JP2011175608A - Image processing device and image processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing device and an image processing method, capable of providing a natural image close to the visual characteristics of humans. <P>SOLUTION: At least one sub-band image is generated from an image signal, and a histogram of luminance value of pixels included in a predetermined local area around a target pixel in the sub-band image is calculated for all pixels of the sub-band image. Furthermore, sub-band gradation conversion characteristics for each sub-band image is calculated based on the histogram; gradation conversion characteristics of the image signal is calculated based on the sub-band gradation conversion characteristics; and gradation conversion processing of the image signal is performed, based on the calculated gradation conversion characteristics. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to an image processing apparatus, and more particularly to an image processing apparatus that performs gradation conversion processing on an image signal.

  Conventionally, in an image processing apparatus applied to a digital still camera, a video camera, or the like, gradation conversion processing is performed on an image signal in consideration of a display system and human visual characteristics. As an example, there is a method of calculating gradation conversion characteristics using local information of an image signal or the like and performing gradation conversion processing based on the gradation conversion characteristics. For example, Patent Document 1 discloses a technique that enables accurate gradation conversion according to image characteristics by performing multi-resolution decomposition on an input image signal and performing weighted gradation conversion using frequency components. Has been.

  On the human retina, there are cone cells and rod cells, which are types of photoreceptor cells. The pyramidal cell is responsible for resolution and expresses visual substances having different wavelength characteristics, and is the basis of color vision, while it has low sensitivity and requires a sufficient amount of light. Since rod cells express only a single visual substance, they are not involved in color vision, but are highly sensitive. In addition, the fovea in the human retina contributes to high-definition central vision and plays an important role in all of reading, watching TV and movies, driving, and other visual activities. This fovea has a feature that the density of cone cells is high, and there are almost no rod cells at the center, and the number of cone cells decreases from the center to the periphery. For this reason, it has been reported that the foveal characteristic, in other words, the human visual characteristic, has a high resolution and a narrow dynamic range. Therefore, when attention is paid to a local area of an image, when the area is high in brightness, depending on the area, it may be more natural for the human eye to display an image after being saturated to some extent.

JP 2006-41744 A

  However, in the technique described in Patent Document 1 described above, for example, when the image signal includes a very high-luminance light source as exemplified by a point light source, the above-described human visual characteristics are inherent. There is a problem that the details of the light source that cannot be recognized are also reproduced, resulting in an unnatural image.

  The present invention has been made to solve the above problem, and an object thereof is to provide an image processing apparatus and an image processing method capable of obtaining a natural image close to human visual characteristics.

  In order to solve the above problems, the present invention employs the following means.

  The present invention is an image processing apparatus that performs a gradation conversion process on an image signal, a subband image generating unit that generates at least one subband image from the image signal, and the subband image, Histogram calculation means for calculating a luminance value histogram of pixels included in a predetermined local region centered on the target pixel for all pixels of the subband image, and a subband level for each subband image based on the histogram. Subband gradation conversion characteristic calculating means for calculating tone conversion characteristics, gradation conversion characteristic calculating means for calculating gradation conversion characteristics of the image signal based on the subband gradation conversion characteristics, and the gradation conversion characteristics The image processing apparatus is provided with gradation conversion processing means for performing gradation conversion processing of the image signal based on the above.

  The present invention is also an image processing method for performing gradation conversion processing on an image signal, the step of generating at least one subband image from the image signal, and a pixel of interest for the subband image. Calculating a histogram of luminance values of pixels included in a predetermined local area centered on the image for all pixels of the subband image, and subband gradation conversion characteristics for each subband image based on the histogram. Calculating a gradation conversion characteristic of the image signal based on the subband gradation conversion characteristic; performing a gradation conversion process of the image signal based on the gradation conversion characteristic; An image processing method is provided.

  According to the present invention, a natural image close to human visual characteristics can be obtained.

1 is a block diagram showing a schematic configuration of an image processing apparatus according to a first embodiment of the present invention. It is a block diagram which shows schematic structure of the subband image generation part in the image processing apparatus concerning the 1st Embodiment of this invention. It is explanatory drawing in the case of producing | generating a subband image by subband decomposition | disassembly by the subband image generation part in the image processing apparatus concerning the 1st Embodiment of this invention. It is explanatory drawing in the case of producing | generating a subband image by subband decomposition | disassembly by the subband image generation part in the image processing apparatus concerning the 1st Embodiment of this invention. It is explanatory drawing in the case of calculating the histogram of a local region in the histogram calculation part in the image processing apparatus concerning the 1st Embodiment of this invention. It is explanatory drawing in the case of calculating a gradation conversion curve based on a histogram in the subband gradation conversion characteristic calculation part in the image processing apparatus concerning the 1st Embodiment of this invention. It is a flowchart which shows the process of the image processing in the image processing apparatus concerning the 1st Embodiment of this invention. It is a block diagram which shows schematic structure of the subband image generation part in the image processing apparatus concerning the 2nd Embodiment of this invention. It is a flowchart which shows the process of the image processing in the image processing apparatus concerning the 2nd Embodiment of this invention.

  Embodiments of an image processing apparatus according to the present invention will be described below with reference to the drawings.

[First Embodiment]
FIG. 1 is a block diagram showing a schematic configuration of an image processing apparatus according to the first embodiment of the present invention. As shown in FIG. 1, the image processing apparatus according to the present embodiment includes a lens system 100, an aperture 101, a color filter 102, a CCD 103, an AF motor 104, an A / D conversion unit 105, a buffer 106, an imaging control unit 107, an image. A processing unit 120, a compression unit 114, an output unit 115, a control unit 116, and an external I / F unit 117 are provided.

  The lens 100 system is arranged so as to form an image of a subject through a diaphragm 101 and form an image on a CCD 103 which is an image sensor through a color filter 102. The AF motor 104 is driven and controlled by an imaging control unit 107 described later, and is driven to focus the lens system 100 on the subject based on a control signal from the imaging control unit 107. The CCD 103 generates electrical image information based on the subject image formed by the lens system 100 and outputs the electrical image information to the A / D conversion unit 105. In the present embodiment, the description will be made assuming that the CCD 103 uses an RGB primary color single-plate CCD. The A / D conversion unit 105 converts the image information generated by the CCD 103 into an image signal as discrete digital data that can be subjected to predetermined processing in the image processing unit 120, and temporarily converts the converted image signal into the buffer 106. And output from the buffer 106 to the image processing unit 120. A plurality of image signals are acquired by repeating the above processing. The compression unit 114 compresses the image signal that has been subjected to predetermined processing by the image processing unit 120 described later, and outputs the compressed image signal to the output unit 115.

  The imaging control unit 107 controls the electronic shutter speed of the diaphragm 101 and the CCD 103 for adjusting the amount of incident light using a luminance level in the image signal or a luminance sensor (not shown). The control by the imaging control unit 107 is performed as many times as the number of times of photographing in the main photographing. That is, the above-described processing is performed a plurality of times, and image signals of a plurality of captured images are temporarily stored in the buffer 106, and the image signals are sequentially output from the buffer 106 to the image processing unit 120 one by one.

  The control unit 116 includes an imaging control unit 107, a signal processing unit 108, a subband generation unit 109, a histogram calculation unit 110, a subband gradation conversion characteristic calculation unit 111, and a gradation conversion characteristic calculation unit of the image processing unit 120 described later. 112, the gradation conversion unit 113, and the compression unit 114 are bidirectionally connected to drive and control each of them. The external I / F unit 117 includes a power switch (not shown), a shutter button, and an interface for switching various modes during shooting.

  The image processing unit 120 includes a signal processing unit 108, a subband generation unit 109, a histogram calculation unit 110, a subband gradation conversion characteristic calculation unit 111, a gradation conversion characteristic calculation unit 112, and a gradation conversion unit 113.

  The signal processing unit 108 reads a single-panel image signal input from the buffer 106 based on the control of the control unit 116, and performs predetermined image processing such as interpolation processing, white balance adjustment processing, electronic zoom processing, and noise suppression processing. The three-plate image signal of each pixel RGB is generated and output to the subband generation unit 109 and the gradation conversion unit 113. Further, the RGB signal may be converted into a YCbCr signal using the following equation (1).

  The subband generation unit 109 generates a plurality of subband images by down-sampling the image signal, for example. Therefore, as shown in FIG. 2, the subband generation unit 109 includes a decomposition unit 200, a specific luminance region detection unit 201, a decomposition level setting unit 202, and a buffer 203. The decomposition unit 200 includes the specific luminance region detection unit 201. In addition, the specific luminance area detection unit 201 is connected to the decomposition level setting unit 202 and the buffer 203, and the decomposition level setting unit 202 is connected to the decomposition unit 200.

  The decomposition unit 200 hierarchically subband decomposes the image signal transferred from the signal processing unit 108 to a desired decomposition level, for example, downsamples hierarchically to a desired decomposition level at a predetermined pixel interval d. Then, a plurality of subband images having different resolutions are generated. This predetermined pixel interval can be arbitrarily determined. In the present embodiment, the case of downsampling with d = 2 will be described as an example. FIG. 3 is an explanatory diagram of subband image generation by downsampling. As shown in FIG. 3, the decomposition unit 200 sets the image signal transferred from the signal processing unit 108 as the first stage, and subband images at each i stage (i is an integer of 1 to n) up to the nth stage. Is generated. In FIG. 3, the downsampling of d = 2 is performed by thinning out even rows and even columns or odd rows and odd columns of image signals. The subband image generated by the decomposition unit 200 is transferred to the specific luminance area detection unit 201.

  In the specific luminance area detection unit 201, threshold processing is performed on the subband image transferred from the decomposition unit 200, and an area composed of pixels having luminance values in a predetermined luminance range is set as the specific luminance area from the processed subband image. To detect. Here, the specific luminance region is an extremely high luminance region such as the sun or an electric lamp. These regions may become unnatural if local gradation conversion is performed in the same procedure as other regions. However, the present invention is not limited to these high-luminance areas, and only specific luminance areas having a certain range of luminance can be subjected to gradation conversion different from other areas. Note that the method for detecting the specific luminance region is not limited to this, and the region and the area related to the specific luminance may be detected using a known region detection process such as a region expansion method, a water divide method, or MSER (Maximally Stable Extremal Regions). Good.

  In the decomposition level setting unit 202, the decomposition level detected by the specific luminance region detection unit 201 includes the area S (i) of the specific luminance region included in the i-th subband image and a predetermined level in the histogram calculation unit 110 described later. The decomposition level is set based on the area N of the local region. Specifically, the number of subband decomposition stages as the decomposition level is set based on the following equation (2).

  The number of subband decomposition stages is set so that the maximum number of stages i satisfying Equation 2 becomes the maximum number n of subband decomposition stages. FIG. 4 is an explanatory diagram of setting the number of subband decomposition stages. Here, even when a plurality of subband images are generated, the area N of the local region to be compared with the subband images is constant. On the other hand, as shown in FIG. 4, as the subband image is generated in a plurality of stages, the area of the specific luminance region included in the subband image becomes smaller. Therefore, in the processing based on Equation 2, until the subband image in which the specific luminance area is completely included in the predetermined local area and the ratio of the specific luminance area in the local area is maximized is generated. Subband decomposition such as downsampling is repeated hierarchically, and as a result, a plurality of subband images are generated.

  In this way, by limiting the number of steps to be decomposed as the decomposition level of subband decomposition, not only the amount of calculation is reduced, but also ideal gradation that does not include inappropriate subband images that make the local area wider than the specific luminance area Conversion characteristics can be calculated. In other words, if the local area is wider than the specific luminance area, the gradation conversion characteristics for the specific luminance area are applied to areas other than the specific luminance area, and desired gradation conversion can be performed. Disappear. For this reason, the area S (i) of the specific luminance region and the area N of the local region are compared, and the number of subband decomposition stages is set in a range that satisfies the above equation (2).

  The histogram calculation unit 110 calculates, for each of the plurality of subband images transferred from the subband generation unit 109, a histogram of luminance values of pixels included in a predetermined local region centered on the target pixel. The size of the local region can be arbitrarily determined. FIG. 5 is an explanatory diagram regarding a case where a histogram is generated for a predetermined local region. As shown in FIG. 5, as an example, the histogram of the local area indicated by the rectangular area differs depending on the position to be calculated. The histogram calculation unit 110 calculates a histogram in the above-described local region for all the pixels of the subband image, and transfers the calculated histogram to the subband gradation conversion characteristic calculation unit 111.

  The subband tone conversion characteristic calculation unit 111 calculates the tone conversion characteristic for each subband image by performing, for example, accumulation processing on the histogram transferred from the histogram calculation unit 110. Here, when the accumulation process is performed, the clip process is performed based on the statistics of the histogram of a predetermined local region centered on the target pixel. Specifically, if the mode value m of the histogram satisfying the following equation (3) is within ± r with respect to the specific luminance value I, the clipping process is performed with a predetermined clip value c, otherwise the clip is performed. The tone conversion curve is calculated without processing. By controlling the clipping process in this way, it is possible to clip only the specific luminance area. Further, the luminance region near the specific luminance value can be controlled by the value of r.

  Here, in the present embodiment, the mode value is used as the statistics of the histogram, but the mode value is not limited to this, and any one of the median value, mode value, minimum value, maximum value, and variance value is used. Also good. FIG. 6 is an explanatory diagram for calculating a gradation conversion curve based on a histogram. FIG. 6A shows a histogram before clip processing and a clip value c, and FIG. 6B shows a histogram after clip processing. Accumulation processing is performed on the transferred histogram to calculate gradation conversion characteristics. FIG. 6C shows tone conversion characteristics calculated using histogram accumulation processing. The solid line is the gradation conversion characteristic calculated from the histogram before clip processing, and the dotted line is the gradation conversion characteristic calculated from the histogram after clip processing. By performing clipping processing from FIG. 6C, the gradation conversion characteristic becomes close to linear, and the gradation conversion characteristic does not significantly enhance the input luminance. The clip value c used in the clip processing may be changed according to the statistics of the histogram. For example, as shown in Equation 4 below, the clip value c is controlled by the variance value v of the histogram. Here, w is a normalization coefficient.

  When the clip value c is controlled by the above-described equation (4), when the histogram is concentrated on the specific luminance, the clip value becomes low, and gradation conversion that does not emphasize the input luminance is possible. Here, although the variance value is used as the statistical amount of the histogram, any one of an average value, a median value, a mode value, a minimum value, and a maximum value may be used. By performing the clip processing as described above, the cumulative histogram when the histogram is concentrated on the specific luminance becomes smooth, and the gradation conversion characteristics in the subband image can be calculated without emphasizing the specific luminance region. The calculated gradation conversion characteristic in the subband image is transferred to the gradation conversion characteristic calculation unit 112.

  The gradation conversion characteristic calculation unit 112 calculates the gradation conversion characteristic using the gradation conversion characteristic of the subband image calculated by the subband gradation conversion characteristic calculation unit 111. An example of a method for calculating the gradation conversion characteristics of the specific luminance region is shown in Formula 5.

  Here, LHE (i) is a subband gradation conversion characteristic of each subband image transferred from the subband gradation conversion characteristic calculation unit 111, α (i) is a weighting coefficient corresponding to the number of stages i, and k is Normalization factor. The weighting coefficient α (i) is a coefficient that increases with the number of stages i. That is, a large weight is applied to the subband gradation conversion characteristics in which the areas of the local area and the specific luminance area are substantially equal, and a gradation conversion characteristic corresponding to the size of the subject having the specific luminance can be obtained. As the gradation conversion characteristics other than the specific luminance region, only an arbitrary i-th gradation conversion characteristic can be used as shown in the following equation (6). In equation (6), the smaller i is, the more the gradation conversion characteristic that places more importance on the local area.

  The gradation conversion unit 113 performs gradation conversion on the basis of the gradation conversion characteristics transferred from the gradation conversion characteristic calculation unit 112 for each of the image signal RGB values transferred from the signal processing unit 108, and displays Perform gamma conversion considering the display system. The image signal subjected to the gradation conversion process is transferred to the compression unit 114, and the compression unit 114 performs compression processing such as known JPEG on the image signal transferred from the gradation conversion unit 113 and transfers the image signal to the output unit 115. . The output unit 115 records and saves the compressed image signal to a memory card or the like or outputs it to an external display.

  Hereinafter, the process of image processing in the image processing apparatus of the present embodiment configured as described above will be described with reference to the flowchart of FIG.

  In the image processing apparatus according to the present embodiment, after setting shooting conditions such as ISO sensitivity, exposure, and focal length via the external I / F unit 117, a shutter button (not shown) is pressed halfway to enter the pre-shooting mode. enter. When the pre-photographing is completed, the real photographing is performed by fully pressing a shutter button (not shown) via the external I / F unit 117. The actual photographing is performed based on the focusing condition and the exposure condition obtained by the photographing control unit 107, and information at the time of photographing is transferred to the control unit 116.

  In step S11, header information is read from raw data obtained by adding header information such as the image signal size and photographing conditions to the image signal from the CCD 103, and in the next step S12, the image signal is input. In step S <b> 13, the signal processing unit 108 performs the predetermined signal processing on the image signal, and outputs the processed image signal to the subband generation unit 109.

  In the next step S14, the image signal is subband decomposed by the decomposition unit 200 of the subband generation unit 109 to generate a subband image. Then, the generated subband image is output to the specific luminance area detection unit 201. In the next step S15, the decomposition level setting unit 202 determines whether the subband image generated by the decomposition unit 200 is a subband image having a desired decomposition level. Here, the desired decomposition level is set to the decomposition level of the subband image in which the specific luminance area is included in the local area and the area other than the specific luminance area is the smallest in the local area. That is, it is determined whether or not the above-described Expression 2 is satisfied.

  Specifically, as processing based on Equation 2, it is determined as follows whether the generated subband image is a subband image of a desired decomposition level. In the specific luminance area detection unit 201, information relating to the specific luminance area detected from the subband image is output to the decomposition level setting unit 202, and the input specific luminance area and the target pixel are centered in the decomposition level setting unit 202. If the specific luminance region is not included in the local region, it is determined that the subband image generated by the decomposition unit 200 is not a subband image having a desired decomposition level. To do. In addition, when the specific luminance region is included in the local region as a result of comparing the input specific luminance region with a predetermined local region centered on the target pixel, the subband image generated earlier is further included. The subband image that maximizes the ratio of the specific luminance region in the local region is selected, and the selected subband image is determined as a desired subband image.

  As a result of the determination by the determination method as described above, when it is determined that the subband image to be determined is not a subband image of a desired decomposition level, the process returns to step S14, and the subband decomposition is performed again by the decomposition unit 200. . By repeating the processing of step S14 and step S15, the image signal is hierarchically subband decomposed to a desired decomposition level, and a plurality of subband images are generated. As a result of the determination in step S15, when it is determined that the subband image to be determined is a subband image having a desired decomposition level, the process proceeds to step S16.

  In step S16, for each subband image, for each subband image, a histogram of luminance values of pixels included in a predetermined local region centered on the pixel of interest is calculated for all pixels of the subband image; Proceed to the next Step S17. In step S17, subband tone conversion characteristics are calculated for each subband image based on the histogram calculated in step S16.

  In the next step S18, the gradation conversion characteristic for the image signal is calculated based on the subband gradation conversion characteristic. Specifically, the weighting coefficient corresponding to the decomposition level is calculated for the subband gradation conversion characteristics of each subband image, and the subband gradation conversion characteristics are weighted and averaged according to the weighting coefficient, thereby obtaining an image. A gradation conversion characteristic for the signal is calculated. In the next step S19, the tone conversion unit 113 performs tone conversion processing on the image signal using the tone conversion characteristics calculated in step S18, and this routine is terminated.

  In the gradation conversion process, the gradation conversion characteristic calculated in step S18 is applied only to the specific luminance area of the image signal, and arbitrary decomposition is applied to the area other than the specific luminance area of the image signal. Subband tone conversion characteristics for level subband images can be applied.

  As described above, according to the present embodiment, gradation conversion can be performed in accordance with the size and luminance of the subject. For example, when there is a point light source having a small high luminance in the image signal, the point light source With respect to, gradation correction can be suppressed, and a natural image signal close to human visual characteristics can be obtained.

[Second Embodiment]
Hereinafter, a second embodiment of the present invention will be described with reference to the drawings. In the image processing apparatus of this embodiment, the configuration of the subband decomposition unit 209 is different from the configuration of the subband generation unit 109 in the image processing apparatus of the first embodiment described above. That is, in the subband decomposition unit 209, after detecting the specific luminance region, the decomposition level setting unit 202 sets a decomposition level in advance for this region, and information on the set decomposition level is sent to the decomposition unit 200. Send. Based on this, the decomposition unit 200 generates a subband image for the target image, and performs subsequent processing such as histogram calculation.

  Therefore, the connection relationship between the processing units of the subband generation unit 209 is different between the subband generation unit 109 in the first embodiment and the subband image generation unit 209 in the present embodiment. Specifically, the signal processing unit 108 is connected to the specific luminance region detection unit 201, the specific luminance region detection unit 201 is connected to the decomposition level setting unit 202 and the decomposition unit 200, and the buffer 203 is connected to the decomposition unit 200 and the histogram calculation unit 110, respectively. This point is different from the subband generation unit 109 in the first embodiment. Hereinafter, configurations common to the first embodiment will be denoted by the same reference numerals, description thereof will be omitted, and different configurations will be mainly described.

  FIG. 8 is a block diagram showing a schematic configuration of an image processing apparatus according to the second embodiment of the invention. As shown in FIG. 8, the decomposition level setting unit 202 does not obtain the maximum number n as the decomposition level by comparing the area S (i) of the specific luminance region with the area N of the local region. And it calculates based on Formula 8. The area S (i) of the i-th specific luminance region is reduced in inverse proportion to the pixel interval d as shown in Equation 7.

  From Equation 7, the maximum stage number n that satisfies Equation 2 is Equation 8 below.

  Information relating to the maximum number of stages n, which is the decomposition level set by the decomposition level setting unit 202, is output to the decomposition unit 200, and the decomposition unit 200 generates a plurality of subband images in a hierarchical manner. Is transferred to and stored in the buffer 203. Further, the subband images accumulated in the buffer 203 are output to the histogram calculation unit 110. Each subband image holds information regarding the number of stages i as the decomposition level of each subband image.

  Hereinafter, the process of image processing in the image processing apparatus of the present embodiment configured as described above will be described with reference to the flowchart of FIG.

  In the image processing apparatus according to the present embodiment, after setting shooting conditions such as ISO sensitivity, exposure, and focal length via the external I / F unit 117, a shutter button (not shown) is pressed halfway to enter the pre-shooting mode. enter. When the pre-photographing is completed, the real photographing is performed by fully pressing a shutter button (not shown) via the external I / F unit 117. The actual photographing is performed based on the focusing condition and the exposure condition obtained by the photographing control unit 107, and information at the time of photographing is transferred to the control unit 116.

  In step S21, header information is read from raw data obtained by adding header information such as the image signal size and photographing conditions to the image signal from the CCD 103, and in the next step S22, the image signal is input. In step S <b> 23, the signal processing unit 108 performs the predetermined signal processing on the image signal, and outputs the processed image signal to the subband generation unit 109.

  In step S24, the decomposition level is set by the decomposition level setting unit 202. Specifically, the specific luminance area detection unit 201 detects a specific luminance area from the image signal, that is, a specific luminance area with the number of stages n = 1, and substitutes it into Equation 8 to calculate the maximum number of stages n as a decomposition level. In the next step S25, the decomposition unit 200 hierarchically subband decomposes the image signal up to the previously calculated decomposition level to generate a plurality of subband images having different resolutions.

  In step S26, for each subband image, a histogram of luminance values of pixels included in a predetermined local region centered on the target pixel is calculated for all the subband images for each subband image, Proceed to the next Step S27. In step S27, subband tone conversion characteristics are calculated for each subband image based on the histogram calculated in step S26.

  In the next step S28, a gradation conversion characteristic for the image signal is calculated based on the subband gradation conversion characteristic. Specifically, the weighting coefficient corresponding to the decomposition level is calculated for the subband gradation conversion characteristics of each subband image, and the subband gradation conversion characteristics are weighted and averaged according to the weighting coefficient, thereby obtaining an image. A gradation conversion characteristic for the signal is calculated. In the next step S29, the tone conversion unit 113 performs tone conversion processing on the image signal using the tone conversion characteristics calculated in step S28, and this routine is terminated. In the gradation conversion process, the gradation conversion characteristic calculated in step S28 is applied only to the specific luminance area of the image signal, and arbitrary decomposition is applied to the area other than the specific luminance area of the image signal. Subband tone conversion characteristics for level subband images can be applied.

  As described above, according to the present embodiment, gradation conversion can be performed in accordance with the size and luminance of the subject. For example, when there is a point light source having a small high luminance in the image signal, the point light source With respect to, gradation correction can be suppressed, and a natural image signal close to human visual characteristics can be obtained. Further, in the present embodiment, unlike the first embodiment, the decomposition level setting unit 202 does not need to compare the area S (i) of the specific luminance region and the area N of the local region, and can further reduce the amount of calculation. .

  In the first and second embodiments described above, processing by hardware is assumed, but it is not necessary to be limited to such a configuration. For example, it can be realized separately by software, in other words, a general-purpose or dedicated computer and a program operating on the computer. Such a computer includes, for example, a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), and a storage device, and the CPU or the like realizes all or part of the above processing. The above-described image processing apparatus reads out a program recorded on a computer-readable recording medium in which a program for recording is read, expands the program in a ROM, a RAM, and the like, and executes information processing / calculation processing The image processing by is realized.

  Here, the computer-readable recording medium means a magnetic disk, a magneto-optical disk, a CD-ROM, a DVD-ROM, a semiconductor memory, or the like. Alternatively, the computer program may be distributed to the computer via a communication line, and the computer that has received the distribution may execute the program.

109 Subband generation unit 110 Histogram calculation unit 111 Subband tone conversion characteristic calculation unit 112 Tone conversion characteristic calculation unit 113 Tone conversion unit 200 Decomposition unit 201 Specific luminance region detection unit 202 Decomposition level setting unit 209 Subband generation unit

Claims (9)

An image processing apparatus that performs gradation conversion processing on an image signal,
Subband image generation means for generating at least one subband image from the image signal;
Histogram calculation means for calculating a histogram of luminance values of pixels included in a predetermined local region centered on the target pixel with respect to the subband image for all pixels of the subband image;
Subband gradation conversion characteristic calculating means for calculating a subband gradation conversion characteristic for each subband image based on the histogram;
Gradation conversion characteristic calculating means for calculating gradation conversion characteristics of the image signal based on the subband gradation conversion characteristics;
Gradation conversion processing means for performing gradation conversion processing of the image signal based on the gradation conversion characteristics;
An image processing apparatus comprising: The subband image generation means includes
A specific luminance area detecting means for detecting, as a specific luminance area, an area composed of pixels having a predetermined luminance value among the pixels included in the image signal;
Based on the local area and the specific luminance area, the first subband image in which the specific luminance area is included in the local area and the ratio of the specific luminance area in the local area is maximized The image processing apparatus according to claim 1, wherein: The subband image generation means includes
Decomposition means for hierarchically subband decomposition of the image signal to a desired decomposition level to generate a plurality of subband images having different resolutions;
A specific luminance area detecting means for detecting, as a specific luminance area, an area composed of pixels having a predetermined luminance value among the pixels included in the image signal;
Decomposition level setting means for setting the decomposition level based on the shape or size of the specific luminance region;
The image processing apparatus according to claim 1, further comprising:   The decomposition level setting means, based on the local area and the specific luminance area, the specific luminance area is included in the local area, and the ratio of the specific luminance area in the local area is the maximum The image processing apparatus according to claim 3, wherein a decomposition level of the first subband image is set as the decomposition level. The subband gradation conversion characteristic calculating means includes:
Clipping means for performing a clipping process on the histogram based on the statistics of the histogram,
5. The image processing apparatus according to claim 1, wherein the subband tone conversion characteristics are calculated based on the histogram after the clipping process. 6.   The histogram statistic is an average value, median value, mode value, minimum value, maximum value, and variance value of the histogram of the local region of each subband image for each subband image. The image processing apparatus according to claim 5, wherein the image processing apparatus is provided. The gradation conversion characteristic calculating unit is configured to, based on the subband gradation conversion characteristic, a first gradation conversion characteristic for the specific luminance region of the image signal and a portion other than the specific luminance region of the image signal. Calculating a second gradation conversion characteristic for each region,
The gradation conversion processing means performs gradation conversion processing of the specific luminance area based on a first gradation conversion characteristic, and for areas other than the specific luminance area based on the second gradation conversion characteristic. The image processing apparatus according to any one of claims 1 to 6, wherein gradation conversion processing is performed.   The gradation conversion characteristic calculation means calculates a weighting coefficient corresponding to the decomposition level of the subband image for the subband gradation conversion characteristic, and calculates the subband gradation conversion characteristic based on the weighting coefficient. The image processing apparatus according to claim 1, wherein the first gradation conversion characteristic is calculated by weighted averaging. An image processing method for performing gradation conversion processing on an image signal,
Generating at least one subband image from the image signal;
Calculating a histogram of luminance values of pixels included in a predetermined local region centered on a target pixel with respect to the subband image for all pixels of the subband image;
Calculating a subband gradation conversion characteristic for each subband image based on the histogram;
Calculating a gradation conversion characteristic of the image signal based on the subband gradation conversion characteristic;
Performing gradation conversion processing of the image signal based on the gradation conversion characteristics;
An image processing method comprising:

Images