JP2020156016A - Image processing system, image processing method, and program - Google Patents

Abstract

To provide an image processing system capable of performing brightness correction while suppressing the influence of color bending and increasing the gradation of brightness.SOLUTION: An image processing system (101) includes: processing means (102) that performs a series of white balance processing on the input image signal; correction means (105) that performs a series of luminance correction processing on the high-luminance region of the image signal after the white balance processing by the processing means (102); and gradation processing means (106) that performs a series of gradation processing to expand the luminance gradation in the high luminance region of the image signal after the correction processing by the correction means (105).SELECTED DRAWING: Figure 1

Description

The present invention relates to an image processing technique for correcting an image.

Conventionally, various correction processes have been devised so that a good image can be obtained in a high-luminance region of an image captured by an imaging device. As such a process, there is a shine correction process for correcting the brightness value in the human face image area and the area having high brightness. Further, a high-luminance gradation improvement process for correcting a gradation that is lost when a pixel value is saturated in a high-luminance region is also known.
For example, in Patent Document 1, a region brighter than a predetermined region is detected in a human face image region, and this region (hereinafter referred to as a shiny region) is painted so as to gradually approach the surrounding color to reduce the brightness value. The shine correction process is disclosed. Further, in Patent Document 2, the saturation level of the captured image data for each color signal is calculated, the input upper limit value is determined, and the replacement ratio that changes toward the saturation level of the color signal is obtained according to the input upper limit value. The technology is disclosed. Then, in Patent Document 2, each of the color signals is adjusted according to the input upper limit value, and the color signal having a high saturation level is replaced with the color signal having a low saturation level based on the replacement ratio, so that the gradation of high brightness is obtained. It is possible to perform high-luminance gradation improvement processing that expands.

Japanese Unexamined Patent Publication No. 2013-161348 Japanese Unexamined Patent Publication No. 2015-156616

By the way, conventionally, there has been a demand for expanding the gradation of a high-luminance region while performing a shine correction for the shine region. However, when both processes are performed, a sufficient effect may not be obtained depending on the content of the process.
For example, if the high-luminance gradation improvement process is used to improve the high-luminance gradation and then the shine correction is performed, it becomes difficult to detect the skin color region due to the influence of the color signal replacement process. That is, in the high-luminance gradation improvement process, the color signal having a high saturation level is replaced with the color signal having a low saturation level to maintain the wide gradation of the color signal having a high saturation level, and the color of the high-luminance portion is white. This prevents unnatural coloring from occurring. However, in a region such as a shiny region where the brightness value continuously increases while maintaining the hue of the skin color region, the region that was originally flesh-colored becomes a different hue due to the color replacement process, so-called color. May cause bending. Therefore, when the shine correction process is performed after the high-luminance gradation improvement process, it is necessary to make a correction in consideration of color bending, and it may be difficult to generate a complementary color of the skin color for the shine correction. In addition, due to this color bending, the skin color region may be erroneously detected.

Further, if the high-luminance gradation improvement processing is performed after the shine correction, the signal having a high saturation level is clip-processed, which makes it difficult to improve the gradation. That is, in the shine correction, since it is necessary to detect a region that is a skin color in a human face image region, it is necessary to perform white balance processing and clip processing a signal having a saturation level or higher that causes unnecessary coloring. Therefore, when the high-luminance gradation improvement process is subsequently performed, the signal having a high saturation level is lost by the clip process, so that the luminance gradation cannot be expanded.

Therefore, an object of the present invention is to be able to perform luminance correction while suppressing the influence of color bending to a small extent, and to make it possible to enlarge the gradation of luminance.

The image processing apparatus of the present invention has a processing means for performing white balance processing on an input image signal and a high-luminance region of the image signal after the white balance processing is performed by the processing means. It has a correction means for performing correction processing and a gradation processing means for performing gradation processing for enlarging the luminance gradation in a high-luminance region of the image signal after the correction processing is performed by the correction means. It is characterized by.

According to the present invention, it is possible to perform luminance correction while suppressing the influence of color bending to a small extent, and to expand the luminance gradation.

It is an overall block diagram of the 1st Embodiment. It is a flowchart of the process of 1st Embodiment. It is a figure which showed the graph of each signal value which concerns on 1st Embodiment. It is an overall block diagram of the 2nd Embodiment. It is a flowchart of the process of 2nd Embodiment. It is a figure which showed the graph of each signal value which concerns on 2nd Embodiment. It is an overall block diagram of the 3rd Embodiment. It is a flowchart of the process of 3rd Embodiment. It is a figure which showed the graph of each signal value which concerns on the shine correction processing. It is a figure which showed the graph of each signal value which concerns on high-luminance gradation improvement processing.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. The configuration shown in the following embodiments is only an example, and the present invention is not limited to the illustrated configuration.
<First Embodiment>
The image processing method and apparatus according to the first embodiment of the present invention will be described with reference to FIGS. 1 to 3, and the effects of the present embodiment will be described while comparing with conventional examples. In the following description, an example is given in which the image processing apparatus according to the present embodiment is implemented by a circuit configuration, but a CPU or the like executes a program for a part or all of the processes performed in each circuit described later. May be carried out by.

FIG. 1 is a diagram showing a configuration example of the image processing device 101 of the present embodiment that corrects a high-luminance region. An image signal captured by an image pickup device such as a digital camera (not shown) is input to the image processing device 101. The input image signal may be an image signal read from a recording medium or a recording device, or may be an image signal transmitted via a network or the like.

As shown in FIG. 1, the image processing device 101 includes circuits of a WB processing unit 102, a clip processing unit 103, a correction amount determination unit 104, a correction processing unit 105, and a gradation processing unit 106. There is.
The WB processing unit 102 multiplies the R signal and the B signal of the image signal including the R (red) signal, the G (green) signal, and the B (blue) signal by a predetermined white balance gain value to obtain white balance. Perform white balance processing to adjust. Hereinafter, the white balance will be referred to as WB.
The clip processing unit 103 performs a process of clipping a signal value having a value equal to or more than a predetermined value to a predetermined value for each signal of the R signal, the G signal, and the B signal.

The correction amount determination unit 104 determines an area in the image to be subjected to the shine correction process and an amount to be corrected for the area. In the case of the present embodiment, the correction amount determination unit 104 outputs a complementary color signal of a specific color region required for shine correction, for example, a flesh color complementary color signal, as a shine correction amount.
The correction processing unit 105 performs the shine correction process by subtracting the shine correction amount from the image signal after the WB process.
The gradation processing unit 106 performs high-luminance gradation improvement processing for replacing a color signal having a high saturation level with a color signal having a low saturation level.
The details of the processing in these configurations will be described later. Further, in the following description, when the R signal, the G signal and the B signal are not distinguished, they are collectively referred to as an RGB signal.

Next, the procedure of image processing and the details of image processing according to the present embodiment will be described with reference to the flowchart of FIG. 2 and the graphs of the signal values shown in FIGS. 3A to 3E. In each of the graphs shown in FIGS. 3A to 3E, the horizontal axis represents the amount of light incident on the image sensor, and the vertical axis represents the signal intensity of the input image signal or the image signal after each processing. In the graphs of FIGS. 3A to 3E, unless otherwise specified, the amount of light incident on the image sensor increases from right to left on the horizontal axis, and increases from bottom to top on the vertical axis. It shows that each signal strength increases. Unless otherwise specified, in the graphs of FIGS. 3A to 3E, the broken line represents the R signal, the solid line represents the G signal, and the alternate long and short dash line represents the B signal.

In the flowchart of FIG. 2, step S202 is an input image data acquisition step. In step S202, the image processing device 101 of FIG. 1 acquires, for example, an image signal output from an image sensor of an image pickup device. Here, the input image signal is a signal in a state before the WB processing is performed. Further, in the following description, the signal value of the R signal of the input image signal is expressed as Rin, and similarly, the signal value of the G signal is expressed as Gin and the signal value of the B signal is expressed as Bin. Further, it is assumed that the signal values of the RGB signals in the input image signals are, for example, R signal 301, G signal 302, and B signal 303 as shown in FIG. 3A. That is, in the G signal 302, the signal value Gin is the sensor saturation level 304 in the section 305 and the image sensor is saturated, but the signal value Rin of the R signal 301 and the signal value Bin of the B signal 303 are Suppose it is not saturated.

Next, step S203 is a WB processing step. In step S203, the WB processing unit 102 performs WB processing for applying WB gain to the input image signal. The WB process is a commonly known WB process. At this time, the WB processing unit 102 determines the WB gain Wr for the R signal 301 and the WB gain Wb for the B signal 303, multiplies the R signal 301 by the WB gain Wr, and WB gains the B signal 303. Multiply Wb.

FIG. 3B is a graph showing an RGB signal after WB processing is performed on the RGB signal shown in FIG. 3A. That is, as a result of the WB processing, the R signal 301 becomes the R signal 306 whose signal value is represented by Rin * Wr, and the B signal 303 becomes the B signal 307 whose signal value is represented by Bin * Wb. On the other hand, the G signal 302 remains a signal whose signal value is represented by Gin. Further, in FIG. 3B, the section 308 is a section in which any signal value of the RGB signal after the WB processing is equal to or higher than the sensor saturation level. In the present embodiment, this section 308 is a section that is the target of the high-luminance gradation improvement processing described later. Further, the section 309 is a section having high brightness and skin color, and in the present embodiment, it is a section corresponding to the skin color region to be subject to the shine correction described later.

The R signal 306, the G signal 302, and the B signal 307 after the WB processing in step S203 can be sent to the correction processing unit 105 and the clip processing unit 103 in FIG.
Step S204 of FIG. 2 is a clip processing step. The clip processing unit 103 performs clip processing in step S204. The RGB signal after the clip processing is sent to the correction amount determination unit 104 of FIG.

Step S205 in FIG. 2 is a shine correction amount determination step. The correction amount determination unit 104 determines the shine correction amount in step S205. The process of determining the amount of shine correction is a commonly used process. The correction amount determination unit 104 determines the shine correction amount by, for example, a process as described in FIGS. 9 (c) to 9 (f) described later. In the present embodiment, the correction amount determination unit 104 determines the shine correction amount Tr for the R signal, similarly, the shine correction amount Tg for the G signal and the shine correction amount for the B signal. Determine Tb. In FIG. 9 (f) described later, the shine correction amount 917 is the shine correction amount Tr for the R signal, the shine correction amount 918 is the shine correction amount Tg for the G signal, and the shine correction amount 919 is the shine correction amount Tb for the B signal. Represents. The signals indicating the shiny correction amounts Tr, Tg, and Tb determined by the correction amount determination unit 104 in this way are sent to the correction processing unit 105.

Step S206 in FIG. 2 is a shiny correction processing step. In step S206, the correction processing unit 105 uses the shine correction amount determined in step 205 to perform shine correction processing on the image signal after WB processing in step S203. In the present embodiment, the correction processing unit 105 subtracts the shine correction amount Tr from the R signal 306 after the WB processing, and similarly, subtracts the shine correction amount Tg from the G signal 302 after the WB processing after the WB processing. The shine correction amount Tb is subtracted from the B signal 307. That is, the correction processing unit 105 performs a process of subtracting the shiny correction amount determined in step S205 from the image signal that is the signal after the WB gain is applied in step S203 and has not been clipped. Do.

FIG. 3C is a graph showing the RGB signal after the shine correction processing. As a result of the shine correction processing, the R signal 306 becomes the R signal 310 whose signal value is represented by Rin * Wr-Tr. Similarly, the G signal 302 becomes the G signal 311 whose signal value is represented by Gin-Tg, and the B signal 307 becomes the B signal 312 whose signal value is represented by Bin * Wb-Tb. That is, the R signal 310 is a signal obtained by subtracting the shine correction amount 917 of FIG. 9 (f) from the R signal 306 of FIG. 3 (b), and similarly, the G signal 311 is a signal obtained by subtracting the shine correction amount 918 from the G signal 302. , B signal 312 is a signal obtained by subtracting the shine correction amount 919 from the B signal 307. Further, the section 313 in FIG. 3C is a section in which the shine correction is performed. The signal after the shine correction processing is performed by the correction processing unit 105 in this way is sent to the gradation processing unit 106.

Step S207 in FIG. 2 is a high-luminance gradation improvement step. In step S207, the gradation processing unit 106 performs high-luminance gradation improvement processing on the RGB signal after the shine correction processing in step S206. In the case of the present embodiment, the gradation processing unit 106 replaces the G signal and the B signal having the lowest saturation level with the signal (R signal) having the highest saturation level among the signals after the shine correction processing in step S206. Performs high-brightness gradation improvement processing. After this step S207, the image processing device 101 ends the processing of the flowchart of FIG.

FIG. 3D is a graph showing a state of high-luminance gradation improvement processing for an RGB signal. In the case of the present embodiment, as described above, the gradation processing unit 106 has the G signals 311 and B having a low saturation level due to the signal value of the R signal 310 having the highest saturation level among the RGB signals after the shine correction processing. High-luminance gradation improvement processing that replaces the signal value of signal 312 is performed. That is, the G signal 311 becomes the G signal 314 by replacing the signal value with the R signal 310, and the B signal 312 becomes the B signal 315 by replacing the signal value with the R signal 310. The R signal 310 is the same signal. Further, the section 316 of FIG. 3D is a section corresponding to the section 309 of FIG. 3B, which is a high-luminance and flesh-colored section and is a flesh-colored section subject to shine correction. Further, the section 317 is a section in which the signal value replacement processing is performed on the G signal and the B signal, and corresponds to the section 308 in FIG. 3 (b). Further, the section 318 is a section other than the section 317, and is a section in which the signal value replacement process is not performed. Further, the section 319 and the section 320 are sections obtained by dividing the section 316, and in particular, the section 320 is a section in which the signal value is completely replaced.

Here, the replacement process in the high-luminance gradation improvement process is performed using a replacement ratio table as described with reference to FIG. 10 (c) described later. That is, in the replacement ratio table, the replacement ratio is 0.0 in the section corresponding to the section 318, the replacement rate increases in the section corresponding to the section 319, and the replacement rate is 1.0 in the section corresponding to the section 320. It is made like a table. By using this replacement ratio table, in the section 318, the G signal 314 has the same signal value as the G signal 311 in FIG. 3C, and similarly, the B signal 315 has the same signal value as the B signal 312. Will remain. Further, the G signal 314 gradually increases in the ratio of being replaced with the signal value of the R signal 310 in the section 319, and becomes a signal completely replaced with the signal value of the R signal 310 in the section 320. Similarly, the B signal 315 gradually increases in the ratio of being replaced with the signal value of the R signal 310 in the section 319, and becomes a signal completely replaced with the signal value of the R signal 310 in the section 320. On the other hand, the section 316 is included in the section 318 in which the replacement process is not performed. Therefore, in this section 316, the effect of the shine correction process performed in FIG. 3C is maintained.

FIG. 3 (e) is a graph obtained by converting the RGB signal of FIG. 3 (d) into a luminance signal. In FIG. 3E, the solid line represents the luminance signal 321. That is, the luminance signal 321 is a signal in which the luminance value of the sensor saturation level 304 or higher is generated and the gradation is expanded, and the luminance value is lowered in the section 316, so that the effect of the shine correction processing is maintained. It becomes a signal of the state.

As described above, the image processing apparatus 101 of the present embodiment branches the signal after the WB processing, and on the one hand, performs lip processing to obtain the shine correction region and the shine correction amount, and on the other hand, the signal without the clip processing is shine. Shine correction is performed using the correction amount. After that, the image processing device 101 performs high-luminance gradation improvement processing. That is, according to the image processing apparatus 101 of the first embodiment, both the shine correction processing and the high-luminance gradation improvement processing can be applied to the same high-luminance region.

<Comparison between image processing in the first embodiment and conventional image processing>
Here, the processing result of the conventional image processing apparatus is compared with the effect of the image processing apparatus of the present embodiment.
First, the shine correction process in the conventional image processing apparatus will be described, and then the high-luminance gradation improvement process will be described.
In the shine correction process, for example, a skin color region in a human face region and a high-luminance region is specified as a shine region, and the shine correction process is performed on the shine region. At this time, if only the inside of the shiny region is corrected, the difference in brightness and color becomes conspicuous at the boundary between the shiny region and another region, which makes it unnatural. Therefore, for example, as disclosed in Patent Document 1, in the shine correction process, the color of the shine region is gradually painted closer to the surrounding color from the brightest region of the shine region toward the periphery. Processing is done. Therefore, when performing shine correction, it is necessary to detect a high-luminance region in the skin color region of a human face image and correctly obtain the skin color at the boundary.

9 (a) to 9 (g) are graphs showing changes in RGB signals when shine correction processing is performed by a conventional method. In each of the graphs of FIGS. 9A to 9G, the horizontal axis represents the amount of incident light on the image sensor, and the vertical axis represents the signal intensity of the input image signal or the image signal after each processing. Further, in FIGS. 9 (a) to 9 (g), unless otherwise specified, the amount of incident light on the image sensor increases from right to left on the horizontal axis, and each increases from bottom to top on the vertical axis. This indicates that the signal strength increases. Unless otherwise specified, in FIGS. 9 (a) to 9 (g), the broken line represents the R signal, the solid line represents the G signal, and the alternate long and short dash line represents the B signal.

FIG. 9A is a graph showing the signal values of the R signal 901, the G signal 902, and the B signal 903 output from the image sensor, and the same example as in FIG. 3A is shown. Further, as in FIG. 3A, the signal value of the G signal 902 is the sensor saturation level 904 in the section 905, and the image sensor is saturated, but the signals of the R signal 901 and the B signal 903 are generated. The value is not saturated.

The conventional image processing device corrects the difference in the sensor sensitivity of the image sensor for these RGB signals, and performs WB processing so that the white image becomes a white signal.
FIG. 9B is a graph showing an RGB signal after WB processing such as multiplying the R signal and the B signal by the WB gain is performed. In FIG. 9B, the R signal 906 is the R signal after the WB gain is applied, and the B signal 907 is the B signal after the WB gain is applied. Further, the section 908 of FIG. 9B is a section of a high brightness region in which the R signal 906 after the WB processing or the B signal 907 after the WB processing is the sensor saturation level 904 or higher. That is, the signal in this section 908 is in a state in which accurate color information cannot be obtained and so-called false color is generated. Therefore, normally, the signal in this section 908 is clipped to prevent false color in the high luminance region.

FIG. 9C is a graph showing an RGB signal after clipping processing is performed at a value corresponding to the sensor saturation level 904. That is, in FIG. 9C, the R signal 909 is the R signal after the clip processing, and the B signal 910 is the B signal after the clip processing. Since the G signal 902 does not originally have a signal value of the sensor saturation level 904 or higher, it is a signal as it is.

Further, in the shine correction process, a human face image is detected, the brightness value of the skin color area is examined, and when there is a high brightness area in this skin color area, the correction is performed so as to lower the brightness value of the high brightness area. Will be done. In the case of the example of FIG. 9C, the section 911 is a section corresponding to a high-luminance region in the skin color region, which is the target of the shine correction. Further, the section 912 is a high-intensity white region, which corresponds to a so-called overexposed region. In the shine correction process, the brightness value of the section 911 is lowered to correct the shine, but it is desirable to correct the brightness value of the whiteout area in order to maintain the continuity with the whiteout section 912.

Here, when obtaining the shiny correction amount, the conventional image processing apparatus obtains the luminance signal Y from the RGB signal after the clip processing as shown in FIG. 9C, for example. The luminance signal Y is obtained from the signal value of the RGB signal by, for example, the calculation of the equation (1). In addition, Rc in the formula (1) shows the signal value of the R signal after the clip processing, similarly, Gc is the signal value of the G signal after the clip processing, and Bc is the signal value of the B signal after the clip processing. Shown.

Y = 0.299 * Rc + 0.587 * Gc + 0.114 * Bc formula (1)

FIG. 9D is a graph showing the luminance signal 913 obtained in this way.
Next, the image processing device obtains the luminance value Yt of the section 915 in which the luminance signal 913 becomes a constant value 914 or more. The luminance value Yt having a constant value of 914 or more is obtained by using the luminance signal Y obtained by the equation (1) and the equations (2) and (3) when the constant value 914 is represented by the threshold value th.

Yt = 0 (Y <th) Equation (2)
Yt = Y-th (Y ≧ th) Equation (3)

FIG. 9E is a graph showing a signal 916 having a luminance value Yt in a section 915 in which the luminance signal 913 in FIG. 9D is equal to or higher than a certain value (threshold value th).
Next, the conventional image processing apparatus obtains the complementary color of the skin color, which is proportional to the brightness value Yt of the signal 916, and obtains the shine correction amount.
Here, it is assumed that the ratio of the RGB signal values of the complementary colors to the skin color of the section 911 in FIG. 9C is R: G: B = a: b: c. In this case, the shine correction amount Tr for the R signal is obtained by the equation (4), similarly, the shine correction amount Tg for the G signal is obtained by the equation (5), and the shine correction amount Tb for the B signal is obtained by the equation (6).

Tr = a * Yt equation (4)
Tg = b * Yt equation (5)
Tb = c * Yt equation (6)

FIG. 9 (f) is a graph showing the amount of shine correction for the RGB signal thus obtained. In FIG. 9 (f), the shine correction amount 917 indicates Tr, the shine correction amount 918 indicates Tg, and the shine correction amount 919 indicates Tg. These shine correction amounts are required to be proportional to the brightness value of the signal 916 having a constant value of 914 or more. That is, the section 915 is a section in which the shine correction amount exceeds the 0 value. Therefore, the shine correction process is performed on the RGB signal in this section 915.

That is, the conventional image processing apparatus lowers the luminance value in the shiny region by subtracting the shiny correction amount as shown in FIG. 9 (f) from the RGB signal of FIG. 9 (c). FIG. 9 (g) shows an RGB signal after performing a shine correction process on the RGB signal of FIG. 9 (c) using the shine correction amount shown in FIG. 9 (f). In FIG. 9 (g), the R signal 920 shows the R signal after the shiny correction processing, similarly, the G signal 921 shows the G signal after the shiny correction processing, and the B signal 922 shows the B signal after the shiny correction processing. ing. Further, the section 923 is a correction section in which the shine correction process is performed. As a result of performing such a shine correction process, the RGB signal in the section 923 is in a state where the signal strength is lowered as a whole.

Next, the high-luminance gradation improvement processing in the conventional image processing apparatus will be described.
The conventional image processing apparatus replaces a color signal having a high saturation level with a color signal having a low saturation level with respect to the image signal after performing the WB processing as described above to expand the gradation in the high luminance region. High-luminance gradation improvement processing such as is performed.

10 (a), 10 (e), 10 (d), and 10 (e) are graphs showing changes in RGB signals when high-luminance gradation improvement processing is performed. In each of these graphs, the horizontal axis represents the amount of light incident on the image sensor, the vertical axis represents the signal intensity of the image signal, the dashed line represents the R signal, and the solid line represents the G signal. The chain line represents the B signal.

FIG. 10A is a graph showing the signal value of the RGB signal output from the image sensor, and shows the same example as in FIGS. 3A and 9A. In FIG. 10A, it is assumed that the R signal 1001, the G signal 1002, and the B signal 1003 are the same signals as the R signal 301, the G signal 302, and the B signal 303 in FIG. 3A. Further, similarly to the above, the signal value of the G signal 1002 is the sensor saturation level 1004 in the section 1005, and the image sensor is saturated, but the signal values of the R signal 1001 and the B signal 1003 are not saturated. ..

FIG. 10 (b) is a graph showing RGB signals after WB processing for multiplying R signal and B signal by WB gain is performed, as in the above-mentioned examples of FIGS. 3 (b) and 9 (b). Is. That is, in FIG. 10B, the R signal 1006 is the R signal after the WB gain is applied, the B signal 1007 is the B signal after the WB gain is applied, and the G signal 1002 is the same signal as the original. is there. Further, as in the above example, as a result of the WB processing, as shown in FIG. 10B, the signal values of the R signal 1006 and the B signal 1007 exceed the sensor saturation level 1004 in the section 1008. In particular, in this section 1008, the signal value of the R signal 1006 always greatly exceeds the sensor saturation level 1004.

Therefore, in the conventional image processing apparatus, in this section 1008, the signal having the highest saturation level (R signal) is used to replace the signal having the lowest saturation level (G signal, B signal), respectively. Luminance gradation improvement processing is performed. Note that the section 1009 is not subject to the high-luminance gradation improvement processing and the replacement processing is not performed because all the signal values of the RGB signals are small values below the sensor saturation level 1004.

The conventional image processing apparatus uses, for example, a replacement ratio table as shown in FIG. 10C when performing replacement processing for high-luminance gradation improvement processing. In FIG. 10 (c), the horizontal axis represents the signal strength, the vertical axis represents the substitution ratio, and the solid line represents the function value 1010 of the substitution characteristic. In the function value 1010 shown in FIG. 10C, the replacement ratio is set to 0.0 in the section 1013 having a signal strength value of 0 or more and a signal strength value of less than 1011. Further, the function value 1010 is set so that the replacement ratio gradually increases in the section 1014 having a signal strength value of 1011 or more and a constant signal strength value of less than 1012, and the replacement ratio becomes 1.0 when the signal strength value is 1012 or more. ing. For example, when the signal strength is x and the substitution ratio is y, the function ratio y is expressed by the following equation (7).

y = 0 (0 ≦ x <T1011) Equation (7)
y = min (((x−T1011) / (T1012-T1011) ^ 2,1.0) (T1011 ≦ x) Equation (8)

Here, T1011 in the formula (7) is the signal strength value 1011 in FIG. 10 (c), and T1012 is the constant signal strength value 1012 in FIG. 10 (c). In addition, min () is a function for finding the minimum value. That is, the substitution ratio y is 0.0 in the section 1013 where the signal strength x is 0 or more and less than T1011, changes significantly in the section 1014 where the signal strength x is T1011 or more and less than T1012, and 1 in the section 1015 where the signal strength x is T1012 or more. It becomes 0.0.

In the high-luminance gradation improvement processing, the replacement ratio y is used to replace the R signal having the highest saturation level as the replacement source signal and the G signal and the B signal having the lowest saturation level as the replacement target signals. Will be done. Here, the substitution ratio for the G signal is yg, and the substitution ratio for the B signal is yb. Further, the signal value of the G signal before replacement is set to G1, the signal value of the G signal before replacement is set to G2, similarly, the signal value of the B signal before replacement is B1 and the signal value of the B signal after replacement is B2. Further, the signal value of the replacement source R signal is R1. In this case, the replacement process is represented by the following equations (9) and (10).

G2 = G1 + (R1-G1) * yg equation (9)
B2 = B1 + (R1-B1) * yb equation (10)

As described above, since the replacement ratio is 0.0 in the section 1013, the signal to be replaced (G signal or B signal) is output as it is. On the other hand, in the section 1014, the signal value of the signal to be replaced starts to be replaced by the signal value of the signal of the replacement source. Since the replacement ratio is 1.0 in the section 1015, the signal value of the signal to be replaced (G signal or B signal) completely matches the signal value of the replacement source signal (R signal). Become.

The function value 1010 of the substitution characteristic can be obtained by setting the above-mentioned signal strength value 1011 and a constant signal strength value 1012, and can also be set for each color signal of the RGB signal. In the following, for the sake of simplicity, it is assumed that the G signal and the B signal are set to be replaced at the same position with the same replacement ratio.

Further, the signal strength value when obtaining the function value 1010 of the substitution characteristic can be set based on, for example, the signal value of the R signal 1006 in FIG. 10B and the sensor saturation level 1004. For example, a case where the signal value of the R signal 1006 is set to the signal strength value 1011 and the sensor saturation level 1004 is set to the constant signal strength value 1012 and a value slightly higher than the sensor saturation level 1004 is set will be described. Here, the section 1009 in FIG. 10B is a section in which the signal value of the R signal 1006 is smaller than the sensor saturation level 1004. Therefore, the replacement ratio in the section 1009 is set to 0.0 corresponding to the range of the section 1013 in the replacement rate table of FIG. 10C. Further, the section 1008 in FIG. 10B is a section in which the signal value of the R signal 1006 is the sensor saturation level 1004 or higher. Therefore, the replacement ratio in the section 1008 corresponding to the combined range of the section 1014 and the section 1015 in the replacement ratio table of FIG. 10C is a value exceeding 0.0. That is, this section 1008 is a section in which the replacement process is performed.

FIG. 10 (d) shows how the G signal and the B signal are replaced by using the function value 1010 of the replacement characteristic. In FIG. 10D, the G signal 1016 is the G signal after the replacement process, and the B signal 1017 is the B signal after the replacement process. Further, in FIG. 10 (d), the section 1018 is a section corresponding to the section 1008 of FIG. 10 (b), the replacement ratio of FIG. 10 (c) is a value exceeding 0.0, and the G signal and B This is the section in which the signal is replaced. The section 1019 is a section corresponding to the section 1009 in FIG. 10 (b), the replacement ratio in FIG. 10 (c) is 0.0, and the replacement process is not performed. Further, the section 1018 to be replaced is divided into a section 1020 and a section 1021. In the section 1020, the replacement ratio in FIG. 10C is 1.0, and the signals to be replaced (G signal and B signal) completely match the signal values of the replacement source signal (R signal). It is a section. As a result, in the section 1019, the G signal and the B signal remain the same values as the G signal 1002 and the B signal 1007 in FIG. 10B. On the other hand, in the section 1020, the G signal and the B signal are gradually replaced by the signal values of the R signal, and in the section 1021, they are completely replaced by the signal values of the R signal 1006.

FIG. 10 (e) is a graph obtained by converting the RGB signal of FIG. 10 (d) after the replacement process into a luminance signal. That is, in FIG. 10E, the luminance signal 1022 is a signal in which the luminance value of the sensor saturation level 1004 or higher is generated and the gradation is enlarged.
In addition, in FIG. 3D of the first embodiment, the section 318 which does not perform the replacement process corresponds to the section 1019 which does not perform the replacement process of FIG. 10D, and the section 317 which performs the replacement process is shown in FIG. It corresponds to the section 1018 in which the replacement process of 10 (d) is performed.

Here, in the conventional image processing apparatus, problems in performing both the shine correction processing and the high-luminance gradation improvement processing will be described.
For example, if the high-luminance gradation improvement process is performed after the shine correction process, the signal after the shine correction process is in the state shown in FIG. 9 (g). As is clear from FIG. 9 (g), the R signal 920 and the B signal 922 have lost signals having a high saturation level due to the clip processing performed after the WB processing. Therefore, it is not possible to improve the high luminance gradation by replacing the signal having a high saturation level with the signal having a low saturation level.

Further, for example, if the shine correction process is performed after the high-luminance gradation improvement process, the signal after the high-luminance gradation improvement is in the state shown in FIG. 10 (d). In this case, in the section 1018, the G signal 1016 and the B signal 1017 are replaced by the R signal 1006 according to their respective replacement ratios. That is, in this section 1018, the ratio of the RGB signals is different from the ratio of the original RGB signals, and color bending occurs. Therefore, in order to perform the shine correction process in this section 1018, it is necessary to perform the correction in consideration of this color bending, and it becomes difficult to generate the complementary color of the skin color. In addition, a color different from the skin color may be erroneously detected as a skin color region due to color bending, and the shine correction processing may be performed on the erroneous region.

On the other hand, as described above, the image processing apparatus 101 of the present embodiment performs the shine correction processing on the image signal that has not been clip-processed after the WB processing, and then performs the high-luminance gradation. It is designed to perform improvement processing. That is, in the case of the image processing apparatus 101 of the present embodiment, as shown in FIG. 3D, the R signal 310 after the shine correction processing maintains a high saturation level, so that this high saturation level signal is used. The high-luminance gradation improvement processing used is possible. Further, in the section 317 where the replacement process is performed, the same color bending as in the conventional case occurs, but in the case of the present embodiment, the shine correction process has already been performed, and therefore, the color bending region is erroneously detected and erroneously. No shiny correction processing is performed on the area.

As described above, in the present embodiment, when both the processing of performing the luminance correction and the processing of expanding the luminance gradation are performed in the high luminance region, the influence of the color bending is suppressed to a small extent and the luminance correction is performed. Moreover, it is possible to enlarge the gradation of brightness.

<Second embodiment>
Next, the image processing method and the apparatus according to the second embodiment of the present invention will be described with reference to FIGS. 4 to 6. In the second embodiment, a case where the shine correction process is performed before the WB process, that is, a case where the WB process is performed as an intermediate process between the shine correction process and the high-luminance gradation improvement process will be described. In the second embodiment, in the WB processing performed as the intermediate processing, the signal to which the WB gain is applied is suppressed to a small range, and the signal value to be processed is suppressed to a value as small as possible, which is advantageous for hardware processing and the like. In the second embodiment as well, an example in which the image processing device is implemented according to the circuit configuration is given as in the first embodiment, but a CPU or the like programs a part or all of the processes performed in each circuit. It may be carried out by carrying out.

FIG. 4 is a diagram showing a configuration example of the image processing device 401 according to the second embodiment. The input image signal to the image processing device 401 has been transmitted via an image signal captured by the image pickup device, an image signal read from a recording medium, or the like, or a network or the like, as in the above-described embodiment. It may be any of the image signals.

As shown in FIG. 4, the image processing apparatus 401 includes a WB processing unit 102, a clip processing unit 103, a correction amount determination unit 104, a reverse WB processing unit 402, a correction processing unit 403, a WB processing unit 404, and a gradation processing unit 405. It is configured to have each circuit of.
The WB processing unit 102, the clip processing unit 103, and the correction amount determination unit 104 of FIG. 4 perform the same processing as the WB processing unit 102, the clip processing unit 103, and the correction amount determination unit 104 shown in FIG. Their description will be omitted.

The inverse WB processing unit 402 multiplies the shine correction amount for the R signal and the B signal determined by the correction amount determination unit 104 by a value having a reciprocal relationship with the WB gain value used by the WB processing unit 102. Perform processing. Then, the reverse WB processing unit 402 sends the value of the shine correction amount for the R signal and the B signal after the reverse WB processing is performed to the correction processing unit 403.

The correction processing unit 403 performs the shine correction processing by subtracting the shine correction amount after the reverse WB processing by the reverse WB processing unit 402 from the input image signal that has not been WB processed. The image signal after the shiny correction processing by the correction processing unit 403 is sent to the WB processing unit 404.

The WB processing unit 404 is a processing unit that performs WB processing similar to the WB processing unit 102. The WB processing unit 404 performs WB processing such that a predetermined WB gain value is multiplied by the R signal and the B signal of the image signal after the shine correction processing by the correction processing unit 403. The image signal after WB processing by the WB processing unit 404 is sent to the gradation processing unit 405.
Similar to the gradation processing unit 106 of FIG. 1, the gradation processing unit 405 performs high-luminance gradation improvement processing that replaces a color signal having a high saturation level with a color signal having a low saturation level.

Next, the procedure of image processing and the details of image processing according to the second embodiment will be described with reference to the flowchart of FIG. 5 and the graphs of the signal values shown in FIGS. 6 (a) to 6 (f). In each of the graphs shown in FIGS. 6A to 6F, the horizontal axis represents the amount of light incident on the image sensor and the vertical axis represents the signal intensity of the image signal after each processing, as in the graphs described above. , The broken line represents the R signal, the solid line represents the G signal, and the alternate long and short dash line represents the B signal.

In step S502 of the flowchart of FIG. 5, the image processing apparatus 401 of FIG. 4 acquires an image signal in a state before the WB processing is performed. Further, as in the first embodiment, the signal value of the R signal of the input image signal is Rin, the signal value of the G signal is Gin, and the signal value of the B signal is Bin. Further, as in the graph described above, it is assumed that the input image signal and the RGB signal are the R signal 601, the G signal 602, and the B signal 603 shown in FIG. 6A. In the G signal 602, the signal value Gin is the sensor saturation level 604 in the section 605 and saturation occurs in the image sensor, but the signal value Rin of the R signal 601 and the signal value Bin of the B signal 603 are saturated. Not.

Steps S503 to S505 of FIG. 5 are processing steps corresponding to steps 203 to S205 of FIG. That is, similarly to the above, in step S503, a process of applying the WB gain is performed, and in step S504, a process of clipping a predetermined value or more is performed. Further, in step S505, the skin color to be corrected is obtained, and the skin color correction amount is further obtained, which is determined as the shine correction amount. The signal values obtained in steps S503 to S505 are the same as the signal values in FIGS. 9 (a) to 9 (f) described in the first embodiment, and therefore, it is determined in step S505 in FIG. It is the same shine correction amount as in (f). Further, also in the second embodiment, as in the first embodiment, the shine correction amounts determined for the RGB signals are Tr, Tg, and Tb, respectively. In the case of the flowchart of FIG. 5, after step S505, the image processing apparatus 401 proceeds to step S506.

Step S505 is a reverse WB processing step. In step S505, the reverse WB processing unit 402 performs reverse WB processing in which the complementary color signal of the skin color, which is the amount of shine correction generated in step S505, is multiplied by the reciprocal of the WB gain. Here, the reciprocal of the WB gain Wr with respect to the R signal is represented by 1 / Wr, and the reciprocal of the WB gain Wb with respect to the B signal is represented by 1 / Wb. Therefore, as a result of the reverse WB processing by the reverse WB processing unit 402, the shine correction amount used for the shine correction processing becomes a value represented by Tr / Wr for the R signal, and Tg and B signals for the G signal. Is a value represented by Tb / Wb.

FIG. 6B is a graph showing the signal value of the shine correction amount obtained as a result of the reverse WB processing by the reverse WB processing unit 402. The shine correction amount for the RGB signal shown in FIG. 6B is a value obtained by multiplying the shine correction amount shown in FIG. 9F by the reciprocal of the WB gain. That is, the shine correction amount 606 for the R signal in FIG. 6B is a value obtained by multiplying the shine correction amount 917 in FIG. 9F by the reciprocal of the WB gain. Similarly, the shine correction amount 607 for the G signal is the value obtained by multiplying the shine correction amount 918 by the reciprocal of the WB gain, and the shine correction amount 608 for the B signal is the value obtained by multiplying the shine correction amount 919 by the reciprocal of the WB gain. Further, the section 609 of FIG. 6B is a section in which the shine correction amount exceeds 0, and is a section in which the shine correction process is performed.

Next, in step S507, the correction processing unit 403 of FIG. 4 performs a shiny correction process. That is, the correction processing unit 403 subtracts the shine correction amount obtained for each color signal of the RGB signal in step S506 from the input image signal that has not been WB-processed acquired in step S502.

FIG. 6C is a graph showing the RGB signal after the shiny correction processing by the correction processing unit 403. That is, as a result of the shine correction processing, the R signal of the input image signal becomes the R signal 610 whose signal value is represented by Rin-Tr / Wr. Further, the G signal of the input image signal is a G signal 611 whose signal value is represented by Gin-Tg, and the B signal of the input image signal is a B signal 612 whose signal value is represented by Bin-Tb / Wb. .. Further, the section 613 represents a section in which the shine correction has been performed, and corresponds to the section 609 in FIG. 6B.

Next, in step S508, the WB processing unit 404 of FIG. 4 performs WB processing for multiplying the WB gain. That is, the WB processing unit 404 applies the same WB gain Wr as the WB processing unit 102 to the R signal 610 after the shine correction processing is performed by the correction processing unit 403, and the B signal 612 after the shine correction processing is applied. On the other hand, the same WB gain Wb as that of the WB processing unit 102 is applied.

FIG. 6D is a graph showing RGB signals after WB processing by the WB processing unit 404. As a result of the WB processing by the WB processing unit 404, the R signal becomes the R signal 614 whose signal value is represented by (Rin-Tr / Wr) * Wr = Rin * Wr-Tr. Further, the B signal is a B signal 615 whose signal value is represented by (Bin-Tb / Wb) * Wb = Bin * Wb-Tb. The G signal is still the G signal 611 whose signal value is represented by Gin-Tg.

Next, in step S509, the gradation processing unit 405 of FIG. 4 performs high-luminance gradation improvement processing on the RGB signal after WB processing by the WB processing unit 404. That is, in step S509, the gradation processing unit 405 performs a process of replacing the G signal and the B signal having the lowest saturation level with the R signal having the highest saturation level among the RGB signals after the WB processing by the WB processing unit 404. Do. After this step S509, the image processing apparatus 401 ends the processing of the flowchart of FIG.

FIG. 6E is a graph showing a state of high-luminance gradation improvement processing for an RGB signal by the gradation processing unit 405. That is, the G signal after the replacement process by the high-luminance gradation improvement process becomes the G signal 616 by replacing the signal value with the R signal 614, and similarly, the B signal after the replacement is the signal value of the R signal 614. By being replaced with, the B signal 617 is obtained. Further, the section 618 is a section corresponding to the section 309 of FIG. 3B, which is a section having high brightness and a skin color, and is a skin color section targeted for shine correction. The section 619 is a section in which the replacement process is performed on the G signal and the B signal, and corresponds to the section 308 in FIG. 3 (b). The section 620 is a section other than the section 619, and the replacement process is not performed. Further, section 621 and section 622 are sections obtained by dividing section 619, and in particular, section 622 is a section in which the signal value to be replaced is completely replaced by the signal value of the replacement source.

Further, in the second embodiment as well as described in the first embodiment, the replacement process is performed using the replacement ratio table of FIG. 10C described above. That is, in the replacement ratio table, the replacement ratio is 0.0 in the section corresponding to the section 620, the replacement rate increases in the section corresponding to the section 621, and the replacement rate is 1.0 in the section corresponding to the section 622. It is made like a table. By using this replacement ratio table, in the section 620, the G signal 616 remains the same signal value as the G signal 611 in FIG. 6 (d), and the B signal 617 remains the same signal value as the B signal 615. .. Further, the G signal 616 and the B signal 617 are gradually replaced by the R signal 614 in the section 621, and are completely replaced by the R signal 614 in the section 622. On the other hand, since the section 618 is included in the section 620 in which the replacement process is not performed, the effect of the shine correction process performed in FIG. 6D is maintained.

FIG. 6 (f) is a graph obtained by converting the RGB signal of FIG. 6 (e) into a luminance signal. In FIG. 6 (f), the solid line represents the luminance signal 623. That is, the luminance signal 623 is a signal in which the luminance value of the sensor saturation level 604 or higher is generated and the gradation is expanded, and the luminance value is lowered in the section 618, so that the effect of the shine correction processing is maintained. There is.

As described above, the image processing apparatus 401 of the second embodiment performs the shine correction processing on the input image signal before the WB processing, applies the WB gain, and then performs the high-luminance gradation improvement processing. ing. As a result, the image processing device 401 of the second embodiment can provide the effects of the shine correction processing and the high-luminance gradation improvement processing on the same high-luminance region. Further, the image processing device 401 performs WB processing as an intermediate processing between the shine correction processing and the high-luminance gradation improvement processing, suppresses the signal to which the WB gain is applied within a small range, suppresses the processed signal to a value as small as possible, and performs hard processing. It has an advantageous configuration.
<Third embodiment>
Next, the image processing method and the apparatus according to the third embodiment of the present invention will be described with reference to FIGS. 7 and 8. The image processing apparatus of the third embodiment has a configuration in which shine correction processing and high-luminance gradation improvement processing are selectively performed. In the third embodiment, as in the first and second embodiments, the image processing apparatus is implemented according to the circuit configuration, but some or all of the processes performed in each circuit are CPUs and the like. May be implemented by running the program.

FIG. 7 is a diagram showing a configuration example of the image processing device 701 according to the third embodiment. The input image signal to the image processing device 701 is an image signal as in the first and second embodiments described above.
The image processing device 701 shown in FIG. 7 includes a WB processing unit 102, a clip processing unit 103, a correction amount determination unit 104, a determination unit 702, selection circuits 703 and 704, a clip processing unit 705, a correction processing unit 706, and gradation processing. Each circuit of the part 707 is included.
The WB processing unit 102, the clip processing unit 103, and the correction amount determination unit 104 of FIG. 7 perform the same processing as the WB processing unit 102, the clip processing unit 103, and the correction amount determination unit 104 shown in FIG. Their description will be omitted.

The determination unit 702 examines the value of the shine correction amount determined by the correction amount determination unit 104, and determines whether or not all the shine correction amounts are 0 values. Then, the determination result of the determination unit 702 is used as a control signal for the selection process in the selection circuits 703 and 704.
The selection circuits 703 and 704 are controlled to select the circuit to be used according to the determination result of the determination unit 702. That is, the selection circuits 703 and 704 select whether to use the path passing through the clip processing unit 705 and the correction processing unit 706 or the path passing through the gradation processing unit 707 based on the determination result of the determination unit 702.

Similar to the clip processing unit 103, the clip processing unit 705 performs a process of clipping a signal having a predetermined value or more with respect to the image signal after the WB processing by the WB processing unit 102.
The correction processing unit 706 performs a shine correction process of subtracting the shine correction amount determined by the correction amount determination unit 104 from the image signal after the clip processing by the clip processing unit 705.
The gradation processing unit 707 performs high-luminance gradation improvement processing for replacing the image signal after WB processing by the WB processing unit 102 with a color signal having a high saturation level and a color signal having a low saturation level.

Next, the procedure of image processing and the details of image processing according to the third embodiment will be described with reference to the flowchart of FIG.
Since the processes from step S802 to step S805 in FIG. 8 are the same as the processes from steps S202 to S205 in FIG. 2, their description will be omitted.
In the case of the flowchart of FIG. 8, after step S803 and after step S805, the image processing apparatus 701 proceeds to step S806.

Step S806 is a determination step. In step S806, the determination unit 702 performs a determination process using the shine correction amount determined by the correction amount determination unit 104. When the determination unit 702 determines that all the shine correction amounts are 0 values (Yes), the process of the image processing device 701 proceeds to the high-luminance gradation improvement process of step S809. On the other hand, when the determination unit 702 does not determine that all the shine correction amounts are 0 values (No), the processing of the image processing device 701 proceeds to the clip processing of M step S807 and the subsequent shine correction processing of step S808. .. That is, in the image processing device 701 of FIG. 7, when the determination unit 702 determines that all the shine correction amounts are 0 values, the selection circuits 703 and 704 select the path of the gradation processing unit 707. On the other hand, when the determination unit 702 does not determine that all the shine correction amounts are 0 values, the selection circuits 703 and 704 select the routes of the clip processing unit 705 and the correction processing unit 706.

Proceeding to the process of step S807, the clip processing unit 705 performs clip processing on the image signal after WB processing in step S803. That is, the clip processing unit 705 clips a signal having a predetermined clip value or more.
After that, in step S808, the correction processing unit 706 performs a shine correction process of subtracting the shine correction amount determined in step S805 from the image signal after the clip process in step S807. Since the graph of the signal value at this time is the same as the example of the shine correction process described with reference to FIG. 9, the description thereof will be omitted. After this step S808, the image processing device 701 ends the processing of the flowchart of FIG.

On the other hand, when the process proceeds to step S809, the gradation processing unit 707 performs high-luminance gradation improvement processing on the image signal after the WB processing in step S803. That is, the gradation processing unit 707 performs high-luminance gradation improvement processing such as replacing a color signal having a high saturation level with a color signal having a low saturation level. Since the graph of the signal value at this time is the same as the example of the high-luminance gradation improvement processing described with reference to FIG. 10, the description thereof will be omitted. After this step S809, the image processing apparatus 701 ends the processing of the flowchart of FIG.

As described above, the image processing apparatus 701 of the third embodiment selects the shine correction process in the region where it is determined that the shine correction process should be performed because all the shine correction amounts are not 0 values, and if not, the shine correction process is selected. Select high-brightness gradation improvement processing. That is, the image processing apparatus 701 of the third embodiment performs the same shine correction processing as usual when it is not determined that all the shine correction amounts are 0 values. On the other hand, when it is determined that the shine correction amounts are all 0 values, the image processing device 701 performs the same high-luminance gradation improvement processing as usual.

As described above, according to the third embodiment, the shine correction process is performed in the area requiring the shine correction process, and the high-luminance gradation improvement process is performed in the other areas. The correction process for the high-luminance region will be appropriately performed.
Also in the third embodiment, as in each of the above-described embodiments, when both the processing of performing the luminance correction and the processing of expanding the luminance gradation are performed in the high luminance region, the luminance is not affected by the color bending. It is possible to make corrections and enlarge the gradation of brightness.

Although each embodiment of the present invention has been described in detail above, the present invention is not limited to these embodiments, and various embodiments within the scope of the gist of the present invention are also included in the present invention.
The processing of each function and the flowchart related to the image processing apparatus of each of the above-described embodiments may be realized only by the hardware configuration, or may be configured as a software module by executing a program by a CPU or the like. Further, a part may be configured as a hardware configuration and the rest may be configured as a software module. The program for configuring this software module is not only when it is prepared in advance and stored in an internal memory or the like, but also when it is acquired from a recording medium such as an external memory or via a network (not shown). You may.
In the above-described embodiment, an example of processing an image signal acquired from a digital camera or the like has been given, but the image signal is acquired by a surveillance camera, an industrial camera, an in-vehicle camera, a medical camera, a camera of a smartphone or a tablet terminal, or the like. It may be a signal.

A program that realizes one or more functions in signal processing according to the present invention can be supplied to a system or device via a network or storage medium, and is read and executed by one or more processors of the computer of the system or device. It is feasible by being done. It can also be realized by a circuit (for example, ASIC) that realizes one or more functions.

The above-described embodiments are merely examples of implementation of the present invention, and the technical scope of the present invention should not be construed in a limited manner by these. That is, the present invention can be implemented in various forms without departing from the technical idea or its main features.

101: Image processing device, 102: WB processing unit, 103: Clip processing unit, 104: Correction amount determination unit, 105: Correction processing unit, 106: Gradation processing unit

Claims (18)

A processing means that performs white balance processing on the input image signal,
A correction means for performing brightness correction processing on a high-luminance region of an image signal after the white balance processing by the processing means has been performed.
A gradation processing means that performs gradation processing that enlarges the luminance gradation in the high-luminance region of the image signal after the correction processing by the correction means is performed.
An image processing device characterized by having. A clip means that performs clip processing on an image signal after the white balance processing is performed by the processing means, and
A correction amount determining means for determining a correction amount when the correcting means corrects the brightness based on an image signal after the clipping process is performed by the clipping means.
The image processing apparatus according to claim 1, wherein the image processing apparatus has. The correction means performs, as the correction process, a correction for lowering the brightness of a color signal in a specific color region and a high-luminance region among the image signals after the white balance processing is performed by the processing means. The image processing apparatus according to claim 1 or 2. The gradation processing means performs a process of replacing a color signal having a high saturation level with a color signal having a low saturation level in a high-luminance region of an image signal after the correction processing by the correction means is performed. The image processing apparatus according to any one of claims 1 to 3, wherein the image processing apparatus is performed as a process. A correction means that corrects the brightness of the high-brightness region of the input image signal, and
A first processing means that performs white balance processing on an image signal after the correction processing by the correction means, and
A gradation processing means that performs gradation processing that enlarges the luminance gradation in the high-luminance region of the image signal after the white balance processing by the first processing means is performed.
An image processing device characterized by having. A second processing means that performs white balance processing on the input image signal,
A clip means that performs clip processing on an image signal after the white balance processing is performed by the second processing means, and
A correction amount determining means for determining a correction amount when the correcting means corrects the brightness based on an image signal after the clipping process is performed by the clipping means.
The image processing apparatus according to claim 5, wherein the image processing apparatus has. The correction amount determining means obtains a correction amount according to the image signal after the clip processing by the clip means is performed, and uses the correction amount in the white balance processing by the second processing means. The image processing apparatus according to claim 6, wherein a value obtained by multiplying the reciprocal of the obtained white balance gain is determined as a correction amount when the correction means corrects the brightness. Any of claims 5 to 7, wherein the correction means performs correction for lowering the brightness of a color signal in a specific color region and a high-luminance region among the input image signals as the correction process. The image processing apparatus according to item 1. The gradation processing means performs a process of replacing a color signal having a high saturation level with a color signal having a low saturation level in a high-luminance region of an image signal after the white balance processing by the first processing means has been performed. The image processing apparatus according to any one of claims 5 to 8, wherein the image processing apparatus is performed as the gradation processing. A processing means that performs white balance processing on the input image signal,
A correction means for performing brightness correction processing on a high-luminance region of an image signal after the white balance processing by the processing means has been performed.
A gradation processing means that performs gradation processing that enlarges the luminance gradation in the high-luminance region of the image signal after the white balance processing by the processing means is performed.
A selection means for selecting one of the correction process by the correction means and the gradation process by the gradation processing means, and
An image processing device characterized by having. It has a first clipping means that performs clipping processing on an image signal after the white balance processing by the processing means is performed.
The image processing apparatus according to claim 10, wherein the correction means performs the correction process on an image signal after the clip process is performed by the first clip means. A second clipping means that performs clipping processing on the image signal after the white balance processing by the processing means has been performed, and
A correction amount determining means for determining a correction amount when the correcting means corrects the brightness based on an image signal after the clipping process is performed by the second clipping means.
10. The image processing apparatus according to claim 10 or 11. A control means for controlling whether the selection means selects the correction process by the correction means or the gradation process by the gradation processing means based on the correction amount determined by the correction amount determining means. The image processing apparatus according to claim 12, wherein the image processing apparatus has. The control means
When it is determined that all the correction amounts determined by the correction amount determining means are 0 values, the selection means is made to select to perform the gradation processing by the gradation processing means.
The claim is characterized in that when it cannot be determined that all the correction amounts determined by the correction amount determining means are 0 values, the selection means is made to select to perform the correction processing by the correction means. 13. The image processing apparatus according to 13. An image processing method executed by an image processing device.
A processing process that performs white balance processing on the input image signal,
A correction processing step of performing brightness correction processing on a high-luminance region of an image signal after the white balance processing by the processing step is performed,
A gradation processing step of performing gradation processing for expanding the luminance gradation in a high-luminance region of the image signal after the correction processing is performed by the correction processing step, and
An image processing method characterized by having. An image processing method executed by an image processing device.
A correction processing step that corrects the brightness of the high-brightness region of the input image signal, and
A first processing step of performing white balance processing on an image signal after the correction processing is performed by the correction processing step, and
A gradation processing step of performing gradation processing for expanding the luminance gradation of the high-luminance region of the image signal after the white balance processing by the first processing step is performed, and
An image processing method characterized by having. An image processing method executed by an image processing device.
A processing process that performs white balance processing on the input image signal,
A correction processing step of performing brightness correction processing on a high-luminance region of an image signal after the white balance processing by the processing step is performed,
A gradation processing step of performing gradation processing for expanding the luminance gradation of a high-luminance region of the image signal after the white balance processing by the processing step is performed, and
A selection step of selecting one of the correction process by the correction processing step and the gradation processing by the gradation processing step.
An image processing method characterized by having. A program for causing a computer to function as each means of the image processing apparatus according to any one of claims 1 to 14.

Images