JPH10294911A - Method and device for compressing dynamic range of picture - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent a picture from being painted out even in the case that a difference between a bright part and a dark part is large by eliminating a density level where the number of picture elements is less than a prescribed value, shifting the density level to one side for an eliminated part and obtaining picture data. SOLUTION: In a histogram computing circuit 38, the histograms for respective colors R and G and B of the picture data are prepared. In a density conversion circuit 44, the density level where the number of picture elements is less than the prescribed value is eliminated, the density level where the number of picture elements is more than the prescribed value is shifted to the lower part of the density level for the eliminated density level part and the picture data provided with a second density level range are prepared. In a full density average conversion circuit 45 and a reference density conversion circuit 46, a reference density level in a third density level range compressed based on second picture data is set, a compression coefficient in the range of the upper and lower density levels is computed and obtained and the density level after compression is computed and obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for compressing a dynamic range of an image, and further measures a physical quantity to be measured which changes with time such as voltage or current of an electric signal, temperature and humidity. A method and apparatus for compressing the dynamic range of a detector.

[0002]

2. Description of the Related Art A technique for compressing a dynamic range of an image is required, for example, for displaying and outputting wide dynamic range image data by a cathode ray tube or a liquid crystal display device having a smaller dynamic range. In the prior art dynamic range compression, for example, as shown in FIG. 26, the density level range of the wide dynamic range image data of FIG. 26 (1) is linearly narrowed as shown in FIG. Compressed to range image data. In this prior art, the brightness or illuminance of the subject on the left side of FIGS. 26 (1) and 26 (2) is small, and in a dark part, a change in density level is not sufficiently expressed, so that a solid image is obtained. turn into. Similarly, FIG.
Also in the part (1) and the right side of FIG. 26 (2) where the luminance of the subject is large and bright, the change in the density level is not sufficiently expressed, and the image looks like a solid.

[0003]

SUMMARY OF THE INVENTION An object of the present invention is to maintain the relative relationship between the brightness, that is, the density level even after the dynamic range compression, and to provide a case where the difference between a bright portion and a dark portion is large. It is still another object of the present invention to provide a method and apparatus for compressing the dynamic range of an image, which can prevent the image from becoming a solid image. It is another object of the present invention to provide a method and apparatus capable of compressing the dynamic range of a detector while maintaining the relative relationship between local changes in the value of a physical quantity to be measured.

[0004]

SUMMARY OF THE INVENTION According to the present invention, there is provided a method for removing a first density level in which the number of pixels in a first image data in which the density level of each pixel falls within a first density level range is less than a predetermined value. The density level whose number of pixels is equal to or greater than the predetermined value is shifted to one side of the density level by the density level deleted within the level range, and the pixel having the deleted density level has one of the above-mentioned ones. By setting the density level shifted to the side, the second image data having the second density level range is obtained, and the average value Em of the density levels of the first image data is set to the assigned value Zm1, within the second density level range. Z
m2, and set the values Zm1 and Zm2.
In correspondence with the predetermined reference density level Cc in the third density level range CL to CH which is smaller than the first density level range of the first image data, the upper and lower density levels assigned values Zm1 and Zm2 of the second image data are changed. This is a method for compressing a dynamic range of an image, wherein each range is allocated to a range above and below a density level related to a reference density level Cc of a third density level range. Further, according to the present invention, each pixel obtains first image data having a plurality of density levels within a first density level range.
The average value Em of the density levels of the image data is obtained, and the density level of the first image data whose number of pixels is less than a predetermined value is deleted. The above density levels are shifted toward the lower or higher density level, and the pixels having the deleted density levels are set to the shifted density, and the second density level is set. Obtaining the second image data having the range, and when the number of pixels having the average value Em is less than the predetermined value, the deleted pixels including the average value Em within the second density level range of the second image data. The density level on the side shifted by the density level of
m, and when the number of pixels having the average value Em is equal to or larger than the predetermined value, the second image data having the average value Em within the second density level range of the second image data is set.
The density level of the pixel of the image data is set to an assigned value Zm2, and these assigned values Zm1 and Zm2 are set to a predetermined reference density level Cc in a third density level range CL to CH narrower than the first density level range. Assigned values Zm1, Z in the second density level range ZL to ZH
The density level in each range between m2 and the values ZL and ZH at both ends of the density level in the second density level range is referred to as a reference density level Cc in the third density level range, and the density in the third density level range. This is a method for compressing the dynamic range of an image, which is set by distributing and setting the density level in each range between the values CL and CH at both ends of the level. According to the present invention, the number of pixels in the first image data, which is the first density level range that is a wide dynamic range whose dynamic range is to be compressed, is less than a predetermined value;
For example, a density level that is zero or less than a value close to zero is deleted, and then, by the deleted density level, on one side of the density level, that is, on the dark side where the density level is small or on the bright side where the density level is large. Stagger. The density level shifted to the one side is set to the pixel having the deleted density level. Thus, the second image data having the second density level range is created. An average value Em of the density levels of the first image data before the deletion of the density levels is calculated and obtained.
Within the second density level range of the second image data,
It is set by allocating to the allocation values Zm1 and Zm2. in this case,
When the average value Em is the density level of the number of pixels less than the predetermined value, the density on the shifted side after the density level is shifted to the one side is set as the assigned value Zm1. . When the average value Em is a density level having the number of pixels equal to or larger than the predetermined value,
A density level corresponding to the average value Em in the range of the shifted density level in the second image data is set to the assigned value Zm2. Then, the allocation values Zm1 and Zm2
Is set to a predetermined reference density level Cc in a third density level range CL to CH which is a narrow dynamic range after compression, for example, the third density level range CL to C
H, corresponding to the central density level, and with respect to the reference density level Cc, the assigned values Zm1 and Zm2 and the lower limit value in the second density level range are set in the range of the lower limit value CL of the third density level range CL to CH. Are similarly set and the range between the reference density level Cc and the upper limit value CH within the third density level range CL to CH is set to the second density level.
The allocation is performed between the assigned values Zm1 and Zm2 in the density level range and the upper limit value of the second density level range.
Therefore, the density level having a frequency of zero or small is deleted, and the relative relationship of the brightness is maintained.
Even if the difference between the light and dark areas is large,
The density change in each part is expressed, and therefore, a solid image is not obtained. In this way, a change in density level is expressed in each of the bright portion and the dark portion, and an apparently natural image having a gradation change in density level can be obtained. Furthermore, according to the present invention, even in the case of a wide dynamic range image having a bias toward darker or brighter as a whole,
An easily viewable dynamic range compressed image converted to an appropriate level is obtained. In this way, for a mountain having a density distribution that is separated, a white image and a black image are prevented from being lost during compression of a wide dynamic range image, and a smooth image in which a change in gradation is emphasized can be obtained. In this way, an effective compression technique for showing an image with a natural gradation is realized.

[0005] Further, the present invention provides
With respect to the reference density level Cc in the third density level range, the first compression coefficient kL, kL = (Cc-CL) / (Zm-ZL) in the lower density level range is obtained, and the reference density level in the third density level range is obtained. C
For c, the second compression factor k in the range of the above density levels
H, kH = (CH-Cc) / (ZH-Zm), and the density level C within the third density level range is calculated as the second density level.
If the density level Z in the second density level range ZL to ZH of the image data is in a range where Z <Zm, C = Cc-kL (Zm-Z), and the density in the second density level range ZL to ZH is obtained. 3. The method for compressing the dynamic range of an image according to claim 2, wherein C = Cc-kH (Z-Zm) in a range where the level Z satisfies Z ≧ Zm. According to the present invention, the reference density level Cc in the third density level range is lower than the reference density level Cc in the upper and lower density levels.
And the second compression coefficients kL and kH are determined, and the reference density level Cc is calculated using the first and second compression coefficients kL and kH.
The distribution of the density level in the range of each of the lower and upper parts can be performed.

Further, according to the present invention, the first image data includes red R,
Color image data for each color of green G and blue B;
The average value Em is an average value of the density levels of all colors. According to the present invention, when the first image data is color image data for each of the colors R, G, and B, the average value Em is used.
Is the average value of the density levels of the image data for each of the colors R, G, and B. This prevents the image from becoming a filled image of each of the colors R, G, and B, and the floor for each of the colors R, G, and B. It is possible to obtain smooth image data in which tone change is emphasized.

The present invention also provides (a) a memory for storing first image data in which each pixel has a plurality of density levels within a first density level range, and (b) a first image stored in the memory. Means for calculating an average value Em of the density levels of the data; and (c) deleting, from the first image data, the density levels whose number of pixels is less than a predetermined value, The density level whose number of pixels is equal to or greater than the predetermined value is shifted to the lower or higher density level side, and the density of the shifted side is set for pixels having the deleted density level. Means for creating second image data having a second density level range; and (d) in response to the output of the second image data creating means, the number of pixels having an average value Em is less than the predetermined value. When Within the second density level range of the two image data, the density level on the side shifted by the density level of the deleted pixel including the average value Em is set to the assigned value Zm1 of the average value Em, and the average value Em is set. Is greater than or equal to the predetermined value, the second image data of the second
Assignment value setting means for setting the density level of the pixel of the second image data having the average value Em to the assignment value Zm2 within the density level range; (e) second image data creation means;
In response to each output with the assigned value setting means, these assigned values Z
m1 and Zm2 are set corresponding to a predetermined reference density level Cc in a third density level range CL to CH which is smaller than the first density level range, and a second density level range ZL is set.
To ZH, the density levels in the respective ranges between the values ZL and ZH at both ends of the density levels in the second density level range, the reference density levels Cc in the third density level range, Means for distributing and setting the density levels in the respective ranges between the values CL and CH at both ends of the density levels in the third density level range to create third image data. This is a device for compressing the dynamic range. According to the present invention, the first image data having a wide dynamic range is read out from the memory, the second image data is created based on the histogram, and the average value Em of the density levels of the first image data before deletion is obtained. Within the range of the density level of the second image data, the third dynamic range, which is a narrow dynamic range to be set and assigned to the assigned values Zm1 and Zm2 and compressed.
The reference density levels Cc of the density level ranges CL to CH are set corresponding to the assigned values Zm1 and Zm2. Within the third density level ranges CL to CH, the lower and upper density level ranges with respect to the reference density level Cc are set. Then, the density levels of the allocation values Zm1 and Zm2 in the second density level range are allocated so as to correspond to the lower and upper ranges, thus creating the third image data. In this way, it is possible to automatically compress wide dynamic range image data into narrow dynamic range image data.

Further, according to the present invention, the third image data creating means includes: (a) a first compression coefficient k in a lower density level range with respect to the reference density level Cc in the third density level range;
L, kL = (Cc-CL) / (Zm-ZL), a first compression coefficient calculating means, and (b) a second compression of the upper density level range with respect to the reference density level Cc in the third density level range. A second compression coefficient calculating means for calculating coefficients kH, kH = (CH-Cc) / (ZH-Zm); and (c) responding to each output of the first and second compression coefficient calculating means, and In the range where the density level Z in the second density range ZL to ZH is Z <Zm, C = Cc-kL (Zm-Z), and the density level C in the second density range ZL to ZH is obtained. In a range where the density level Z satisfies Z ≧ Zm, a compression density level calculating means for obtaining C = Cc−kH (Z−Zm) is included. According to the present invention, the first and second compression coefficients k1 and k2 are calculated and obtained in the same manner as in the above-described claim 3, and the lower and upper reference density levels Cc in the third density level ranges CL to CH are obtained. Allocation to each concentration level range is automatically enabled.

Further, according to the present invention, the first image data includes red R,
Color image data for each color of green G and blue B;
The average value calculation means is an average value of density levels of color image data of all colors. According to the present invention, similarly to the above-described fourth aspect of the present invention, image data having a gradation change of a density level that maintains a relative relationship of brightness for each of the colors R, G, and B of the color image data. Can be obtained by compression of the dynamic range.

The present invention also relates to (a) extending image data of a plurality of screens obtained by imaging the same subject with different exposure amounts to obtain a single image data having an extended dynamic range; Of the image data having different amounts, the density levels at both ends or one end of the dynamic range of the reference image data in the same object portion are adjusted to the density levels α1 and α2 of the image data having different exposure amounts. , The amount of change in the density level (F
b-Fa, Fd-Fc), the amount of change (F2b-F2a, F1d-) of the density level of image data having different exposure amounts.
(B) correcting the one image data with the other image data by using the exposure coefficient β2m, β1m which is the ratio of F1c) to extend the dynamic range of the image;
The image data with the extended dynamic range obtained as described above is used as the first image data, and when the dynamic range of the first image is compressed, the number of pixels in the first image data in which the density level of each pixel is the first density level range Is deleted, and the first level is deleted.
The density level whose number of pixels is equal to or more than the predetermined value is shifted to one side of the density level by the density level deleted within the density level range, and the pixel having the deleted density level has , The second image data having the second density level range is obtained, and the average value Em of the density levels of the first image data is calculated as follows.
The assigned values Zm1 and Zm2 in the second density level range are assigned and set, and the assigned values Zm1 and Zm2 are set in advance in the third density level ranges CL to CH that are smaller than the first density level range of the first image data. Assigned values Zm1 and Zm of the density level of the second image data corresponding to the determined reference density level Cc.
An image processing method characterized by distributing upper and lower ranges with respect to m2 to respective ranges above and below a density level with respect to a reference density level Cc of a third density level range. The present invention also provides: (a) a first memory for storing first image data having a plurality of density levels captured at a certain exposure amount;
(B) a second memory storing second image data having a plurality of density levels obtained by imaging the same subject as the first image data with an exposure amount larger than the exposure amount of the first image data; and (c). Within the dynamic range of the density level of the first image data stored in the first memory, out of the plurality of predetermined density levels Fa and Fb on the side corresponding to the larger exposure amount, the outside of the dynamic range The coordinates of the pixels of the first memory having the density level Fa closer to the outside and the density level further outward and the density level Fb inside the predetermined outward density level Fa are defined as follows. (D) second extraction means for extracting a pixel in the second memory having the same coordinates as the coordinates extracted by the first extraction means, and (e) second extraction means. Means for calculating average values F2a and F2b of density levels of pixels extracted from the second memory for each of the predetermined density levels Fa and Fb in response to the output of the means; In response to the output of the arithmetic means, the density level of the second image data is adjusted to the density level of the first image data. The density level differences α2 and α2 with respect to the density level Fa outside the dynamic range for adjusting the density level. = Fa−F2a, and the exposure coefficient β2m, β2m, which is the ratio of the gradient of the density level of the first image data to the gradient of the density level of the second image data = (Fb−Fa) / (F2b−F2a) And (g) responding to the output of the second extracting means, and determining the predetermined density level Fa,
When the density level of the pixel of the second memory having the same coordinates as the coordinates of each pixel having the density level further outside the dynamic range than Fb is F2old, the density levels F2new, F2new of the image data after correction = Fa + (F2old + α2-Fa) × β
2h, and (h) the corrected image data obtained by the second arithmetic means is first image data in which each pixel has a plurality of density levels within the first density level range. Yes, the third storing the first image data
A memory; (i) an average value calculating means for calculating an average value Em of density levels of the first image data stored in the third memory; and (j) a number of pixels of the first image data whose number is less than a predetermined value. Is deleted, the density level whose number of pixels is equal to or more than the predetermined value is shifted toward the lower or higher density level by the deleted density level, and the deleted density level is deleted. Means for setting a density on the shifted side for a pixel having a level to generate second image data having a second density level range;
(K) responding to the output of the second image data generating means, when the number of pixels having the average value Em is less than the predetermined value, the average value Em is set within the second density level range of the second image data. The density level on the side shifted by the density level of the deleted pixel is set to the assigned value Zm1 of the average value Em, and when the number of pixels having the average value Em is equal to or greater than the predetermined value, the second image Assignment value setting means for setting the density level of the pixel of the second image data having the average value Em to the assignment value Zm2 within the second density level range of the data; (l) second image data creation means; In response to each output with the assigned value setting means, these assigned values Zm1, Zm2
Is set in correspondence with a predetermined reference density level Cc in a third density level range CL to CH narrower than the first density level range, and assigned values Zm1 and Zm2 in the second density level ranges ZL to ZH are set. The density level in each range between the values ZL and ZH at both ends of the density level in the second density level range is defined as a reference density level Cc in the third density level range, and the density level in both ends of the third density level range. Means for generating third image data by distributing and setting density levels in respective ranges between values CL and CH. The present invention also provides (a) a first memory for storing first image data having a plurality of density levels captured at a certain exposure amount, and (b) an exposure amount smaller than the exposure amount of the first image data. A second memory for storing second image data having a plurality of density levels obtained by imaging the same subject as the first image data; and (c) a dynamic density level of the first image data stored in the first memory. Among the plurality of predetermined density levels Fd and Fc on the side corresponding to the smaller one of the exposure amounts within the range, the density level Fd that is closer to the outside of the dynamic range and the density level that is further closer to the outside of the dynamic range, A first extracting means for extracting coordinates of a pixel in a first memory having a density level of a density level Fc which is inside of the predetermined outward density level Fd; A second extracting means for extracting a pixel of the second memory having the same coordinates as the coordinates extracted by the extracting means; and (e) responding to the output of the second extracting means for each of the predetermined density levels Fd and Fc. , Average values F1d, F1 of the density levels of the pixels extracted from the second memory
and (f) responding to the output of the average value calculating means to set the density level of the second image data outside the dynamic range for adjusting the density level to the density level of the first image data. A density level difference α1, α1 = Fd−F1d with respect to the density level Fd closer thereto is obtained, and an exposure coefficient β1m which is a ratio of a gradient of the density level of the first image data to a gradient of the density level of the second image data. , Β1m = (Fd−Fc) / (F1d−F1c), and (g) in response to the output of the second extracting means, the predetermined density level Fd,
When the density level of a pixel in the second memory having the same coordinates as the coordinates of each pixel having a density level further outside the dynamic range than Fc is F1old, the density levels F1new, F1new of the image data after correction = Fd + (F1old + α1-Fd) × β
1m, and (h) the corrected image data obtained by the second arithmetic means is first image data in which each pixel has a plurality of density levels within the first density level range. Yes, the third storing the first image data
A memory; (i) an average value calculating means for calculating an average value Em of density levels of the first image data stored in the third memory; and (j) a number of pixels of the first image data whose number is less than a predetermined value. Is deleted, the density level whose number of pixels is equal to or more than the predetermined value is shifted toward the lower or higher density level by the deleted density level, and the deleted density level is deleted. Means for setting a density on the shifted side for a pixel having a level to generate second image data having a second density level range;
(K) responding to the output of the second image data generating means, when the number of pixels having the average value Em is less than the predetermined value, the average value Em is set within the second density level range of the second image data. The density level on the side shifted by the density level of the deleted pixel is set to the assigned value Zm1 of the average value Em, and when the number of pixels having the average value Em is equal to or greater than the predetermined value, the second image Assignment value setting means for setting the density level of the pixel of the second image data having the average value Em to the assignment value Zm2 within the second density level range of the data; (l) second image data creation means; In response to each output with the assigned value setting means, these assigned values Zm1, Zm2
Is set corresponding to a predetermined reference density level Cc in a third density level range CL to CH narrower than the first density level range, and the assignment values Zm1 and Zm2 in the second density level ranges ZL to ZH are set. The density level in each range between the values ZL and ZH at both ends of the density level in the second density level range is defined as a reference density level Cc in the third density level range, and the density level in both ends of the third density level range. Means for generating third image data by distributing and setting the density levels in respective ranges between values CL and CH. According to the present invention, a single image data with an extended dynamic range is obtained by using a plurality of, for example, three screens of image data obtained by imaging the same subject with mutually different exposure amounts. Since the dynamic range is compressed using the image data having the extended dynamic range, the relative relationship between the density levels in the wide dynamic range image data can be maintained even after the compression.

The present invention is characterized in that when there is no density level in which the number of pixels in the second image data is less than the predetermined value, the first image data is used as the second image data. According to the fifth, ninth, eleventh and fourteenth aspects of the present invention, even when the first image data is not a separated density distribution but a continuous density distribution, the dynamic range can be compressed by the same arithmetic processing.

Further, according to the present invention, the detection level of a plurality of pieces of detection data with the passage of time that is less than a predetermined value in the number of pieces of first detection data within the first detection level range is deleted. The detection level whose number of data is equal to or more than the predetermined value is shifted to one side of the detection level by the deleted detection level within the detection level range, and the data having the deleted detection level includes Is set to a second detection level having a second detection level range, and the average value Em of the detection levels of the first detection data is set to the assigned value Zm1 within the second detection level range. , Zm2, and set the values.
Zm2 is made to correspond to a predetermined reference detection level Cc within a third detection level range CL to CH narrower than the first detection level range of the first detection data, and the detection level allocation values Zm1 and Zm2 of the second detection data are set. The upper and lower ranges with respect to
This is a method for compressing the dynamic range of the detector, characterized in that the detection range is allocated to upper and lower ranges of the detection level related to the reference detection level Cc in the third detection level range.
Further, according to the present invention, (a) a plurality of pieces of detection data over time are stored in a memory for storing first detection data having a plurality of detection levels within a first detection level range, and (b) a memory for storing the first detection data. Means for calculating an average value Em of the detection levels of the first detection data; and (c) deleting, from the first detection data, detection levels whose number of data is less than a predetermined value, and deleting the deleted detection levels. By the amount, the detection level at which the data is equal to or more than the predetermined value is shifted to the lower or higher side of the detection level, and the pixel having the deleted detection level is provided with the detection level of the shifted side. And (d) responding to the output of the second detection data generating means, wherein the data having the average value Em is the predetermined value. If it is less than the value, the detection level on the side shifted by the detection level of the deleted data including the average value Em within the second detection level range of the second detection data is set to the assigned value Zm1 of the average value Em. And the average value E
When the data having m is equal to or greater than the predetermined value,
Within the second detection level range of the second detection data, the detection level of the second detection data having the average value Em is determined by the assignment value Z
In response to the outputs of the assigned value setting means for setting m2, (e) the second detection data creating means and the assigned value setting means, these assigned values Zm1 and Zm2 are narrower than the first detection level range. Assigned values Zm1 and Zm2 in second detection level ranges ZL to ZH are set corresponding to predetermined reference detection levels Cc in third detection level ranges CL to CH.
And values Z at both ends of the detection level in the second detection level range.
The detection level in each range between L and ZH is determined by detecting the range between the reference detection level Cc in the third detection level range and the values CL and CH at both ends of the detection level in the third detection level range. Means for creating the third detection data by distributing and setting the levels to generate the third detection data. According to the present invention, the dynamic range of the detector is compressed while maintaining the relative magnitude of the physical quantity to be measured that changes with the passage of time, such as the voltage or current of the electric signal, the temperature and the humidity, and the like. Will be able to

Further, the present invention is characterized in that when there is no detection level in which the number of data in the first detection data is less than the predetermined value, the first detection data is used as the second detection data. According to the present invention, when the detection level of the first detection data is not only the distant distribution but also the continuous distribution, the dynamic range can be compressed by the same processing.

[0014]

FIG. 1 is a block diagram showing the overall configuration of an embodiment of the present invention. The same subject is imaged with different exposure amounts by the camera 7 serving as an imaging means, and image data for a plurality of screens is obtained. This image data is color image data, and includes image data for each of red R, green G, and blue B. Image data of each color R, G, B from the camera 7 is digitized in an analog / digital conversion circuit 8 and converted into an 8-bit digital signal for each pixel of each color R, G, B. The digitized image data from the analog / digital conversion circuit 8 is supplied to the dynamic range expansion device 9 of the present invention, where the density level adjustment is performed for a plurality of digitized images having different exposure amounts. As well as
A screen composed of image data for each of the colors R, G, B is synthesized while correcting the exposure coefficient β2m, β1m described later.

Thus, the dynamic range is extended. The image data expanded by the dynamic range expansion device 9 can be given to an image processing device to perform color analysis and the like. The image data with the extended dynamic range is also given to the dynamic range compression device 10 for each color R, G, B synthesis, and the dynamic range extended image of the synthesized image is converted into, for example, a color cathode ray tube for displaying. Alternatively, the dynamic range is compressed so as to be provided to an output device such as a liquid crystal display.

The camera 7 has a condenser lens 12 for forming an image of a subject on an image sensor 11. Image sensor 11
Is composed of, for example, a CCD (charge storage element), and a total of three pixels for each color R, G, B constitute one combination, and a large number of such combinations are arranged in a matrix. Each pixel is provided with a filter 13 of colors R, G, and B. In order to set the exposure amount, an exposure amount setting unit 14 including an aperture and a shutter is provided, and the charge accumulation time of the image sensor 11 is adjusted. Image sensor 11
The image data of each pixel is supplied to the analog / digital conversion circuit 8 and digitized as described above.

FIG. 2 is a block diagram showing the configuration of the dynamic range extending device 9. The digitized image data from the analog / digital conversion circuit 8 is input for each of the colors R, G, and B, and stored in the memories M1 to M3 for each screen.
, Through an image switching circuit 15. Memory M1
Stores image data of one screen imaged by an intermediate exposure amount. Image data captured with a large exposure amount is stored in the memory M2. Image data captured with a small exposure amount is stored in the memory M3. In FIG. 2, a solid line represents image data for each of the colors R, G, and B, and a broken line represents a control signal other than the image data. The density level of the image data having an intermediate exposure amount stored in the memory M1 is defined by a predetermined density level Fa closer to the darker side of the dynamic range and another predetermined density level Fd closer to the lighter side of the dynamic range. The determination circuit 16 determines whether image data exists outside the range. Memory M out of specified range
When at least a part of the image data stored in No. 1 exists, the arithmetic units 17 and 18 determine the density level difference α
2, α1 are calculated by the arithmetic circuits 19 and 20, and the exposure coefficients β2m and β1m are calculated by the arithmetic circuits 21 and 22.

If all of the image data stored in the memory M1 is within the specified range, the image data is
Without being subjected to the calculation, the signal is directly passed to the image processing device through the replacement circuit 23 and is also supplied to the dynamic range compression device 10. When it is determined by the determination circuit 16 that the image data exists outside the specified range, each of the arithmetic units 17 and 18 performs the above-described calculation using the image data of the memory M1 and the image data of the memories M2 and M3. . The density level differences α2, α1 and the exposure coefficient β2m, β1 obtained by the arithmetic circuits 17 and 18 in this manner.
m is supplied to the density conversion circuits 24 and 25 and stored in the memory M1.
In the image data stored in the image data, the density level of a pixel having a density level outside the specified range is converted into corrected image data to extend the dynamic range.
The new corrected image data thus obtained is subjected to a correction operation in the color temperature correction circuits 26 and 27, and the replacement circuit 23
Given to. Therefore, image data within the specified range
The image data stored in the memory M1 is used as it is.
The image data is replaced with the image data given by the image data 6, 27, and the image data in which the dynamic range of one screen is extended is derived.

FIG. 3 is a diagram showing the relationship between the dynamic range of image data from the camera 7 and the density level of output image data when the exposure amount is changed. The horizontal axis in FIG.
The coordinates of the pixel in the horizontal direction of the subject are shown. This subject is configured for the sake of convenience such that the brightness or illuminance of the subject increases in the horizontal direction from left to right and becomes brighter, and the vertical axis in FIG. The scale is graduated from dark to light. The vertical axis in FIG. 3 represents the density level of the image data for each of the colors R, G, and B, and the reference symbol DR indicates the dynamic range of the camera 7. Image data captured with an intermediate exposure has the characteristic indicated by L4, and image data captured with a large exposure has the characteristic indicated by line L5, and an image captured with a smaller exposure. The data has the characteristics shown by line L6. Even if the brightness of the subject is the same, the density level of the image data differs depending on the value of the exposure amount,
It can also be seen that the gradient of the density level is different.

FIG. 4 is a diagram for explaining a subject. FIG. 4A shows a subject. As described above, the subject is configured such that the luminance or illuminance changes smoothly from dark to bright as going from left to right in the horizontal direction. In the vertical direction of the subject in FIG. 4A, the brightness of the subject is the same. FIG. 4B shows the brightness corresponding to the coordinates of the horizontal pixel of the subject shown in FIG. FIG.
The mutual interval Δb1 between the vertical lines in (1) corresponds to the brightness of the subject, and indicates that the brightness increases as Δb1 increases.

FIG. 5 is a diagram for explaining the operation when the camera 7 takes an image with an intermediate exposure amount, and FIG. 6 is a diagram for explaining the operation when the camera 7 takes an image with a large exposure amount. FIG. 7 is a diagram for explaining an operation when an image is captured by the camera 7 with a small exposure amount. Among these exposure conditions, image data of the subject obtained by large and small is shown in FIG. 5A, FIG. 6A and FIG. 7A, and the above-mentioned memories M1, M2, M3 , FIG. 5 (2), FIG. 6 (2) and FIG. 7 (2) are stored.

Referring to FIG. 5 (2), in the memory M1, the density levels obtained by the camera 7 at the coordinates A and D of the horizontal pixel coordinates of the subject shown on the horizontal axis are indicated by reference numerals Fa and Fd. And the concentration levels Fa to
Fd is set within the above-mentioned specified range. The density level of the dynamic range DR is, for example, in this embodiment,
A total of 256 gradations from 0 to 255 are set. If the density level is outside the specified range below Fa and beyond Fd, according to the present invention, the memory M2 uses the image data captured under the condition of the large exposure amount in FIG. Each correction is made using image data captured under the condition that the exposure amount shown in (2) is small. The interval Δb2 between the vertical lines in FIGS. 5 (1), 6 (1) and 7 (1) indicates the density level, and the greater the interval Δb2, the greater the density level, that is, the brighter.

The memory M1 stores density levels f1 to fr corresponding to pixel coordinates (x1, y1) to (xm, yn) for each of the colors R, G, and B. The same applies to M2 and M3.

[0024]

[Table 1]

FIG. 8 is a histogram obtained according to the present invention for image data of one color, for example, red R in the memory M1. The horizontal axis in FIG. 8 indicates the density level, and the vertical axis in FIG. 8 indicates the number of pixels, that is, the frequency.

FIG. 9 is a block diagram showing a simplified operation of the entire dynamic range extending device 9 shown in FIG. The dynamic range expansion device 9 performs a first operation 28 on the image data stored in the memory M1, performs a second operation 29 on the image data stored in the memory M2, and stores the image data in the memory M3. A third operation 30 is performed on the image data that has been set. Specific steps a1 to a9 of these first to third operations 28 to 30; a10 to a17; a20 to a
27 is shown in FIGS. 10 to 12, respectively.

In the exposure amount imaged by the camera 7,
The large and small image data are stored in the memory M as described above.
1, M2, and M3. First, the image data of the memory M1 is read out for each of the colors R, G, and B in step a1, and it is determined in step a2 whether the image data is within the specified range of the density level Fa or more and the density level Fd or less. The determination is performed by the determination circuit 16.
If it is determined that all the image data in the memory M1 exists within the specified range Fa to Fd and does not exist outside the specified range, the operation is terminated without performing the dynamic range expansion operation in step a3. .

If it is determined in step a4 that at least a part of the image data in the memory M1 exists outside the specified range, in the next step a5, the image data outside the specified range is classified into cases 1-3. . In case 1, at least a part of the image data of the memory M1 is
This is the case where Fd exceeds the upper limit Fd. Case 2
The case where at least a part of the image data in the memory M1 exists below the lower limit density level Fa. Case 3
Is the case where both the cases 1 and 2 described above are satisfied, that is, at least a part of the image data in the memory M1 exists beyond the density level Fd, and the density level F
This is the case when the value is less than a.

In step a6, the histogram described with reference to FIG. 8 is created for the image data in the memory M1. This histogram represents a frequency which is the number of pixels for each density level of the image data in the memory M1.

In step a7, with respect to the image data in the memory M1, in the case 1 or 3, the density level Fc which is equal to or lower than the upper limit density level Fd of the specified range and equal to or higher than the predetermined specified frequency Q is converted into the histogram. Search using That is, Fc is defined as the first density level having a density level equal to or larger than the predetermined number of pixels Q inside the dynamic range (left side in FIG. 8) from the density level Fd.
Is determined.

In step a8, with respect to the image data in the memory M1, a density level Fb equal to or higher than the lower limit density Fa of the specified range and equal to or higher than the specified frequency Q is searched. This density level Fb
Is within the dynamic range from the density level Fa (FIG. 8).
Is defined as the first density level having a density level equal to or larger than the predetermined number Q of pixels. Each density level F
The specified frequencies Q of c and Fb may be different from each other. Thus, the density levels Fc and Fb are determined based on the density levels Fa and Fd and the specified frequency Q.

In step a9, the coordinates of the pixels having the density levels Fa to Fd obtained in the respective cases 1 to 3 are extracted based on Table 1 described above.

In step a9, when the density level is F
Coordinates having a density level lower than a and higher than Fd are also extracted based on the image data in the memory M1.

In case 2 or 3 above, FIG.
In step 1, further arithmetic processing is performed based on the stored contents of the memory M2 in which the image data captured with a large exposure amount is stored. FIG. 11 shows the image memory M2
3 is a flowchart for explaining the processing operation of the image data stored in the storage device. In step a10, the memory M2 having the same coordinates as the pixels of the memory M1 having the density levels Fa and Fb obtained in step a9 among the image data stored in the memory M1 captured with the intermediate exposure amount. And the density level in the memory M2 is read out. Table 2 shows the results of this extraction. The coordinates of M2 in Table 2 are the coordinates of the pixels of the memory M2 having the density levels Fa and Fb in the memory M1.

[0035]

[Table 2]

In step a11, each density level Fa,
For each Fb, average values F2a and F2b of the density levels of the pixels extracted from the memory M2 are calculated and calculated based on Expressions 1 and 2.

[0037]

(Equation 1)

Fi is the density level of the pixels in the memory M2 having the same coordinates as the coordinates having the density level Fa of the image data in the memory M1, and the total number of pixels is q. Fj is the density level in the memory M2 having the same coordinates as the pixel having the density level Fb of the image data in the memory M1, and r is the total number of pixels.

In step a12, a density level difference α2 is calculated and obtained. α2 = Fa−F2a (3) In step a13, the exposure coefficient β2 is calculated and obtained. β2 = (Fb−Fa) / (F2b−F2a) (4) The above-described steps a1 to a13 are performed for each of the colors R, G, and B of the image data.

In step a14, the average value β2m of the average values β2R, β2G, and β2B for each of the colors R, G, and B obtained by Equation 4 is calculated. β2m = (β2R + β2G + β2B) / 3 (5) In step a15, the density level of the pixel of the image data in the memory M2 having the same coordinates as the pixel of the image data of the memory M1 having the density level less than the density level Fa Using F2old, and using the calculated density level difference α2 and the exposure coefficient β2m calculated above, Equation 6
Based on the density level F2new of the corrected image data
Is calculated. F2new = Fa + (F2old + α2-Fa) × β2m (6) In the arithmetic unit 17, the above-described steps a5 to a14 are executed, and the density conversion circuit 24 executes the operation of the above-described step a15. In step a16, the color temperature correction circuit 26 performs a color temperature correction calculation of the density level F2new of the corrected image data obtained by the equation (6).

For color temperature correction, memories provided in the color temperature correction circuit 26 include memories M2 and M3.
Corresponding to each color R, G, B as shown in Table 3.
Gains kR, kG, and kB are set for each.

[0042]

[Table 3]

The image data stored in the memory M2 is image data obtained by imaging the subject with darkness and therefore with a large exposure amount. In this case, since the color temperature is low, the gain kR of the red R is obtained. Is the gain kG of the other colors G and B,
It is set to a value smaller than kB, and for example, as shown in Table 3, kR = 0.8 and kG = kB = 1.0. The density level F2new for each color thus corrected in color temperature
R2, F2newG2, F2newB2 are obtained by Equations 7-9.

F2newR2 = F2newR × kR (7) F2newG2 = F2newG × kG (8) F2newB2 = F2newB × kB (9) In step a17, in the replacement circuit 23, the memory M
The image data within the specified range of the density levels Fa to Fd is directly derived without replacement, and
For the image data less than the above, the image data in the memory M2 is calculated as described above to derive the density level of the corrected image data for each color obtained by the above Expressions 7 to 9. In this way, the dynamic range is extended to the one with the larger exposure amount.

In case 1 or 3 described above, FIG.
The operation shown in FIG. 2 is performed. FIG. 12 is a flowchart for explaining the operation in case 1 or 3. Steps a20 to a27 in FIG. 12 correspond to steps a10 to a17 in FIG. 8 described above, and image data stored in the memory M3 of the subject captured with a small exposure amount is used.

In step a20, of the image data stored in the memory M1 picked up at the intermediate exposure amount, the density levels Fd,
The coordinates of the coordinates of the memory M3 having the same coordinates as the pixels of the memory M1 having Fc are extracted, and the density level in the memory M3 is read out in the same manner as in Table 2 described above.

In step a21, each density level Fc,
For each Fd, average values F1d and F1c of the density levels of the pixels extracted from the memory M3 are calculated and calculated based on Expressions 1 and 2.

[0048]

(Equation 2)

Fk is the density level of the pixels in the memory M3 having the same coordinates as the coordinates having the density level Fd of the image data in the memory M1, and the total number of pixels is u. Fh is the density level in the memory M3 having the same coordinates as the pixel having the density level Fc of the image data in the memory M1, and v is the total number of pixels. In step a22, a density level difference α1 is calculated and obtained. α1 = Fd−F1d (12) In step a23, the exposure coefficient β1 is calculated and obtained. β1 = (Fd−Fc) / (F1d−F1c) (13) Regarding the above-described steps a20 to a23,
This is performed for each of the colors R, G, and B of the image data. In step a24, each of the colors R, G, B obtained by Expression 13
The average value β1m of each of the average values β1R, β1G, and β1B is calculated and obtained. β1m = (β1R + β1G + β1B) / 3 (14) In step a25, the density level of the pixel of the image data in the memory M3 having the same coordinates as the pixel of the image data in the memory M1 having the density level lower than Fa. Using F1old, and using the calculated density level difference α1 and the exposure coefficient β1m calculated above,
5, the density level F1ne of the corrected image data
Calculate and obtain w. F1new = Fd + (F1old + α1−Fd) × β1m (15) In the arithmetic unit 17, the above-described steps a20 to a24 are executed, and the density conversion circuit 24 executes the above-described operation of step a25. In step a26, the color temperature correction circuit 26 performs a color temperature correction calculation of the density level F1new of the corrected image data obtained by equation (15). The image data stored in the memory M3 is image data obtained by imaging the subject brightly and therefore with a small amount of exposure. In this case, since the color temperature is high, the gain kB of blue B is changed to another The gain is set to a value smaller than the gains kR and kG of the colors R and G,
For example, as shown in Table 3 above, kB = 0.8 and kR = kG
= 1.0 is set. The density levels F1newR1 and F1new for each of the colors R, G, and B thus corrected in color temperature.
G1 and F1newB1 are obtained by Expressions 16 to 18. F1newR1 = F1newR × kR (16) F1newG1 = F1newG × kG (17) F1newB1 = F1newB × kB (18) In step a27, in the replacement circuit 23, the memory M
The image data within the specified range of the density levels Fa to Fd is directly derived without replacement, and the density level Fd
Is calculated, the image data in the memory M3 is calculated as described above to derive the density level of the corrected image data for each color obtained by Expressions 16 to 18 described above. In this way, the dynamic range is extended to the one with the larger exposure amount.

FIG. 13 shows the dynamic range extending device 9.
FIG. 4 is a diagram for explaining image data obtained by the above. The density level in the specified range of the density levels Fa to Fd in the area P1 of the image data obtained at the intermediate exposure amount and the gradient of the density level correspond to the density adjustment in the large and small exposure areas P2 and P3. A correction operation using the exposure coefficient is performed, and thus the characteristic L4 is added to the characteristic L4.
5, an extended dynamic range in which L6 is linearly continued. In the above-described embodiment, the dynamic range DR is such that the density level is determined to be 0 to 255, and for example, Fa = 10, Fb = 20, Fc = 230, Fd =
240.

In the dynamic range compressor 10 to which the corrected image data from the dynamic range extending device 9 is given, for example, when the dynamic range in a cathode ray tube or a liquid crystal display device is narrow as indicated by reference numeral 31 in FIG. In order to maintain a relative brightness relationship, and to obtain an easy-to-view image that has been converted to an appropriate level even if a wide dynamic range image is biased toward dark or bright sides as a whole. Can be. In another embodiment of the present invention, the range 31 of this dynamic range is, for example, 0 to
A density level range of 255 may be used.

FIG. 14 shows the dynamic range compression device 10.
15 is a block diagram showing a specific configuration of the dynamic range compression device 10. FIG. 15 is a flowchart showing the overall operation of the dynamic range compression device 10 in a simplified manner. Operation 3 in FIG.
Step 3 is achieved by steps b1 to b4 in FIG. The operation 34 includes steps b5 and b6 shown in FIG.
Achieved by Act 35 corresponds to step b in FIG.
7 to b9. Operation 36 is accomplished by steps b10 and b11 in FIG. In order to execute the operations shown in FIGS. 15 to 19, the dynamic range compression device 10 is provided with the wide dynamic range image data from the dynamic range expansion device 9 and stored in the memory 37. Is the color image data as described above, and is given to the histogram calculation circuit 38 and the average value calculation means 39. FIG.
4 indicate image data for each of the colors R, G, and B, and a broken line indicates a control signal. Histogram operation circuit 3
8, each color R,
The histograms shown in FIGS. 20 (1) to 20 (3) for each of G and B are created. In this histogram, the horizontal axis represents, for example, density levels from 0 to 1000 on an equally spaced scale, and the vertical axis represents the frequency, that is, the number of pixels.
Such histogram generation is performed by the step b1 in FIG.
From step b2.

In the next step b3, for each density level, the number of pixels of all the colors R, G, B is less than a predetermined value, for example, zero in this embodiment, Search for a distribution image. In FIG. 20, reference numerals 41 and 42 indicate a common density level in which the number of pixels is zero for all the colors R, G and B. Such a density level search is achieved by the density level search circuit 43 in FIG.

In step b4, the average value Em of the density levels of all the colors R, G, and B of the image data having a wide dynamic range from the memory 37 is calculated.

[0055]

(Equation 3)

Where fi is one density level,
ki is the number of pixels having the density level fi, i
Are all density levels of an image with a wide dynamic range, and K is the total number of pixels of the image data in the memory 37.

If it is determined in step b3 that there is no density level whose number of pixels is less than the predetermined value, the flow advances from step b5 to step b6 in FIG. Is used as it is, and the process proceeds to step b10 in FIG. 19 described later. In this case, it can be seen that the image data is a smooth image, that is, an image having a continuous density level distribution having many intermediate density levels.

If it is determined in step b3 that there is a density level in which the number of pixels is less than the predetermined value, the process proceeds from step b7 in FIG. 18 to step b8, and the density conversion circuit 44 in FIG. A density level whose number is less than a predetermined value is deleted, and a density level whose number of pixels is equal to or more than the predetermined value is shifted to a lower density level by the deleted density levels 41 and 42, and The pixel having the deleted density level is set to the lower density of the density level which is the shifted side, and the second color common to each of the colors R, G and B is set as shown in FIG. Image data having a density level range is created.

FIG. 21 shows a histogram of the second image data having the second density level ranges ZL to ZH after deleting the density levels in which the number of pixels is less than the predetermined value, as shown in FIGS. 21 (3) shows a histogram for each of the colors R, G, B. Deleted density levels 41, 4
The position of 2 is indicated by the same reference numerals 41 and 42 in FIG. Thus, in order from the low concentration level, that is, FIG.
21. In order from left to right in FIG. 21, the density levels of frequencies R, G, and B that are less than the frequency, which is a predetermined number of pixels, are deleted.
The conversion table is created by replacing the density level with the most recently registered density level and allocating density levels equal to or larger than the predetermined number of pixels in the order of density levels. That is, the wide dynamic range image data stored in the memory 37 is referred to as first image data, and the number of pixels for each density level is as shown in Table 4. It is assumed to be stored in For example, in Table 4, it is assumed that the number of pixels is zero at density levels 121 to 129.

[0060]

[Table 4]

In the case of Table 4, in step b8, the density conversion circuit 44 creates the conversion table of Table 5.

[0062]

[Table 5]

The density levels 121 to 129 in Table 4 where the number of pixels is zero are deleted. Also, the density level 1 in Table 4
Even between 31 and 1000, for example, the density level at which the number of pixels is zero is deleted. Thus, as shown in Table 5, the second density level range is, for example, 0 to 340. At step b9, for example, the allocation values Zm1 and Zm2 of the average value Em in the conversion table obtained in Table 5 are searched and set.

When the number of pixels having the average value Em is less than the predetermined value, the total density average conversion circuit 45 of FIG. The density level on the side shifted by the density level of the deleted pixel including the average value Em is set to the assigned value Zm1 of the average value Em.

FIG. 22 is a diagram for explaining the operation of the total density average conversion circuit 45. As shown in FIG. 22A, the density level of the first image data in the memory 37 is 0 to
1000 and the density level 121 as described above.
In the range 41 to 129, the number of pixels is zero, and similarly, in the range 42 of the density levels 250 to 400, the number of pixels is zero. The average value Em is, for example, 350
In this case, the average value Em exists in the range 42 of the density level to be deleted.

FIG. 22 (2) shows that the entire density average conversion circuit 45 uses the deleted density level ranges 41, 4.
The second obtained by shifting the density level to the lower one by 2
3 is a histogram in a density level range of 0 to 340. FIG. 22 (1) corresponds to Table 4, and FIG. 22 (2) corresponds to Table 5. By deleting the density level range 41 in FIG. 22A, the density level 130 in FIG. 22A becomes a new density level 121 in FIG. The range 42 in FIG.
Are eliminated, the density level 249 in FIG. 22A becomes
The density level becomes 240 as shown in FIG.
Is the density level 401 of FIG. 22 (2).
Converted to 1. In this case, the average value Em (for example, 35
0) is set to the density level 240 in FIG. 22B on the side shifted by the density level of the range 42 of the deleted pixel. That is, the assignment value Zm1 is set to 240.

Assuming that the number of pixels having the average value Em is equal to or larger than the predetermined value and the average value Em is 200, for example, as shown in FIG. Concentration level range 41 to be deleted,
42 does not exist. In this case, the density level of the pixel of the second image data having the average value Em shown in FIG. 23B is set to the assigned value Zm2, and this calculation is performed in the reference density conversion circuit 46. For example, FIG.
In (1), the histogram of the first image data has the density level ranges 41 and 4 as in FIG.
2 is the range to be deleted, and the number of pixels is
Assuming that it is different from (1) and that the average value Em = 200, in the second image data of FIG. 23 (2), the average value Em is shifted by the range 41 toward the lower density level. The assigned value Zm2 = 191 is set.

The reference density conversion circuit 46 proceeds from step b6 in FIG. 17 and step b9 in FIG. 18 to step b10 in FIG. The histogram of the second image data in FIG. 23 (2) is also shown in FIG. 24 (1). Based on the second image data shown in FIG. 24 (1), as shown in FIG. 24 (2), the density level C in the third density level range CL to CH where the dynamic range is compressed is calculated. There is a need.

In step b10, first, FIG.
A reference density level Cc in the third density level range CL to CH shown in (2) is set, and the reference density level Cc is set.
With respect to c, the first compression coefficient kL in the lower density level range CL to Cc is calculated and obtained. kL = (Cc-CL) / (Zm-ZL) (20) where z <Zm Here, the assigned values Zm1 and Zm2 are collectively represented by a reference sign Zm. Regarding the reference density level Cc in the third density level range, in the upper density level range Cc to CH,
The second compression coefficient kH is calculated and obtained. kH = (CH−Cc) / (ZH−Zm) (21) where z ≧ Zm The reference density level Cc in the third density level range CL to CH is a predetermined value, for example, this range CL to CH Is set to a value at or near the center thereof. For example, in this embodiment, Cc = 128 is selected.

In step b11, the compressed density level C is calculated and obtained according to the correspondence between the converted density levels Z of the colors R, G, and B and the assigned values Zm1, Zm2 and less. If Z <Zm, C = Cc-kL (Zm-Z) (22) If Z ≧ Zm, C = Cc + kH (Z-Zm) (23) Thus, in step b12, a series of operations is ended. Get.

[0071]

[Table 6]

As shown in FIG. 25, a histogram with a reduced dynamic range is obtained. FIGS. 25 (1) to 25 (3) show the third colors R, G, and B for each color.
It is a histogram of image data. Reference density level Cc
Is set to one value common to each of the colors R, G, and B.

The present invention can be implemented not only in connection with image data but also with other physical quantities to be measured, for example, values that change with time, such as voltage or current of an electric signal, temperature or humidity. A similar implementation can also be used to reduce the dynamic range of the detector to be measured.

For the configuration for extending the dynamic range in the present invention, the following embodiments are possible. The embodiment is a method of extending the dynamic range of an image that obtains a single image data with an extended dynamic range from image data of a plurality of screens obtained by imaging the same subject with different exposure amounts, Of the image data having different exposure amounts, the density levels at both ends or one end of the dynamic range of the reference image data in the same subject portion are adjusted to the density levels α1 and α2 of the image data having different exposure amounts. At the same time, the change amount (F2b-F) of the density level of image data having a different exposure amount with respect to the change amount (Fb-Fa, Fd-Fc) of the reference image data in the same subject portion.
2a, the exposure coefficient β2m, which is the ratio of F1d-F1c),
This is a method of extending the dynamic range of an image, characterized by correcting image data having a different exposure amount from reference image data using β1m.

In this embodiment, the above-described processing is repeated as image data based on the obtained image data of which the dynamic range of a single sheet has been extended, so that the image data can be extended to a desired dynamic range.

In the embodiment, the first image data captured at a certain exposure amount and the second image data captured at an exposure amount larger than the first image data are obtained. A density level Fa represented by image data at one end of the dynamic range, and a density level F2 represented by image data at the other end of the dynamic range of the second image data.
is obtained, and a change amount (Fb-F) of the density level with respect to the subject portion represented by the image data at the one end of the dynamic range in the first image data is obtained.
a), and an exposure coefficient β2m, which is a ratio of a density level change amount (F2b−F2a) of the second image data to the subject portion represented by the image data at the other end of the dynamic range, is obtained. Correcting the density level F2old of the second image data based on the density level difference α2 and the exposure coefficient β2m to obtain corrected image data F2new continuous to the one end of the dynamic range of the first image data. This is a method for extending the dynamic range of an image characterized by the following.

In the embodiment, the density level F2new of the corrected image data is expressed as follows: F2new = Fa + (F2old + α2-Fa) × β
2 m 2.

In this embodiment, first image data captured with a certain exposure amount and second image data captured with a smaller exposure amount than the first image data are obtained. A density level Fd represented by image data at one end of the dynamic range, and a density level F1 represented by image data at the other end of the dynamic range among the second image data.
and a density level difference α1 with respect to the subject portion represented by the one end image data of the dynamic range in the first image data (Fc−F).
d), and among the second image data, an exposure coefficient β1m, which is a ratio of a density level change amount (F1d-F1c) to the subject portion represented by the image data at the other end of the dynamic range, is obtained. Correcting the density level F1old of the second image data based on the density level difference α1 and the exposure coefficient β1m to obtain corrected image data F1new continuous to the one end of the dynamic range of the first image data. This is a method for extending the dynamic range of an image characterized by the following.

In the embodiment, the density level F1new of the corrected image data is expressed as follows: F1new = Fd + (F1old + α1-Fd) × β
1 m 2.

According to the embodiment, a plurality of images (for example, three screens) obtained by imaging the same subject with different exposure amounts are used to synthesize a plurality of images having different exposure amounts. Thus, it is possible to obtain an image having an extended dynamic range in which both the density level and the exposure amount ratio are corrected. Using the image data for the plurality of screens, the density level is adjusted, the ratio of the amount of exposure is determined from the image data for the plurality of screens, and the density conversion is performed using the ratios. In obtaining the density level α2, for example, one side of the dynamic range of the first image data with an intermediate exposure amount (that is, the brightness or illuminance of the subject is low, that is, the region which is dark and is imaged with a large exposure amount) Density level Fa represented by the image data of
And calculating the difference between the second image data captured with the large exposure amount and the density level F2a represented by the image data on the other side of the dynamic range (that is, the side of the area obtained with the intermediate exposure amount). Obtained by The exposure coefficient β2m, which is the ratio of the exposure, is used to determine the amount of change (Fb-Fa) in the density level of the two subject portions based on, for example, image data on the one side of the dynamic range in the first image data. The ratio between the dynamic range of the second image data for the same two subject portions and the change amount (F2b-F2a) of the density level based on the image data on the other side is obtained. In this way, the levels of the first and second image data are adjusted, and the image data on the one side of the dynamic range in the first image data is replaced by calculating and correcting the density level F2old of the second image data. By this calculation, the dynamic range can be further extended near the one side of the dynamic range of the first image data.

In the above description and the following description, description will be made mainly in relation to image data having an intermediate exposure amount and image data having a large exposure amount. The same applies to image data with a smaller exposure amount. In other words, when obtaining the density level α1, for example, of the first image data at the intermediate exposure amount, the one of the dynamic range is shifted toward one side (that is, the brightness or illuminance of the subject is high, that is, the region side where the image is captured with a small exposure amount is bright. ) Represents the density level Fd represented by the image data.
And calculating the difference between the second image data captured with the small exposure amount and the density level F1d represented by the image data on the other side of the dynamic range (that is, the area obtained with the intermediate exposure amount). Obtained by The exposure coefficient β1m, which is the ratio of the exposures, is used to determine the amount of change (Fd−Fc) in the density level of the two object portions based on, for example, image data on the one side of the dynamic range in the first image data. The ratio between the dynamic range of the second image data for the same two subject portions and the change amount (F1d-F1c) of the density level based on the image data on the other side is calculated. In this way, the levels of the first and second image data are adjusted, and the one-side image data of the dynamic range in the first image data is calculated and replaced by correcting the density level F1old of the second image data. By this calculation, the dynamic range can be further extended near the one side of the dynamic range of the first image data.

In the embodiment, the density levels F2new and F1new of the corrected image data are expressed as follows: F2new = Fa + (F2old + α2-Fa) × β
2m, and F1new = Fd + (F1old + α1-Fd) × β
1 m 2.

According to the embodiment, the density levels F2old and F1old of the second image data are used, and the density level difference α2 and the exposure coefficients β2m and β1m are used.
It is possible to obtain corrected density levels F2new and F1new of the second image data to be combined with the first image data.

The image data may be monochrome, for example, but may also be color image data for each of red R, green G and blue B.

In the embodiment, the image data is represented by red R,
Color image data for each color of green G and blue B;
Exposure coefficient β2m, β1m are exposure coefficient β2R, β2G, β2B; β1 obtained from image data of each color.
R, β1G, and β1B are average values.

According to the embodiment, when the first and second image data are color image data, the exposure coefficient β
2m and β1m are individual exposure coefficient β2R, β2G, β2 obtained from image data for each color R, G, B.
B; average value of β1R, β1G, β1B, whereas density level differences α2, α1 are independent values α2R, α2G, α2B; α1R, α for color R, G, B, respectively.
Using 1G and α1B, density levels F2new and F1new of the corrected image data for each color are calculated and obtained.
As a result, even when the sensitivity difference and the amplification factor of the image sensor for each color R, G, and B vary, the color reproducibility is excellent, and the black level difference between the colors R, G, and B is reduced. The influence can be reduced so that a false contour does not occur.

In this embodiment, the corrected image data F2new and F1new for each color are used as the image data F2new used to obtain the corrected image data F2new.
When 2old is the image data with the larger exposure amount, the gain of red is lowered compared to the gain of blue to correct the color temperature, and the image data F1old used to obtain the corrected image data F1new is When the image data has a smaller exposure amount, the gain of the blue color is reduced compared to the gain of the red color and the color temperature is corrected.

According to the embodiment, the color temperature correction of the density levels F2new and F1new of the corrected image data is performed. The subject is dark, so the aperture is large,
In the case of using the second image data in which the charge accumulation time is adjusted to be longer and the exposure amount is increased, since the color temperature of the second image data is low, the red gain is lowered and corrected compared to the blue gain. You. Conversely, for a bright subject, the exposure is adjusted to be small. In this case, the color temperature of the second image data is high, so that the blue gain is lower than the red gain.

Further, according to the embodiment, a first memory for storing first image data having a plurality of density levels imaged at a certain exposure amount, and an exposure amount larger than the exposure amount of the first image data, A second memory for storing second image data having a plurality of density levels obtained by capturing an image of the same subject as the first image data, and exposing light within a dynamic range of density levels of the first image data stored in the first memory; Among the plurality of predetermined density levels Fa and Fb on the side corresponding to the larger amount, the density level Fa outside the dynamic range, the density level further outside the dynamic range, and the predetermined outside level. First extracting means for extracting the coordinates of a pixel in a first memory having a density level of a density level Fb which is inside the density level Fa closer to the first level; A second extraction means for extracting a pixel of the second memory having the extracted coordinates the same coordinate by,
Means for calculating average values F2a and F2b of the density levels of the pixels extracted from the second memory for each of the predetermined density levels Fa and Fb in response to the output of the second extraction means; In response to the output of the arithmetic means, the density level difference α relating to the density level Fa outside the dynamic range for adjusting the density level of the second image data to the density level of the first image data.
2, α2 = Fa−F2a, and further, exposure coefficient β2m, β2m = (Fb−Fa) / (F2b), which is the ratio of the gradient of the density level of the first image data to the gradient of the density level of the second image data. -F2a), and in response to the output of the second extracting means, the predetermined density levels Fa, Fb in the first memory.
When the density level of a pixel in the second memory having the same coordinates as the coordinates of each pixel having a density level further outside the dynamic range is F2old, the density levels F2new, F2new = Fa + of the corrected image data. (F2old + α2-Fa) × β
And a second calculating means for calculating 2m 2.

Further, according to the embodiment, a first memory for storing first image data having a plurality of density levels imaged at a certain exposure amount, and an exposure amount smaller than the exposure amount of the first image data, A second memory for storing second image data having a plurality of density levels obtained by capturing an image of the same subject as the first image data, and exposing light within a dynamic range of density levels of the first image data stored in the first memory; Among the plurality of predetermined density levels Fd and Fc on the side corresponding to the smaller amount, the density level Fd located outside the dynamic range and the density level located further outside the dynamic range, and the predetermined outside. First extracting means for extracting the coordinates of a pixel in the first memory having a density level of the density level Fc which is inside the density level Fd closer to the first extraction means; A second extraction means for extracting a pixel of the second memory having the extracted coordinates the same coordinate by,
Means for calculating average values F1d and F1c of the density levels of the pixels extracted from the second memory for each of the predetermined density levels Fd and Fc in response to the output of the second extraction means; In response to the output of the calculating means, the density level of the second image data is adjusted to the density level of the first image data.
1, α1 = Fd−F1d, and the exposure coefficient β1m, β1m = (Fd−Fc) / (F1d), which is the ratio of the gradient of the density level of the first image data to the gradient of the density level of the second image data. -F1c) in response to the output of the first calculating means and the output of the second extracting means.
When the density level of a pixel in the second memory having the same coordinates as the coordinates of each pixel having a density level further outside the dynamic range is F1old, the density levels F1new, F1new = Fd + of the image data after correction. (F1old + α1-Fd) × β
And a second calculating means for calculating 1m 2.

According to the embodiment, for example, the first extracting means corresponds to any one of the exposure amounts in the first image data stored in the first memory obtained by imaging with the intermediate exposure amount. Side, for example, a plurality of, for example, two, coordinates of pixels of a first memory having predetermined density levels Fa and Fb on the side of the second image data picked up with a large exposure amount. A pixel in the second memory having the same coordinates as the extracted coordinates of the pixel in the memory, in other words,
The density level extracted by the extracting means, the density level of the pixel of the second memory having the coordinates obtained by the second extracting means is obtained, and the average values F2a and F2b are obtained.
The density level difference α2 is obtained based on the average value F2a corresponding to the density level Fa located outside the dynamic range. Also, the average values F corresponding to the density level Fa closer to the outside and the density level Fb located further inward are shown.
Exposure coefficient β2m is obtained using 2a and F2b. By using the average values F2a and F2b in this manner, the accuracy of the density level difference α2 and the exposure coefficient β2m can be improved, and errors due to differences in exposure can be reduced.

In the embodiment, the first and second image data are color image data for each of red R, green G, and blue B, and the first arithmetic means performs the first and second image data for each color. Based on the two image data, the exposure coefficient β2R,
β2G, β2B; β1R, β1G, β1B are obtained, and the exposure coefficient β2R, β2G, β2B for each color is determined.
It is characterized in that average values β2m and β1m of β1G and β1B are obtained and given to the second calculating means.

According to the embodiment, the first and second image data are color image data for each color. In this case, the density levels F2new and F1 of the corrected image data are used.
Exposure coefficient β2 used in obtaining new
m and β1m are exposure amount coefficients β2R and β2 obtained for each color.
G, β2B; average values of β1R, β1G, and β1B.
As a result, even when the sensor sensitivities and amplification factors of the respective colors R, G, and B vary, the color reproducibility is excellent, the influence of the black level difference between the colors R, G, and B is reduced, and the pseudo contour is reduced. It can be prevented from occurring. The density level difference α2 is used to determine the density levels F2new and F1new of the corrected image data for each of the colors R, G, and B.

Also, in the embodiment, the first extracting means includes:
(A) means for creating a histogram representing the number of pixels for each density level of the first image data stored in the first memory; and (b) responding to the output of the histogram creating means, Fb; Fd, Fc, one of the predetermined density levels Fa, Fd is set outside the dynamic range of the other predetermined density level Fb, Fc, and the other predetermined density level Fb, Fc is set. Is defined as a first density level having a density level equal to or more than a predetermined number of pixels within the dynamic range from the one of the predetermined density levels Fa and Fd.

According to the embodiment, a specified range which is closer to the inside of the dynamic range than one of the density levels Fa and Fd is set by one of the density levels Fa and Fd in the first image data. Within the range, the density level closest to the one density level Fa or Fd having a density level equal to or more than a predetermined number of pixels is determined as the other predetermined density level Fb or Fc. This makes it possible to calculate and calculate the exposure coefficient β2m and β1m with high accuracy.

In the embodiment, (a) the first and the second
The image data is color image data for each of red R, green G and blue B. (b) When the exposure amount of the second image data is larger than the exposure amount of the first image data, the gain of the red color Is set lower than the blue gain, and when the exposure amount of the second image data is small, the color temperature correction memory sets the blue gain lower than the red gain, and 2. A color temperature correction calculating means for multiplying the gain stored in the temperature correction memory by the corrected image data F2new and F1new.
Device for extending the dynamic range of an image according to one of the two.

According to the embodiment, when the first and second image data are color image data, the color temperature is corrected as described above, and the exposure in the bright and dark portions of one subject is performed. The difference in color temperature caused by the difference in amount is corrected.

In the embodiment, a first memory for storing first image data having a plurality of density levels imaged at a certain exposure amount, and a first memory having an exposure amount larger than the exposure amount of the first image data. A second memory for storing second image data having a plurality of density levels obtained by imaging the same subject as the one image data; and storing the same subject as the first image data with an exposure smaller than the exposure of the first image data. A third memory for storing third image data having a plurality of imaged density levels and a density level of the first image data within a dynamic range of density levels of the first image data stored in the first memory. It is outside the specified range between a predetermined first density level Fa on one side corresponding to the larger exposure amount and a second density level Fd on the other side corresponding to the smaller exposure amount. , A range judging unit for judging whether it is outside the first density level Fa or outside the second density level Fd, and responding to the output of the range judging unit, The level is the first density level Fa
When the third density level Fb is outside the first density level Fa, the third density level Fb is determined on the one side of the dynamic range to be inside the first density level Fa. First with level
First extracting means for extracting the coordinates of the pixels of the first image data stored in the memory, and second extracting means for extracting the pixels of the second memory having the same coordinates as the coordinates extracted by the first extracting means; , In response to the output of the second extracting means, for each of the first and third density levels Fa and Fb, the average value F2a, of the density levels of the pixels extracted from the second memory.
A first average value calculating means for calculating F2b, and a density level of the second image data in response to an output of the first average value calculating means,
A density level difference α2, α2 = Fa−F2a relating to the first density level Fa located outside the dynamic range for adjusting the density level to the density level of the first image data is obtained. The first calculation means for calculating the exposure coefficient β2m, β2m = (Fb−Fa) / (F2b−F2a), which is the ratio of the gradient of the density level of the first image data to the gradient of the level, and the output of the second extraction means. In response, when the density level of a pixel in the second memory having the same coordinates as the coordinates of each pixel having a lower density than the first density level Fa in the first memory is F2old, the density levels F2new, F2new of the image data after correction are given. Fa + (F2old + α2-Fa) × β
2m, and the density level of the first image data is changed to the second density level Fd in response to the output of the range determining means.
When it is more outward than the second density level, a fourth density level Fc that is inside the dynamic range from the second density level Fd is determined on the other side of the dynamic range, and the density level is equal to or higher than the second density level Fd and the fourth density level Fc. Third extracting means for extracting the coordinates of the pixels stored in the first memory having: and fourth extracting means for extracting the pixels of the third memory having the same coordinates as the coordinates extracted by the third extracting means; Second average value calculating means for responding to the output of the fourth extracting means for obtaining, for each of the second and fourth density levels Fd, Fc, average values F1d, F1c of the density levels of the pixels extracted from the third memory; Responsive to the output of the second average value calculating means, the density level of the third image data is shifted toward the outside of the dynamic range for adjusting the density level to the density level of the first image data. A density level difference α1, α1 = Fd−F1d with respect to the second density level Fd is obtained, and an exposure coefficient β1m, which is a ratio of a gradient of the density level of the first image data to a gradient of the density level of the third image data, β3m = (Fd−Fc) / (F1d−F1c), and the coordinates of each pixel having a density level exceeding the second density level Fd of the first memory in response to the output of the fourth extraction means. When the density level of the pixel in the third memory having the same coordinates as F1old is F1old, the density level F1new of the corrected image data, F1new = Fd + (F1old + α1-Fd) × β
And a fourth calculating means for calculating 1m 2.

According to the embodiment, the first image data is
The second image data is image data obtained by imaging with a large exposure, and the third image data is image data obtained by imaging with a small exposure. With this, the dynamic range on both sides of the first image data can be extended to both upper and lower sides by adjusting the density level and matching the exposure amount ratio.

In the embodiment, the same physical quantity to be measured is measured with mutually different characteristics, a plurality of detection data are obtained at the same time for each characteristic, and both ends or one end of the dynamic range of the reference detection data are obtained. The detection level is adjusted to the detection level of the detection data measured with different characteristics and the detection level α1, α2, and the change of the physical quantity to be measured between the reference data for the same measurement object and the detection data measured with different characteristics Amount (Fb-Fa, Fd-Fc)
(F2b-F2a, F1d)
-F1c) is a method of extending the dynamic range of a detector, wherein the detection data measured with different characteristics from the reference detection data is corrected using the coefficients β2m and β1m.

In the embodiment, a first memory for storing first detection data obtained by detecting a physical quantity to be measured with a predetermined characteristic, and a physical quantity to be measured within a low value range are first detected by a first characteristic. A second memory for storing second detection data obtained by detecting the physical quantity to be measured at the same time as the first detection data by a second characteristic obtained at a detection level higher than the data detection level; Of the plurality of predetermined detection levels Fa and Fb on the side adjacent to the detection level within the dynamic range of the detection level of the first detection data, and the detection level Fa outside the dynamic range and further outside. A first memo having a detection level closer to the outside and a detection level Fb closer to the inside than the predetermined outside detection level Fa. First extracting means for extracting a time on the time axis of the second extracting means, second extracting means for extracting detection data of the second memory at the same time as the time extracted by the first extracting means, and an output of the second extracting means. In response, the detection level difference α2, α2 = Fa−F2a relating to the detection level Fa outside the dynamic range for adjusting the detection level of the second detection data to the detection level of the first detection data. First calculating means for obtaining coefficients β2m, β2m = (Fb−Fa) / (F2b−F2a), which are the ratios of the slope of the detection level of the first detection data to the slope of the detection level of the second detection data. , The predetermined detection levels Fa and Fb of the first memory in response to the output of the second extraction means.
When the detection level of the second memory at each same time having a detection level further outside than that is F2old,
Detection level F2new of the corrected detection data, F2new = Fa + (F2old + α2-Fa) × β2
and a second calculating means for calculating m. 2 is a device for extending the dynamic range of the detector.

Further, in the present embodiment, the first memory for storing the first detection data obtained by detecting the physical quantity to be measured with a predetermined characteristic, and the first detection of the physical quantity to be measured within a high value range by the first characteristic. A second memory for storing second detection data obtained by detecting the physical quantity to be measured at the same time as the first detection data by a second characteristic obtained at a detection level higher than the detection level of the data; Within the dynamic range of the detection level of the first detection data, among the plurality of predetermined detection levels Fd and Fc on the side adjacent to the detection level, the detection level Fd that is closer to the outside of the dynamic range and the detection level Fd that is further outside. A first memo having a detection level closer to the outside and a detection level Fc closer to the inside than the predetermined outside detection level Fd. First extracting means for extracting the time on the time axis of the second extracting means; second extracting means for extracting the detection data of the second memory at the same time as the time extracted by the first extracting means; In response, the detection level difference α1 and α1 = Fd−F1d relating to the detection level Fd located outside of the dynamic range for adjusting the detection level of the second detection data to the detection level of the first detection data. First calculating means for calculating the coefficients β1m, β1m = (Fd−Fc) / (F1d−F1c), which are the ratios of the slope of the detection level of the first detection data to the slope of the detection level of the second detection data. , The predetermined detection levels Fd, Fc of the first memory in response to the output of the second extraction means.
When the detection level of the second memory at each same time having a detection level further outside than the above is F1old,
Detection level F1new of the corrected detection data, F1new = Fd + (F1old + α1-Fd) × β1
and a second calculating means for calculating m. 2 is a device for extending the dynamic range of the detector.

According to the embodiment, it is possible to extend the dynamic range of a detector for measuring a physical quantity to be measured that changes with time, such as the amplitude of an electric signal, such as voltage or current, temperature, and humidity. Embodiments can thus be implemented not only in connection with images, but also in a wide range in connection with other detectors for measuring the physical quantity to be measured.

[0104]

According to the first and second aspects of the present invention, the relationship between the density levels in the first image data having a wide dynamic range is determined by the third density level range which is a narrow dynamic range after dynamic range compression. Can be held even in image data having
Even when there is a large difference between a bright part and a dark part of the image data, a change in the density level in each part is expressed, and a visually natural image having a gradation change in the density level can be obtained. In this way, it is possible to prevent the loss of white and black when compressing a wide dynamic range image on the mountain of the density level distribution that exists at a distance, and to obtain a smooth image emphasizing the gradation change. Can be visualized to make humans look even more natural. Further, according to the present invention, it is possible to obtain a compressed image having an easily viewable narrow dynamic range converted to an appropriate level even if the image has a wide dynamic range image in which the density level is biased in dark or bright areas as a whole. it can.

According to the third aspect of the present invention, in distributing the first image data to the third density level range, the first and second compression coefficients kL and kH are calculated and calculated to obtain the second density level range. , The lower and upper density level ranges related to the assigned values Zm1 and Zm2 in the third density level range CL
It can be linearly distributed with a linear function below and above with respect to the reference concentration level Cc in ~ CH.

According to the fourth aspect of the present invention, when the image data is color image data, the average value E of all the colors R, G, and B is used.
m is calculated, and a dynamic range compression operation is performed for each of the colors R, G, and B, whereby the colors R, G, and B are calculated.
It is possible to prevent the collapse of each gradation.

According to the fifth, ninth, eleventh, and fourteenth aspects of the present invention, the dynamic range can be compressed not only for the separated density distribution but also for the first image data having the continuous density distribution by the same arithmetic processing.

According to the sixth aspect of the present invention, similar to the first and second aspects, the dynamic range is compressed while maintaining the brightness of the first image data, that is, the relative relationship of the density level. be able to.

According to the seventh aspect of the present invention, similarly to the third aspect, the reference density level Cc in the third density level range is provided.
It is automatically possible to make a distribution by a linear function in the range of the lower and upper density levels for.

According to the eighth aspect of the present invention, it is possible to perform dynamic range compression in which color collapse of each of the colors R, G, and B is prevented, as in the above-described fourth aspect.

According to the tenth, twelfth, and thirteenth aspects of the present invention, the dynamic range is temporarily extended based on image data of a plurality of screens obtained by imaging the same subject with different exposure amounts, and thereafter, Because the dynamic range is compressed while maintaining the relative relationship of density levels, emphasis is placed on the gradation change that prevents white and black loss in the image even with a cathode ray tube or liquid crystal display device with a narrow dynamic range. The resulting smooth image can be output by display or the like.

According to the fifteenth and seventeenth aspects of the present invention, the dynamic range of the detector for measuring the physical quantity to be measured with the passage of time is detected while maintaining the relative size of the detected data. The dynamic range of the vessel can be compressed.

According to the sixteenth and eighteenth aspects of the present invention, the first
Even if the detection level of the detection data is not only the separated distribution but also the continuous distribution, the dynamic range of the detector can be compressed by the same arithmetic processing.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an overall configuration of an embodiment of the present invention.

FIG. 2 is a block diagram showing a configuration of a dynamic range extending device 9.

FIG. 3 is a diagram showing the relationship between the brightness of a subject and the density level of output image data when the exposure amount of image data by the camera 7 is changed.

FIG. 4 is a diagram for describing a subject used for describing one embodiment of the present invention.

FIG. 5 is a diagram for explaining image data of the subject in FIG.

FIG. 6 is a view for explaining image data of a subject having a large exposure amount when the subject in FIG.

FIG. 7 is a diagram for explaining image data of a small exposure amount when the subject in FIG. 4A is imaged by the camera 7;

FIG. 8 is a histogram obtained according to the present invention.

FIG. 9 shows a dynamic range extending device 9 shown in FIG.
3 is a simplified block diagram showing the overall operation of FIG.

FIG. 10 is a flowchart specifically showing an operation 28 in FIG. 9;

FIG. 11 is a flowchart specifically showing an operation 29 of FIG. 9;

FIG. 12 is a flowchart specifically showing an operation 30 of FIG. 9;

FIG. 13 is a diagram for explaining image data obtained by the dynamic range extending device 9.

FIG. 14 is a block diagram showing a specific configuration of the dynamic range compression device 10.

FIG. 15 is a flowchart showing the overall operation of the dynamic range compression device 10 in a simplified manner.

FIG. 16 is a diagram showing the operation of steps b1 to b4 in FIG.

FIG. 17 is a diagram showing operations of steps b5 and b6 in FIG.

FIG. 18 is a diagram showing the operation of steps b7 to b9 in FIG.

FIG. 19 is a diagram showing the operation of steps b10 and b11 in FIG.

FIG. 20 shows each color R,
It is a figure showing a histogram for every G and B.

FIG. 21 shows a second example having second density level ranges ZL to ZH after deleting density levels in which the number of pixels is less than a predetermined value.
FIG. 21A shows a histogram of image data, and FIG.
21 (2) shows the histogram of the color G, and FIG. 21 (3) shows the histogram of the color B.

FIG. 22 is a diagram for explaining the operation of the all-density average conversion circuit 45;

FIG. 23 is a diagram illustrating the number of pixels having an average value Em.

24 is a diagram for explaining the operation of the reference density conversion circuit 46. FIG. 24A shows a histogram of the second image data before the dynamic range is compressed, and FIG.
4 (2) shows a histogram of the second image data whose dynamic range has been compressed.

FIG. 25 is a histogram of third image data for each color R, G, B.

FIG. 26 illustrates prior art dynamic range compression.

[Explanation of symbols]

 7 Camera 8 Analog / Digital Converter 9 Dynamic Range Extender 10 Dynamic Range Compressor 11 Image Sensor 12 Condenser Lens 15 Image Switching Circuit 17, 18 Arithmetic Circuit 23 Substitution Circuit 24, 25 Density Conversion Circuit 26, 27 Color Temperature Correction Circuit 37 memory 38 histogram calculation circuit 39 average value calculation means 41, 42 density level 44 density conversion circuit 45 total density average conversion circuit 46 reference density conversion circuit M1, M2, M3 memory

Claims (18)

[Claims] 1. A density level in which the number of pixels in a first image data in which the density level of each pixel is in a first density level range is less than a predetermined value is deleted, and the deleted density level in the first density level range is deleted. The density level whose number of pixels is equal to or greater than the predetermined value is shifted to one side of the density level, and the density level shifted to the one side is assigned to the pixel having the deleted density level. Set,
The second image data having the second density level range is obtained, and the average value Em of the density levels of the first image data is assigned to the assigned values Zm1 and Zm2 within the second density level range, and is set. Zm2 is set to a third density level range CL to CH narrower than the first density level range of the first image data.
The upper and lower ranges of the density level assignment values Zm1 and Zm2 of the second image data are set in correspondence with the predetermined reference density levels Cc in the above. A method for compressing the dynamic range of an image, characterized by distributing the range. 2. A first image data in which each pixel has a plurality of density levels within a first density level range, and an average value Em of the density levels of the first image data is obtained. The density level whose number of pixels is less than the predetermined value is deleted, and the density level whose number of pixels is equal to or more than the predetermined value is shifted toward the lower or higher density level by the deleted density level. hand,
For the pixels having the deleted density levels, the shifted density is set to obtain the second image data having the second density level range, and the number of pixels having the average value Em is determined in advance. If it is less than the value, the density level on the side shifted by the density level of the deleted pixel including the average value Em within the second density level range of the second image data is assigned to the assigned value Zm of the average value Em.
When the number of pixels having the average value Em is equal to or greater than the predetermined value, the density level of the pixel of the second image data having the average value Em is within the second density level range of the second image data. Is set to an assigned value Zm2, and these assigned values Zm1 and Zm2 are set corresponding to a predetermined reference density level Cc in a third density level range CL to CH narrower than the first density level range. 2 Allocation values Zm1 in the concentration level ranges ZL to ZH
The density level in each range between Zm2 and the values ZL and ZH at both ends of the density level in the second density level range is defined as the reference density level Cc in the third density level range and the third density level in the third density level range.
A method for compressing a dynamic range of an image, wherein the dynamic range of an image is set by distributing and setting the density level in each range between values CL and CH at both ends of the density level of the density level range. 3. In the above-mentioned distribution, the first compression coefficient kL, kL = (Cc-CL) / (Zm-ZL) in the lower density level range with respect to the reference density level Cc in the third density level range. With respect to the reference density level Cc in the third density level range, the second compression coefficient kH, kH = (CH-Cc) / (ZH-Zm) in the range of the upper density level is obtained. The density level C is set to the density level Z in the second density level range ZL to ZH of the second image data.
In the range where Z <Zm, C = Cc-kL (Zm-Z) is obtained. When the concentration level Z in the second concentration level range ZL to ZH is Z ≧ Zm, C = Cc− 3. The method for compressing the dynamic range of an image according to claim 2, wherein the value is determined by kH (Z-Zm). 4. The first image data is color image data for each of red R, green G, and blue B, and the average value Em is an average value of density levels of all colors. A method for compressing the dynamic range of an image according to one of the preceding claims. 5. The method according to claim 1, wherein when there is no density level in which the number of pixels in the second image data is less than the predetermined value, the first image data is used as the second image data. A method for compressing the dynamic range of an image according to one of the preceding claims. 6. A memory for storing first image data in which each pixel has a plurality of density levels within a first density level range; and (b) a density of the first image data stored in the memory. Average value calculating means for calculating an average value Em of the level; (c) removing, from the first image data, a density level in which the number of pixels is less than a predetermined value, and reducing the number of pixels by the deleted density level. The density level that is equal to or higher than the predetermined value is shifted to the lower or higher density level side, and the density of the shifted side is set for a pixel having the deleted density level. Means for creating second image data having a density level range of 2; and (d) responding to an output of the second image data creating means, wherein the number of pixels having an average value Em is less than the predetermined value. Two image data In the second density level range of data, the density level of the deleted concentration levels fraction of staggered side of pixels including the average value Em, assigned value of the average value Em Zm
When the number of pixels having the average value Em is equal to or greater than the predetermined value, the density level of the pixel of the second image data having the average value Em is within the second density level range of the second image data. (E) responding to the outputs of the second image data creating means and the assignment value setting means, and assigning these assignment values Zm1 and Zm2 to the first density level. Assigned values Zm1 in the second density level ranges ZL to ZH are set corresponding to predetermined reference density levels Cc in third density level ranges CL to CH that are smaller than the range.
The density level in each range between Zm2 and the values ZL and ZH at both ends of the density level in the second density level range is defined as the reference density level Cc in the third density level range and the third density level in the third density level range.
Means for generating third image data by distributing and setting the density levels in the respective ranges between the values CL and CH at both ends of the density level of the density level range. Device for compressing. 7. The third image data creating means includes: (a) a reference density level Cc in a third density level range
The first compression coefficient calculating means for obtaining the first compression coefficient kL, kL = (Cc-CL) / (Zm-ZL) in the lower density level range, and (b) the reference density level in the third density level range Cc
, A second compression coefficient calculating means for obtaining a second compression coefficient kH, kH = (CH-Cc) / (ZH-Zm) in the range of the above density level; and (c) first and second compression coefficient calculating means. , The density level C in the third density range is changed to the second density range Z.
In a range where the density level Z in L to ZH is Z <Zm, C = Cc-kL (Zm-Z), and the density level Z in the second density range ZL to ZH is Z ≧ Zm. 7. The apparatus for compressing the dynamic range of an image according to claim 6, further comprising: a compressed density level calculating means for determining the range by C = Cc-kH (Z-Zm). 8. The first image data is color image data for each color of red R, green G, and blue B, and the average value calculating means calculates the average value of the density levels of the color image data of all colors. The apparatus for compressing a dynamic range of an image according to claim 6 or 7, wherein: 9. The method according to claim 6, wherein the first image data is used as the second image data when there is no density level in which the number of pixels in the first image data is less than the predetermined value. Apparatus for compressing the dynamic range of an image according to one of the preceding claims. 10. (a) From the image data of a plurality of screens obtained by imaging the same subject with different exposure amounts to obtain a single image data having an extended dynamic range, the exposure amounts are mutually different. In the different image data, the density levels at both ends or one end of the dynamic range of the reference image data in the same subject portion are adjusted to the density levels α1 and α2 of the image data having different exposure amounts, and the same subject The change amount (F2b-) of the density level of image data having different exposure amounts with respect to the change amount (Fb-Fa, Fd-Fc) of the density level of the reference image data in the portion
F2a, an exposure coefficient β2 which is a ratio of F1d-F1c)
m, β1m to correct the other image data with respect to the other image data to extend the dynamic range of the image. As image data, in compressing the dynamic range of the first image, a density level in which the number of pixels in the first image data in which the density level of each pixel is within the first density level range is less than a predetermined value is deleted. The density level whose number of pixels is equal to or more than the predetermined value is shifted to one side of the density level by the deleted density level within one density level range, and the pixel having the deleted density level is A density level shifted to one side is set to obtain second image data having a second density level range, and an average value Em of density levels of the first image data is obtained. Sets assigned to assigned values Zm1, Zm2 in the second concentration level range, the assigned value Zm1, Zm2, narrower than the first concentration level range of the first image data third concentration level range CL~CH
The upper and lower ranges of the density level assignment values Zm1 and Zm2 of the second image data are set in correspondence with the predetermined reference density levels Cc in the above. A method for processing an image, wherein the image is distributed to a range. 11. When there is no density level in which the number of pixels in the first image data is less than the predetermined value,
The image processing method according to claim 10, wherein the first image data is used as the second image data. 12. (a) a first memory for storing first image data having a plurality of density levels captured at a certain exposure amount; and (b) an exposure amount larger than the exposure amount of the first image data. A second memory for storing second image data having a plurality of density levels obtained by imaging the same subject as the first image data; and (c) a dynamic density level of the first image data stored in the first memory. Within the range, among a plurality of predetermined density levels Fa and Fb on the side corresponding to the larger one of the exposure amounts, the density level Fa outside the dynamic range and the density level further outside the dynamic range, And the predetermined outward concentration level F
a first extracting means for extracting the coordinates of a pixel in the first memory having a density level of a density level Fb which is inside of a, and (d) a second extracting means having the same coordinates as the coordinates extracted by the first extracting means. (E) in response to the output of the second extracting means, for each of the predetermined density levels Fa and Fb, the density level of the pixel extracted from the second memory; And (f) responding to the output of the average value calculation means to adjust the density level of the second image data to the density level of the first image data in response to the output of the average value calculation means. A density level difference α2, α2 = Fa−F2a with respect to a density level Fa outside the dynamic range is obtained, and further, the density level of the first image data with respect to the gradient of the density level of the second image data is obtained. (G) responding to the output of the second extracting means, the first calculating means for obtaining the exposure coefficient β2m, β2m = (Fb-Fa) / (F2b-F2a), which is the ratio of the inclination of the first memory. When the density level of a pixel in the second memory having the same coordinates as the coordinates of each pixel having a density level further outside the dynamic range than the predetermined density levels Fa and Fb is F2old, the corrected image data Concentration level F2new, F2new = Fa + (F2old + α2-Fa) × β
(H) the corrected image data obtained by the second operation means is first image data in which each pixel has a plurality of density levels within the first density level range. A third memory for storing the first image data; (i) an average value calculating means for calculating an average value Em of density levels of the first image data stored in the third memory; In the image data, the density level whose number of pixels is less than the predetermined value is deleted, and the density level whose number of pixels is equal to or more than the predetermined value is reduced by the deleted density level. Means for setting the density of the shifted side to a pixel having the density level shifted and removed to the other side to create second image data having a second density level range; k) Second When the number of pixels having the average value Em is less than the predetermined value in response to the output of the image data creating means, the deleted pixels including the average value Em within the second density level range of the second image data. The density level on the side shifted by the density level of is assigned to the assigned value Zm of the average value Em.
When the number of pixels having the average value Em is equal to or greater than the predetermined value, the density level of the pixel of the second image data having the average value Em is within the second density level range of the second image data. (1) responding to the outputs of the second image data creation means and the assignment value setting means, and assigning these assignment values Zm1 and Zm2 to the first density level. Assigned values Zm1 in the second density level ranges ZL to ZH are set corresponding to predetermined reference density levels Cc in third density level ranges CL to CH that are smaller than the range.
The density level in each range between Zm2 and the values ZL and ZH at both ends of the density level in the second density level range is defined as the reference density level Cc in the third density level range and the third density level in the third density level range.
Means for generating third image data by distributing and setting the density levels in the respective ranges between the values CL and CH at both ends of the density level of the density level range. . 13. (a) a first memory for storing first image data having a plurality of density levels captured at a certain exposure amount; and (b) an exposure amount smaller than the exposure amount of the first image data. A second memory for storing second image data having a plurality of density levels obtained by imaging the same subject as the first image data; and (c) a dynamic density level of the first image data stored in the first memory. Among the plurality of predetermined density levels Fd and Fc on the side corresponding to the smaller one of the exposure amounts within the range, the density level Fd that is closer to the outside of the dynamic range and the density level that is further closer to the outside of the dynamic range; And the predetermined outward concentration level F
(d) a second extraction unit having the same coordinates as the coordinates extracted by the first extraction unit. (E) in response to the output of the second extracting means, for each of the predetermined density levels Fd and Fc, the density level of the pixel extracted from the second memory. And (f) responding to the output of the average value calculating means to adjust the density level of the second image data to the density level of the first image data. A density level difference α1, α1 = Fd−F1d relating to a density level Fd located outside of the dynamic range is obtained, and further, the density level of the first image data with respect to the gradient of the density level of the second image data is obtained. (G) responding to the output of the second extracting means, the first calculating means for obtaining the exposure coefficient β1m, β1m = (Fd-Fc) / (F1d-F1c), which is the ratio of the inclination of the first memory. When the density level of a pixel in the second memory having the same coordinates as the coordinates of each pixel having a density level further outside the dynamic range than the predetermined density levels Fd and Fc is F1old, the corrected image data Concentration level F1new, F1new = Fd + (F1old + α1-Fd) × β
(H) the corrected image data obtained by the second operation means is first image data in which each pixel has a plurality of density levels within the first density level range. A third memory for storing the first image data; (i) an average value calculating means for calculating an average value Em of density levels of the first image data stored in the third memory; In the image data, the density level whose number of pixels is less than the predetermined value is deleted, and the density level whose number of pixels is equal to or more than the predetermined value is reduced by the deleted density level. Means for setting the density of the shifted side to a pixel having the density level shifted and removed to the other side to create second image data having a second density level range; k) Second When the number of pixels having the average value Em is less than the predetermined value in response to the output of the image data creating means, the deleted pixels including the average value Em within the second density level range of the second image data. The density level on the side shifted by the density level of is assigned to the assigned value Zm of the average value Em.
When the number of pixels having the average value Em is equal to or greater than the predetermined value, the density level of the pixel of the second image data having the average value Em is within the second density level range of the second image data. (1) responding to the outputs of the second image data creation means and the assignment value setting means, and assigning these assignment values Zm1 and Zm2 to the first density level. Assigned values Zm1 in the second density level ranges ZL to ZH are set corresponding to predetermined reference density levels Cc in third density level ranges CL to CH that are smaller than the range.
The density level in each range between Zm2 and the values ZL and ZH at both ends of the density level in the second density level range is defined as the reference density level Cc in the third density level range and the third density level in the third density level range.
Means for generating third image data by distributing and setting the density levels in the respective ranges between the values CL and CH at both ends of the density level of the density level range. . 14. When there is no density level in which the number of pixels in the first image data is less than the predetermined value,
14. The image processing apparatus according to claim 12, wherein the first image data is used as the second image data. 15. A first detection level range in which a detection level of a plurality of pieces of detection data with the passage of time is within a first detection level range and a detection level in which the number of first detection data is less than a predetermined value is deleted. The detection level whose number of data is equal to or greater than the predetermined value is shifted to one side of the detection level by the amount of the deleted detection level, and the data having the deleted detection level is included in the one side. By setting the shifted detection level, a second detection level having a second detection level range is obtained, and the average value Em of the detection levels of the first detection data is changed to the assigned values Zm1 and Zm2 within the second detection level range. The allocation values Zm1 and Zm2 are set to a third detection level range CL to CH smaller than the first detection level range of the first detection data.
, The upper and lower ranges of the detection level assignment values Zm1 and Zm2 of the second detection data, and the upper and lower ranges of the detection level of the third detection level range with respect to the reference detection level Cc. A method for compressing a dynamic range of a detector, characterized by distributing the range. 16. The detector according to claim 15, wherein when there is no detection level in which the number of data in the first detection data is less than the predetermined value, the first detection data is used as the second detection data. A method for compressing the dynamic range of an image. 17. A memory for storing first detection data having a plurality of detection levels within a first detection level range, wherein: (a) a plurality of detection data over time; and (b) a memory for storing the first detection data. Average value calculating means for calculating an average value Em of the detection levels of the first detection data; and (c) deleting, from the first detection data, detection levels whose number of data is less than a predetermined value, and deleting the deleted detection levels. By the amount, the detection level at which the data is equal to or more than the predetermined value is shifted to the lower or higher side of the detection level, and the pixel having the deleted detection level is provided with the detection level of the shifted side. And (d) responding to the output of the second detection data generating means, and the data having an average value Em is stored in advance in response to the output of the second detection data generating means. When the detected level is smaller than the threshold value, the detection level on the side shifted by the detection level of the deleted data including the average value Em within the second detection level range of the second detection data is assigned to the average value Em. Value Z
m1, and when the data having the average value Em is equal to or larger than the predetermined value, the detection level of the second detection data having the average value Em is assigned within the second detection level range of the second detection data. (E) responding to each output of the second detection data creating means and the assigned value setting means, and setting these assigned values Zm1 and Zm2 to be smaller than the first detection level range. It is set corresponding to a predetermined reference detection level Cc within the narrow third detection level range CL to CH, and the assigned value Zm1 in the second detection level range ZL to ZH is set.
The detection level in each range between Zm2 and the values ZL and ZH at both ends of the detection level in the second detection level range is defined as the reference detection level Cc in the third detection level range and the third detection level in the third detection level range.
Means for generating the third detection data by distributing and setting the detection level in each range between the values CL and CH at both ends of the detection level of the detection level range. A device for compressing a range. 18. The detector according to claim 17, wherein when there is no detection level in which the number of data in the first detection data is less than the predetermined value, the first detection data is used as the second detection data. For compressing the dynamic range of a computer.