JP3424576B2 - Image processing apparatus and image processing method - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus and an image processing method, and can be applied to an image processing apparatus such as a television receiver, a video tape recorder, a television camera and a printer. According to the present invention, when a correction coefficient is generated according to an area to which input image data belongs and a pixel value is corrected, the operation is controlled so that the resolution of the correction coefficient corresponding to the pixel value of the image data is switched. This makes it possible to effectively avoid partial deterioration of contrast and correct the gradation.

[0002]

2. Description of the Related Art Conventionally, in an image processing apparatus such as a television camera, the gradation of image data obtained through image input means such as image pickup means is corrected and output.

FIG. 11 is a characteristic curve diagram showing the input / output characteristics of a signal processing circuit applied to this gradation correction processing.
This type of signal processing circuit reduces the gain when the input level 1 increases above a predetermined reference level lk. As a result, in this type of signal processing circuit, the input level 1 is the reference level lk.
When it further increases, the signal level is suppressed and output, and in this case, the gradation is corrected by sacrificing the contrast of the portion where the signal level is high.

In the characteristic curve diagram shown in FIG. 11, the horizontal axis represents the pixel value 1 which is the input level of the image data,
The vertical axis represents the pixel value T (l) which is the output level of the image data, and Lmax represents the maximum level that each pixel of the input / output image can take. In the following, the function showing the input / output relationship as shown in this characteristic curve diagram is called a level conversion function.

FIG. 12 is a characteristic curve diagram showing the input / output characteristics of the same type of signal processing circuit. The signal processing circuit using the level conversion function reduces the gain when the input level 1 is equal to or lower than the first reference level ls and when it is equal to or higher than the second reference level lb. As a result, the signal processing circuit corrects the gradation at the expense of the contrast between the low signal level portion and the high signal level portion.

On the other hand, in image processing using a computer, gradation is corrected by, for example, histogram equalization.

This histogram equalization is
This is a method of adaptively changing the level conversion function according to the frequency distribution of the pixel values of the input image, and is a method of correcting the gradation by reducing the gradation in the portion where the frequency distribution of the pixel values is low.

That is, as shown in FIG. 13, in this histogram equalization processing, a frequency distribution H which is the total number of pixels based on the pixel value 1 of the input image is used.
Based on (l), the cumulative frequency distribution C (l) is detected by the calculation processing of the following equation.

[0009]

[Equation 1]

In the histogram equalization processing, the cumulative frequency distribution C thus detected is
The level conversion function T (l) is defined by normalizing (l) by the processing of the following equation, and this level conversion function T
The signal level of the input image is corrected according to (l). Here, Fmax is the cumulative frequency distribution C
(L) is the final value, and Lmax is the maximum value of the input / output level.

[0011]

[Equation 2]

Note that such processing for correcting the gradation is performed even when image data is transmitted through a transmission path, displayed on a display device, or stored in a storage device.
For example, for the purpose of suppressing the dynamic range or the like, the processing is appropriately performed as needed.

[0013]

By the way, in the gradation correction processing by the above-mentioned conventional method, the entire gradation is corrected at the expense of the contrast of any part. This is because the level conversion is performed by an input / output function having a monotonic increasing property in order to avoid the generation of an unnatural image in any method.

Therefore, in the case of the conventional method, there is a problem that the contrast is partially lowered in the processed image.

The present invention has been made in consideration of the above points, and an object of the present invention is to propose an image processing apparatus and an image processing method capable of correcting gradation by effectively avoiding a partial decrease in contrast. It is a thing.

[0016]

In order to solve such a problem, in the present invention, in an image processing apparatus or an image processing method, a region to which image data belongs is determined, a determination result is output, and based on this determination result A correction coefficient for correcting the pixel value of the image data is output, and the pixel value of the image data is corrected according to this correction coefficient, and at this time, the resolution of the correction coefficient is switched according to the pixel value of the image data. .

The area to which the image data belongs is determined, the determination result is output, a correction coefficient for correcting the pixel value of the image data is output based on the determination result, and the pixel value of the image data is corrected according to the correction coefficient. For example, it is possible to correct the pixel value in the same area with the same coefficient, maintain the pixel value magnitude relationship in the area, and reverse the pixel value magnitude relationship between pixels belonging to different areas. It is possible to correct the overall gradation while avoiding the deterioration of the specific contrast.

In this way, in correcting the gradation, the contrast between the areas is determined by the correction means and the inclination of the level conversion function which is the input / output characteristic of the image data in the correction processing, and the spatial resolution of the determination result is determined. The higher the value, the greater the influence of the level conversion function on the gradation correction result. Therefore, when the pixel value corresponds to a portion with a small slope in the level conversion function, by increasing the resolution in the correction coefficient, even if the level conversion function does not maintain the monotonic increase property, the effect of this level conversion function is reduced. It is possible to reduce the unnatural change in the contrast between the adjacent regions and the natural contrast between adjacent regions.

[0019]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below in detail with reference to the drawings as appropriate.

(1) First Embodiment (1-1) Configuration of the First Embodiment FIG. 1 is a block diagram showing a television camera according to the first embodiment of the present invention. In this television camera 1, a CCD solid-state image pickup device (CCD) 2 outputs an image pickup result by driving a timing generator (TG) 3.

As shown in an enlarged front view of the image pickup surface in FIG. 2, this CCD solid-state image pickup element 2 has a complementary color Ichimatsu color filter arranged on the image pickup surface. That is, the CCD solid-state image sensor 2 has yellow (Ye) and cyan (Cy)
While the color filter of (1) is repeated for each pixel to form an odd line, the color filters of magenta (Mg) and green (G) are repeated for each pixel to form an even line.

As a result, in the CCD solid-state image pickup device 2, the correlated double sampling circuit associated with this type of image pickup device causes the amplitude-modulated color signals to be sequentially converted into luminance signals by time division as shown in FIG. The superposed image pickup result is output.

When outputting such an image pickup result, C
The CD solid-state image pickup device 2 obtains an image pickup result at a 1/60 [second] cycle according to the charge accumulation time set by the user, and outputs this image pickup result as an image pickup result VN by normal exposure.
Further, the CCD solid-state image pickup device 2 obtains an image pickup result by a charge accumulation time shorter than the charge accumulation time by the normal exposure in the vertical blanking period of the image pickup result VN by the normal exposure, and exposes the image pickup result for a short time. The image pickup result VS is output.

As a result, as shown in FIG. 4, in the CCD solid-state image pickup device 2, the image pickup result VN (FIG. 4A) by normal exposure in which the output level is saturated when the amount of incident light is equal to or more than a predetermined amount, Image pickup result V of short-time exposure in which the output level is not saturated due to the charge accumulation time shorter than V
S (FIG. 4 (B)) is combined and output.

The memory 4N inputs the image pickup result VN by the normal exposure through a correlated double sampling circuit, a defect correction circuit, an analog-digital conversion circuit and the like (not shown), and temporarily holds the image pickup result VN by the normal exposure. Output.

Similarly, the memory 4S inputs the image pickup result VS by this short-time exposure through a correlated double sampling circuit, a defect correction circuit, an analog-digital conversion circuit, etc., which are not shown, and the image-pickup result by this short-time exposure is inputted. Hold VS temporarily and output.

The adder circuit 5 adds a normal exposure image pickup result VN held in the memory 4N and a short time exposure image pickup result VS held in the memory 4S to obtain a wide dynamic range and a sufficient value. The level correction circuit 6 outputs the image pickup result VT by the pixel value, and the level correction circuit 6 outputs the image pickup result VS by the short-time exposure output from the memory 4S so as to ensure practically sufficient linearity in the image pickup result VT by the adder circuit 5. The pixel value of is corrected and output.

As a result, the television camera 1 is adapted to generate the image pickup result VT (FIG. 4C) having a dynamic range which is remarkably larger than that of the conventional one.

The gradation correction circuit 8 corrects the gradation of the image pickup result VT by correcting the pixel value of the image pickup result VT and outputs the corrected image. In the television camera 1,
The subsequent signal processing circuit (not shown) executes various signal processing required for the television camera and outputs the image pickup result to an external device or the like. At this time, the pixel value of the image pickup result is adjusted so as to correspond to the output device. By uniformly suppressing, the dynamic range of the imaging result is suppressed and output.

That is, in the gradation correction circuit 8, the area judgment filter 9 judges the area to which the input image data belongs, and outputs the judgment result.

That is, the area determination filter 9 inputs each pixel value x (i, j) of the imaging result VT to the first and second low pass filters (LPF) 9A and 9B, and limits the band here. As a result, the area determination filter 9 determines to which area of the average brightness level the input image data belongs in the low-pass filters 9A and 9B, respectively, and the low frequency components r0 (i, j) and r1 which are the determination results.
Output (i, j). Further, at this time, the region determination filter 9 simultaneously executes the processes in the low-pass filters 9A and 9B having different passband widths in parallel, so that the low-frequency component r0 (i, j), which is the determination result with different resolutions, is obtained. ) And r1 (i, j). In this embodiment, as shown in FIG. 5, with respect to the imaging result VT input in the raster scanning order, the horizontal direction is indicated by a subscript i and the vertical direction is indicated by a subscript j.

That is, each of the first and second low-pass filters 9A and 9B is a two-dimensional low-pass filter, and the pixel value x (i, j) of the image data sequentially input in the order of raster scanning is expressed by the following equation. The low frequency component r (i, j) represented by the arithmetic expression is detected, and this low frequency component r (i, j) is output as the determination result of each region.

[0033]

[Equation 3]

Note that N and M in the equation (3) are constants representing the size of the neighborhood area for calculating the average value. As a result, the first and second low-pass filters 9A and 9B remove a fine structure in the image from the image obtained by the imaging result VT to extract a region having a relatively flat pixel value, and determine the determination result with different resolutions at a low frequency. The components r0 (i, j) and r1
Output as (i, j). Low-pass filter 9A
1 and 9B are intended for such processing, it is desirable that their bandwidth is relatively narrow.

Further, the first and second low-pass filters 9
A and 9B are, compared to the second low pass filter 9B,
The constants N and M in equation (3) are selected so that the first low-pass filter 9A constitutes a low-resolution low-pass filter.

The weighting coefficient generation circuit 9C performs a calculation process according to the following equation with the first low pass filter 9A as a reference, and outputs the low frequency components r0 (i, j) output from the first and second low pass filters 9A and 9B. ) And r1 (i,
j) generate weighting factors 1-w and w.

[0037]

[Equation 4]

Here, Dmax and Dmin are constants for normalization. Further, wmax and wmin are the maximum value and the minimum value of the values calculated as the weighting coefficient, and both are given values of 0 or more and 1 or less in advance. The function D (l) is a function determined by the coefficient calculation function G used in the subsequent coefficient calculation circuit 11, and is defined by the following equation.

[0039]

[Equation 5]

As a result, the weighting coefficient generation circuit 9C is
When the corresponding pixel value x (i, j) corresponds to a portion where the slope of the level conversion function T (l) described later is small, the value of the weighting coefficient w is increased.

The multiplication circuits 9A and 9B use the weighting factors 1-w and w, respectively, to obtain the low frequency component r0 (i, j).
And r1 (i, j) are weighted, and the subsequent adder circuit 9
F adds up the weighting results from the multiplying circuits 9A and 9B and outputs the result as 1 area determination result r (i, j).

As a result, in the area determination filter 9, the weighted addition processing of the following equation is executed, and for the area corresponding to the portion where the slope of the level conversion function T (l) is large, the low frequency component band-limited by the high resolution. r1
The area determination result r of 1 is obtained by increasing the ratio of (i, j).
While (i, j) is output, conversely, for the region corresponding to the portion where the slope of the level conversion function T (l) is small, the low frequency component r0 (i, j j) is increased and the area determination result r of 1 is increased.
Output (i, j).

[0043]

[Equation 6]

As a result, in the area determination filter 9, the spatial resolution is switched according to the pixel value x (i, j) of the image pickup result VT, that is, the input / output characteristic of the coefficient calculation circuit 11 described later. Level conversion function T (l)
The smaller the slope of, the more the judgment result r
The determination result r (i, j) is generated so that the spatial resolution of (i, j) is reduced.

The coefficient calculation circuit 11 uses the low frequency component r
According to the signal level of (i, j), the contrast correction coefficient g is calculated by the coefficient calculation function G as shown in FIG.
Generate (i, j). Here, this coefficient calculation function G is
For example, the level conversion function T (l) described above with reference to FIG.
Is a function obtained by performing arithmetic processing on

[0046]

[Equation 7]

As a result, the coefficient calculation circuit 11 generates and outputs the contrast correction coefficient g (i, j) by the calculation processing of the following equation, and outputs the low frequency component r (i, j) which is the input level.
For the region where the signal level of j) is equal to or lower than the predetermined reference level lk, the contrast correction coefficient g (i, j) with a constant value gmax of 1 or more is output, and the reference level lk is output.
For the above region, the contrast correction coefficient g (i, j) is output so that the value gradually approaches the value gmin according to the signal level of the low frequency component r (i, j).

[0048]

[Equation 8]

The multiplication circuit 12 multiplies the pixel value x (i, j) of the image pickup result VT by using the contrast correction coefficient g (i, j) generated in this way to obtain the contrast correction coefficient g ( The signal level of the imaging result VT is corrected by i, j) and output.

(1-2) Operation of the First Embodiment In the configuration described above, in the television camera 1 (FIG. 1), the CDD solid-state image pickup device is provided by the color filters arranged on the image pickup surface (FIG. 2). The image pickup result in which the color signal amplitude-modulated by 2 is superimposed on the luminance signal by time division is output (FIG. 3).

In the television camera 1, the image pickup result VN (FIG. 4A) of normal exposure with the charge accumulation time set by the user and the image pickup result VS of short exposure with the short charge accumulation time (FIG. 4 ( B)) are alternately output, and the imaging results VN and VS are respectively stored in the memory 4
Held at N and 4S. With the television camera 1,
These two imaging results VN and VS are the level correction circuit 6,
The image pickup result VT is synthesized by the adder circuit 5 and has a remarkably large dynamic range as compared with the conventional one.
(FIG. 4C) is generated.

In this image pickup result VT, the low-pass filters 9A and 9B of the region judgment filter 9 judge the region to which the input image data belongs with different resolutions and generate the judgment result. More specifically, low-pass filters 9A and 9B having low-frequency components r0 (i, j) and r1 (i, j), which indicate the average luminance level, which is the average value of the pixel values x (i, j), are in different bands. , The fine structure in the image is removed, and a region having a relatively flat pixel value is extracted.

In the image pickup result VT, the two low frequency components r0 (i, j) and r1 (i, j) are combined by the weighted average circuit by the multiplication circuits 9D and 9E and the addition circuit 9F to obtain a low frequency of 1. The component r (i, j) is output as the determination result of each area.

In the image pickup result VT, the subsequent coefficient calculation circuit 11 generates the contrast correction coefficient g (i, j) according to the signal level of the low frequency component r (i, j), and this contrast correction coefficient g The pixel value of the imaging result is corrected in the multiplication circuit 12 by (i, j).

As a result, in the image pickup result VT, the pixel value is corrected by the equal gain in the region where the signal level of the low frequency component r (i, j) is equal, whereas
In a region where the signal level of the low frequency component r (i, j) is different, the pixel values can be made close to each other according to the setting of the level conversion function T (l), and in some cases, the pixel value may have a magnitude relationship. It is also possible to reverse. As a result, the contrast in each area can be naturally increased with respect to the entire gradation, and it is possible to correct the entire gradation by effectively avoiding a partial decrease in contrast.

That is, as shown in FIG. 7, the pixel value x
In the case where (i, j) pulsates at a frequency equal to or higher than the cutoff frequency of the region determination filter 10 which is a low-pass filter, and the DC level of the pixel value x (i, j) rises sharply (FIG. 7). (B)), when the change of the low frequency component r (i, j) corresponding to this rapid change of the DC level crosses the inflection point of the coefficient calculation function G (l) (FIG. 7 (A)), Depending on the conventional level conversion function described above with reference to FIG. A, the contrast is suppressed in the portion where the pixel value is large in the correction result Y (i, j) (FIG. 7C).

However, according to this embodiment, before and after the signal level of the low frequency component r (i, j) rapidly rises, gains corresponding to the signal level of the low frequency component r (i, j) are used. The pixel value x (i, j) is corrected, and the signal level is corrected by setting the coefficient calculation function G (l). At this time, the pixel value x (i,
In the portion where j) is small, the pixel value x (i, j) is corrected by the gain gmax due to the average value level 12 of the peak value 13 and the bottom value 11 so that the pixel value x (i, j) is comparable to the conventional method in the low level region. The contrast can be obtained (FIG. 7D).

On the other hand, on the high level side, similarly, the average value level 15 of the peak value 16 and the bottom value 14 is set.
The pixel value x (i, j) is corrected by the gain g5 by
At this time, the pixel value is corrected by the uniform gain of the peak value 16 and the bottom value 14 so that the contrast between the peak value 16 and the bottom value 14 is
It is amplified with this gain g5.

As a result, in the gradation correction circuit 8 according to the present embodiment, the gradation when viewed as a whole does not change significantly, but the pulsation seen microscopically is the result of the imaging of the input image. It is possible to expand the pulsation due to VT.

Similarly, as shown in FIG. 8, the pixel value x
In the case where (i, j) pulsates and the DC level rises sharply, a large change in the pixel value x (i, j) is biased to a higher level side than the inflection point of the coefficient calculation function G (l). (FIG. 8 (B)), all pixel values x (i, j) may be generated depending on the conventional level conversion function described above with reference to FIG.
The contrast is suppressed by (Fig. 8
(C)).

However, also in this case, on the low level side and the high level side, the average value levels l2 and l5, respectively.
Pixel values are corrected by gains g2 and g5 corresponding to
Although the gradation when viewed as a whole does not change significantly, the pulsation that is viewed microscopically is the imaging result V that is the input image.
It becomes possible to expand the pulsation due to T (Fig. 8).
(D)).

In correcting the gradation of the image pickup result VT in this way, the gradation correction result y is obtained in a constant region in which the value of the low frequency component r (i, j) is maintained at a substantially constant value.
The contrast at (i, j) is determined by the value of the correction coefficient given by the coefficient calculation function G. On the other hand, the contrast between regions having different low frequency components r (i, j) is determined by the slope of the level conversion function. The size of each region corresponds to the spatial resolution of the correction coefficient g (i, j) and is determined by the pass band width of the low-pass filter that extracts the low frequency component r (i, j).

From these facts, the low frequency component r (i,
The wider the pass band width of the low-pass filter for extracting j), that is, the higher the resolution of the correction coefficient, the greater the influence of the level conversion function on the contrast in the gradation correction result y (i, j).

As a result, the area to which the predetermined pixel value belongs is
When the level conversion function corresponds to a portion with a small slope, gradation correction is performed by the area determination result obtained by using a low-pass filter with a narrow pass band width, that is, the resolution of the correction coefficient corresponding to this determination result is improved. If gradation correction is performed by using this method, even if the level conversion function does not maintain a monotonic increase, it is possible to reduce the effect of this level conversion function and reduce unnatural contrast changes between adjacent areas. It will be possible.

On the other hand, when the area to which the predetermined pixel value belongs corresponds to a portion having a large slope in the level conversion function, sufficient contrast is guaranteed by the slope of the level conversion function in the contrast around this pixel. It will be. In this case, in order to improve the accuracy of gradation correction of edges and the like in the gradation correction result, a correction coefficient with a higher resolution, that is, a region judgment result obtained by using a low-pass filter with a wide pass band width, provides a highly accurate step. The key adjustment result can be obtained.

As a result, in this embodiment, the weighting coefficient generation circuit 9C uses the low-frequency component r0 (i, j) having a low resolution to generate the level conversion function T.
For the area corresponding to the large slope of (l),
Low frequency component r1 (i, i,
The area determination result r (i, j) is output by increasing the ratio of j), and conversely, for the area corresponding to the portion where the slope of the level conversion function T (l) is small, the bandwidth is limited due to the low resolution. The region determination result r (i, j) is output by increasing the ratio of the low frequency component r0 (i, j).

That is, in the input / output characteristics of the multiplication circuit 12, the determination result r (i, j) is set so that the spatial resolution is reduced in the area corresponding to the portion where the slope of the level conversion function T (l) is small. Is generated.

As a result, in the gradation correction circuit 8, a correction coefficient having a low resolution is provided for an area having a small gradient of the level conversion function, and an area having a large gradient of the level conversion curve. A high-resolution correction coefficient is generated for the, and the gradation is corrected by the correction coefficient generated in this way, which makes it possible to secure a natural contrast even between adjacent areas. It is possible to correct the gradation.

(1-3) Effects of First Embodiment According to the above configuration, the correction coefficient is generated based on the determination result of the area to which the input image data belongs, and the imaging result is corrected according to this correction coefficient. By doing so, it is possible to make the pixel values close to each other in the same region as needed while maintaining the magnitude relationship of the pixel values with the same coefficient, and to reverse the pixel values in extreme cases. You can also This makes it possible to effectively avoid partial deterioration of contrast and correct the gradation.

At this time, the determination results having different resolutions are combined, a correction coefficient having a low resolution is assigned to an area constituted by the level conversion curve having a small slope, and a level conversion curve having a large slope is allocated. By assigning a high-resolution correction coefficient to the formed area, it is possible to secure a natural contrast even between adjacent areas, and it is possible to more naturally correct the gradation.

Further, at this time, the pixel value of each area is corrected with reference to the low frequency component by the low-pass filter, so that the overall gradation can be corrected with a simple structure while avoiding a partial decrease in contrast. You can

(2) Second Embodiment FIG. 9 is a block diagram showing a gradation correction circuit applied to a television camera according to the second embodiment of the present invention. The gradation correction circuit 18 is applied instead of the gradation correction circuit 8 described above with reference to FIG. In this gradation correction circuit 18, the same components as those of the gradation correction circuit 8 described above are denoted by the corresponding reference numerals, and the duplicated description will be omitted.

In the gradation correction circuit 18, the area judgment filter 19 judges the areas to which the pixel value x (i, j) belongs with different resolutions. The judgment results r0 (i, j), r
1 (i, j) is output.

That is, the area determination filter 19 is a low-pass filter (LPF) 19 having different pass bandwidths.
Each of the low-pass filters 19 is composed of A and 19B.
The pixel value x (i, j) is given to A and 19B, and the corresponding low frequency components are determined as the determination results r0 (i, j) and r1 (i, j).
Output as.

Here, the low-pass filters 19A and 19B
Are the low-pass filters 9 described above with reference to FIG. 1, respectively.
It is formed the same as A and 9B.

The coefficient calculation circuit 21 determines the judgment result r0 (i,
j) and r1 (i, j), the corresponding correction coefficient g0 (i, j
j) and g1 (i, j) are generated, and these correction coefficients g0 (i, j) and g1 (i, j) are combined to generate a correction coefficient g (i, j) of 1.

That is, in the coefficient calculation circuit 21, the coefficient calculation units 21A and 21B respectively determine the determination result r0 (i, i) based on a predetermined coefficient calculation function Gk (k = 0, 1).
j) and r1 (i, j), the corresponding correction coefficient g0 (i, j
j) and g1 (i, j) are generated and output.

The weighting coefficient generation circuit 21C executes the same arithmetic processing as described above for the equation (4) with the pixel value x (i, j) of the image pickup result VT as a reference. As a result, the weighting coefficient generation circuit 21C reduces the value of the weighting coefficient w when the pixel value x (i, j) corresponding to the inclination of the level conversion function is small.

The multiplication circuits 21A and 21B use the weighting factors 1-w and w, respectively, to correct the correction factor g0 (i,
j) and g1 (i, j) are weighted, and the subsequent adder circuit 21F adds the weighted results of the multiplier circuits 21A and 21B and outputs the result as a correction coefficient g (i, j) of 1.

As a result, in the coefficient calculation circuit 21,
The correction coefficient g0 (i, j) according to the pixel value x (i, j),
The spatial resolution of the correction coefficient g (i, j) is switched by operating g1 (i, j).

That is, the pixel value x
The correction coefficient g (i, j) is reduced so that the spatial resolution of the correction coefficient g (i, j) decreases as (i, j) falls.
To generate.

More specifically, the coefficient calculation circuit 21 generates a low frequency component r1 (i, j) whose band is limited by a high resolution for a region corresponding to a large slope of the level conversion function T (l). Correction coefficient g1 (i, j)
The correction coefficient g (i, j) of 1 is increased by increasing the ratio of 1 to 1. On the contrary, in contrast to this, in the region corresponding to the portion where the slope of the level conversion function T (l) is small, the low resolution By increasing the ratio of the correction coefficient g0 (i, j) generated from the low frequency component r0 (i, j) band-limited by
The correction coefficient g (i, j) of is output.

According to the configuration shown in FIG. 9, even if the correction coefficients with different resolutions are generated and then combined to correct the gradation, the same effect as that of the first embodiment can be obtained. .

(3) Other Embodiments In each of the above-described embodiments, a case has been basically described in which the correction coefficient is generated based on the characteristics described above with reference to FIG. 6, but the present invention is not limited to this. Alternatively, the correction coefficient may be generated according to various input / output characteristics.
It is also possible to use a level conversion function with an input / output characteristic such that the output level decreases along the way as the input level increases, as shown in FIG.

That is, in the conventional method, when such a function is used, since this function is not a monotonically increasing function, a pseudo contour may occur in the image as a processing result. However, in the case where the area determination is performed by the low-pass filter as in the above-described embodiment and the processing is performed, the pixel value of the pixel value such that the magnitude relationship of the pixel values is reversed in the neighboring area having the size corresponding to the pass band of the low-pass filter. Change can be prevented. As a result, it is possible to effectively avoid the occurrence of pseudo contours.

Further, in the above-mentioned embodiment, the case where the coefficient calculating function G is generated by the calculation processing of the expression (7) using the level converting function T has been described, but the present invention is not limited to this, and the level converting function is not limited thereto. The coefficient calculation function G may be arbitrarily set without using the function T.

Further, in the above-described embodiment, the case where the gradation is corrected by the gradation correction circuit and the dynamic range is suppressed by the subsequent signal processing circuit has been described, but the present invention is not limited to this, and the level conversion is performed. These processes can be collectively executed by setting the function T and the coefficient calculation function G corresponding thereto.

That is, in the processing of suppressing the dynamic range, it is required that the number of bits of the output pixel value is smaller than the number of bits of the input pixel value, so that the maximum value of the output level in the level conversion function T. Is set to the maximum value allowed for the output image, and the coefficient calculation function G is generated using this, whereby these processes can be collectively executed.

When the coefficient calculation function G is arbitrarily set without using the level conversion function T, the coefficient calculation function G may be set so as to satisfy the following equation. Here, j
Represents the input pixel level, Lmax represents the maximum value of the input pixel level, and L0max represents the maximum value of the output pixel level.

[0090]

[Equation 9]

Further, in the above-mentioned embodiment, the case where the output of the low-pass filter of two systems is adaptively processed to generate the correction coefficient has been described. However, the present invention is not limited to this, and a plurality of systems may be provided as necessary. It is possible to obtain the same effect as that of the above-described embodiment by adaptively processing the determination result of 1 to generate the correction coefficient.

Further, in the above-mentioned first embodiment,
The case of synthesizing low-pass filter outputs of two systems has been described, but the present invention is not limited to this, and by adaptively changing the pass band width of the low-pass filter of one system according to the input image data, the correction coefficient is set. The resolution of the corresponding determination result may be switched.

In the above-described embodiment, the case where the image pickup result in which the amplitude-modulated color signal is sequentially superimposed on the luminance signal by time division is directly processed is described.
The present invention is not limited to this, and after separating the luminance data from such an imaging result to generate a correction coefficient, when correcting the gradation of the imaging result by this correction coefficient, and further when processing a color signal, The present invention can be widely applied to a case of processing a video signal based on a luminance signal and a color difference signal, and further a case of processing a composite video signal in which a chroma signal is superimposed on a luminance signal.

For the composite video signal, a correction coefficient is generated based on the luminance signal generated by YC separation,
By correcting the gradation of the luminance signal and chroma signal, the gradation of the luminance signal and the color difference signal by this correction coefficient, the gradation of the video signal of this kind can be corrected and the video signal by the luminance signal and the color difference signal is processed. In this case, similarly, the correction coefficient is calculated based on the luminance signal, and the gradation of the luminance signal and the color difference signal is corrected by this correction coefficient, whereby the gradation of the video signal of this type can be corrected.

When a color signal is further processed, a luminance signal is generated by the calculation processing of the following equation, and then a correction coefficient is calculated based on this luminance signal, and the gradation of each color signal is corrected by this correction coefficient. This makes it possible to correct the gradation of this type of video signal.

[0096]

[Equation 10]

In the above embodiment, the case where the area to which the input image data belongs is determined by the low-pass filter using the average luminance level as the feature amount of the image data has been described, but the present invention is not limited to this. , For example, in the case of grasping the similarity between an arbitrarily selected pixel and neighboring pixels surrounding the pixel as a feature amount in the image to be processed, and sequentially enlarging the region from this pixel to determine the region to be processed It is possible to obtain the same effect as that of the above-described embodiment by determining the area of the image to be processed by various feature amounts and various processing methods.

Further, in the above-mentioned embodiments, the case where the present invention is applied to the television camera has been described, but the present invention is not limited to this, and various images such as a television receiver, a video tape recorder, a printer and the like. It can be widely applied to processing devices.

[0099]

As described above, according to the present invention, when the correction coefficient is generated based on the determination result of the area to which the input image data belongs and the pixel value is corrected, the correction is performed according to the pixel value of the image data. By controlling the operation so that the spatial resolution of the correction coefficient to be switched is changed, the gradation can be corrected by effectively avoiding the partial deterioration of the contrast. The contrast can be secured.

[Brief description of drawings]

FIG. 1 is a block diagram showing a television camera according to a first embodiment of the present invention.

2 is a plan view showing a color filter of a CCD solid-state image sensor of the television camera of FIG.

FIG. 3 is a signal waveform diagram showing an image pickup result by the color filter of FIG.

FIG. 4 is a characteristic curve diagram for explaining processing of an imaging result in the television camera of FIG.

5 is a schematic diagram showing an array of pixels in the television camera of FIG.

FIG. 6 is a characteristic curve diagram for explaining a contrast correction coefficient g (i, j).

7 is a signal waveform diagram for explaining processing of a gradation correction circuit in the television camera of FIG.

FIG. 8 is a signal waveform diagram for explaining the processing of the gradation correction circuit at an input level different from that of FIG.

FIG. 9 is a block diagram showing a gradation correction circuit applied to a television camera according to a second embodiment of the present invention.

FIG. 10 is a characteristic curve diagram for explaining a level conversion function applied to a gradation correction circuit according to another embodiment.

FIG. 11 is a characteristic curve diagram for explaining a level conversion function applied to a conventional gradation correction suppression process.

FIG. 12 is a characteristic curve diagram for explaining a level conversion function applied to a gradation correction process according to another example different from FIG.

FIG. 13 is a characteristic curve diagram for explaining a process of histogram equalization.

[Explanation of symbols]

1 ... Television camera, 8, 18 ... Tone correction circuit, 9, 19 ... Area determination filter, 9A, 9B, 19
A, 19B ... Low-pass filter, 9C, 21C ... Weighting coefficient generation circuit, 9D, 9E, 12, 21D, 21
E ... Multiplier circuit, 9F, 21F ... Adder circuit, 11 ...
Coefficient calculation circuit

─────────────────────────────────────────────────── ─── Continuation of the front page (56) Reference JP 62-226772 (JP, A) JP 3-158077 (JP, A) JP 4-15783 (JP, A) JP 6- 311392 (JP, A) JP 7-170428 (JP, A) JP 9-312787 (JP, A) JP 10-13854 (JP, A) JP 10-248024 (JP, A) (58) Fields investigated (Int.Cl. ⁷ , DB name) H04N 5/14-5/217 H04N 9/04-9/11 H04N 9/44-9/78

Claims (14)

(57) [Claims] 1. An image processing apparatus for correcting the gradation of image data, comprising: an area determining unit that determines an area to which the image data belongs and outputs a determination result; and a pixel of the image data based on the determination result. A coefficient calculation unit that outputs a correction coefficient that corrects a value; and a correction unit that corrects a pixel value of the image data according to the correction coefficient, wherein the area determination unit or the coefficient calculation unit is a pixel of the image data. An image processing apparatus, wherein the determination result or the correction coefficient is generated so that the resolution of the correction coefficient is switched according to a value. 2. An image processing method for correcting gradation of image data, comprising: a region determination process of determining a region to which the image data belongs and outputting a determination result; and determining a pixel value of the image data according to the determination result. A coefficient calculation process for outputting a correction coefficient to be corrected, and a correction process for correcting the pixel value of the image data according to the correction coefficient, in the area determination process or the coefficient calculation process, the pixel value of the image data Accordingly, the determination result or the correction coefficient is generated so that the resolution of the correction coefficient is switched. 3. In the area determination process or the coefficient calculation process, in the input / output characteristic of the image data in the correction process, the smaller the change in the output value with respect to the change in the input value,
The determination result or the correction coefficient is generated so that the resolution of the correction coefficient is reduced.
The image processing method described in. 4. The area determination process detects a feature amount indicating a feature in a predetermined range near the image data and outputs the determination result, and the coefficient calculation process outputs the correction coefficient according to the feature amount. The image processing method according to claim 2, further comprising: 5. The area determination process according to claim 2, wherein the resolution of the correction coefficient is switched by changing the resolution of the determination result according to the pixel value of the image data. Image processing method. 6. The coefficient calculation process according to claim 2, wherein the resolution of the correction coefficient is switched by correcting the correction coefficient according to a pixel value of the image data. The described image processing method. 7. The image processing method according to claim 2, wherein in the area determination processing, a low frequency component of the image data is extracted and the determination result is output. 8. The area determination processing comprises a signal extraction processing for extracting low frequency components of the image data with different passband widths, and combining the plurality of low frequency components in accordance with the image data to obtain the signal. The image processing method according to claim 2, further comprising a signal synthesizing process for generating a determination result. 9. The image processing method according to claim 8, wherein in the signal combining process, the determination result is generated by performing a weighted average of the plurality of low frequency components. 10. The area determination process extracts low frequency components of the image data with different pass bandwidths and outputs the determination result, and the coefficient calculation process uses the plurality of low frequency components respectively. The image processing method according to claim 2, comprising a partial coefficient calculation process for generating a correction coefficient, and a coefficient synthesizing process for generating the correction coefficient based on the correction coefficient. 11. The image processing method according to claim 2, wherein in the correction processing, the pixel value of the image data is corrected by multiplying the pixel value of the image data by the correction coefficient. 12. The image processing method according to claim 2, wherein the number of bits of the image data obtained by the correction processing is reduced as compared with the number of bits of the input image data. 13. The image data according to claim 2, wherein the image data is data obtained by sampling at a predetermined frequency a signal in which an amplitude-modulated color signal is sequentially superimposed on a luminance signal by time division. Processing method. 14. The image processing method according to claim 2, wherein the image data is data obtained by sampling a luminance signal and a color difference signal at a predetermined frequency.