image in which the maximum pixel value reaches the saturation value (upper part of FIG. 6), the histogram of the highlight portion reaches the saturation value VALu of the WDR composite image, and hmax = VALu. . Note that the fact that the WDR composite image is saturated means that a very high-brightness object that is saturated even in a short-exposure image has been captured.

  Referring to the WDR composite image after compression to the 12-bit range using the added value hmax ′ of the maximum value hmax and the margin value margin (the middle part of FIG. 6), the added value hmax ′ of the maximum value hmax and the margin value margin The range is compressed using, and the highlight part is output using a slightly smaller gradation from saturation. As a result, if only the short-exposure image is seen, it looks saturated and appears white or close to it, but when the synthesized image is seen, it looks like a light gray, not saturated, and looks unnatural.

  On the other hand, referring to the WDR composite image (lower part of FIG. 6) after compression into the 12-bit range using the maximum value hmax, the case where the WDR composite image is saturated is shown. In such a case, it is natural that the compressed WDR composite image is also saturated. That is, it is preferable to compress the WDR composite image using the maximum value hmax without using the added value hmax '.

  As in the example of the effect described above, in the embodiment of the present invention, the maximum value hmax is compared with the saturation value VALu, and when the maximum value hmax reaches the saturation value VALu, the margin value margin is set. Introduce controls such as zero. Accordingly, consideration is given so that a natural result can be obtained even when the WDR composite image is saturated.

  The embodiment of the present invention has been described above. According to the embodiment of the present invention, in the image processing apparatus 1 that performs range compression on the number of bits of each pixel value constituting an input image from A bits to B bits (A> B), the pixel values constituting the input image A maximum value detecting unit 41 for detecting a maximum value of the image, a margin determining unit 42 for determining a margin value based on the maximum value, a coefficient calculating unit 44 for calculating a coefficient based on an added value of the maximum value and the margin value, The image processing apparatus 1 includes a compression unit 45 that performs range compression on an input image using a coefficient.

  According to such a configuration, it is possible to obtain a range-compressed image in which the color reproducibility of a high-brightness / high-saturation object is improved by performing control so as not to use a certain range from the upper limit of the output range. .

  The preferred embodiments of the present invention have been described in detail above with reference to the accompanying drawings, but the present invention is not limited to such examples. It is obvious that a person having ordinary knowledge in the technical field to which the present invention pertains can come up with various changes or modifications within the scope of the technical idea described in the claims. Of course, it is understood that these also belong to the technical scope of the present invention.

  Wide dynamic range technology is particularly important as a network camera differentiation technology. Embodiments of the present invention enable high brightness / saturation objects, which were insufficient in color reproducibility in the prior art, to be reproduced with colors that are closer to real objects, and side effects are suppressed by adaptive control of margin values. Yes. Therefore, it is excellent in that the image quality can be stably improved, the processing is not complicated and can be realized with a realistic circuit scale, and there are few places that need to be controlled.

DESCRIPTION OF SYMBOLS 1 Image processing apparatus 10 Sensor 20 Frame memory 30 Synthesis | combination part 40 Range compression part 41 Maximum value detection part 42 Margin determination part 43 Addition part 44 Coefficient calculation part 45 Compression part coeff Coefficient Exp Exposure ratio hmax Maximum value hmax 'Addition value margin Margin value TH threshold VALu saturation value

Claims (9)

In an image processing apparatus that performs range compression on the number of bits of each pixel value constituting an input image from A bits to B bits (where A> B),
A maximum value detecting unit for detecting a maximum value of pixel values constituting the input image;
A margin determining unit that determines a margin value based on the maximum value;
A coefficient calculation unit that calculates a coefficient based on an added value of the maximum value and the margin value;
A compression unit that performs range compression on the input image using the coefficient;
An image processing apparatus comprising: The margin determining unit determines the margin value based on a comparison result between a saturation value of a pixel value constituting the input image and the maximum value;
The image processing apparatus according to claim 1. The margin determining unit determines the margin value based on a comparison result between a predetermined threshold and the maximum value;
The image processing apparatus according to claim 2. The margin determining unit determines zero as the margin value when the maximum value is equal to or greater than the saturation value;
The image processing apparatus according to claim 3. The margin determining unit determines a specified value as the margin value when the maximum value is less than the threshold;
The image processing apparatus according to claim 4. The margin determining unit determines, as the margin value, a value calculated by linear interpolation when the maximum value is between the saturation value and the threshold;
The image processing apparatus according to claim 5. The input image is a composite image of a short exposure image and a long exposure image,
The saturation value is a multiplication value of the number of bits of each pixel value output from a sensor and the exposure ratio of the short exposure image and the long exposure image captured by the sensor.
The image processing apparatus according to claim 2. The input image is a composite image of a short exposure image and a long exposure image,
The margin determining unit determines the margin value based on a comparison result between a saturation value of a pixel value constituting the short exposure image and a maximum value of a pixel value constituting the short exposure image;
The image processing apparatus according to claim 2. In an image processing method in which the number of bits of each pixel value constituting an input image is range-compressed from A bit to B bit (A> B),
Detecting a maximum value of pixel values constituting the input image;
Determining a margin value based on the maximum value;
Calculating a coefficient based on an added value of the maximum value and the margin value;
Range compressing the input image using the coefficients;
Including an image processing method.

Images