JP2010213745A - Endoscopic image processing device and method and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To generate a spectral estimated image representing the spectral reflectance of an actual object with good accuracy. <P>SOLUTION: A parameter database DB storing parameter sets differing with color is provided. Chromaticity coordinates (x, y) are discriminated as color based on the color of image data (a pixel or a predetermined region) of an endoscopic image P. After that, one of the parameter sets MPS1-MPS6 to be used in matrix operation is selected based on the discriminated chromaticity coordinates (x, y). A spectral estimated image SP is generated using the selected parameter set SMPS. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an endoscopic image processing apparatus, method, and program for generating a spectral estimated image by performing matrix calculation on an endoscopic image acquired using an endoscope.

  In recent years, in an electronic endoscope apparatus using a solid-state imaging device, based on spectral reflectance in a digestive organ (gastric mucosa, etc.), spectral imaging combined with a narrow-band bandpass filter, that is, an electronic endoscope apparatus with a built-in narrow-band filter (Narrow Band Imaging-NBl) is drawing attention. This device is provided with three narrow (wavelength) band-pass filters instead of the surface sequential R (red), G (green), and B (blue) rotary filters. By sequentially performing the same processing as in the case of the R, G, B (RGB) signals while changing the respective weights for the three signals obtained with these illumination lights, the spectral image is output. Is formed. According to such a spectral image, in the digestive organs such as the stomach and the large intestine, a fine structure or the like that has not been obtained conventionally is extracted.

  On the other hand, it has been proposed that a spectral image is formed by arithmetic processing based on an image signal obtained with white light, instead of a frame sequential type using the above-described narrowband bandpass filter (for example, Patent Documents). 1). This is obtained as a matrix data (coefficient set) between the RGB color sensitivity characteristics converted into numerical data and the spectral characteristics of a specific narrowband bandpass converted into numerical data. A spectral image signal obtained by estimating a spectral image obtained through a narrow-band bandpass filter by calculation with an RGB signal is obtained. When a spectral image is formed by such an operation, it is not necessary to prepare a plurality of filters corresponding to a desired wavelength region, and replacement arrangement of these is unnecessary, so that the apparatus can be prevented from being enlarged and reduced in size. Cost can be reduced. The matrix parameters described above are set for each wavelength, for example, at intervals of 5 nm from 400 nm to 700 nm, and a parameter corresponding to the wavelength of the spectral estimation image to be displayed is selected.

  This matrix parameter is calculated from, for example, an image obtained by photographing a color chart having a known spectral reflectance (for example, a 24-color Macbeth chart). Specifically, matrix parameters for each patch of the color chart are detected, and nonlinear parameters such as spline interpolation and Lagrangian interpolation are applied to the detected matrix parameters, thereby generating matrix parameters at intervals of 5 nm from 400 nm to 700 nm. . Thereby, it is possible to acquire matrix parameters having a smooth distribution while increasing the number of parameters.

JP 2003-93336 A

  However, when the matrix parameter is created by performing nonlinear interpolation as described above, there may be a case where an error between the true spectral reflectance of the subject and the spectral estimation image obtained by the calculation becomes large. That is, as described above, when nonlinear interpolation is performed for a wide wavelength range from 400 nm to 700 nm, for example, there is a wavelength range in which the matrix parameter value is deviated from the actual spectral reflectance, and an error in a specific wavelength range. The error may become large. There is a problem that in the wavelength range where this deviation is large, the error becomes large, which causes the accuracy of the spectral estimation image to deteriorate.

  Accordingly, an object of the present invention is to provide an endoscope image processing apparatus, method, and program capable of accurately expressing the spectral reflectance of an actual subject in a spectral estimated image.

  The endoscopic image processing apparatus of the present invention generates a spectral estimation image representing the spectral reflectance of a subject by performing matrix calculation on the endoscopic image using a parameter set to which matrix parameters are assigned for each wavelength. An endoscopic image processing apparatus for determining a color of image data of an endoscopic image, a parameter database storing a plurality of parameter sets different for each color, and a color determined by the color determining means And a parameter selection unit that selects a parameter set corresponding to the parameter database, and a calculation unit that generates a spectral estimation image by performing matrix calculation using the parameter set selected by the parameter selection unit. To do.

  The endoscopic image processing method of the present invention generates a spectral estimation image representing the spectral reflectance of a subject by performing a matrix operation on the endoscopic image using a parameter set to which matrix parameters are assigned for each wavelength. An endoscopic image processing method for determining the color of image data of an endoscopic image and selecting a parameter set corresponding to the determined color from a parameter database storing a plurality of parameter sets different for each color The spectral estimation image is generated by performing a matrix operation using the selected parameter set.

  The endoscopic image processing program of the present invention is a spectral estimation that represents a spectral reflectance of a subject by performing a matrix operation on an endoscopic image using a parameter set in which matrix parameters are assigned to wavelengths for a computer. An endoscopic image processing program for executing generation of an image, determining a color of image data of an endoscopic image, from a parameter database storing a plurality of parameter sets different for each color, A parameter set corresponding to the discriminated color is selected, and a spectral calculation image is generated by performing matrix calculation using the selected parameter set.

  Here, the parameter database only needs to store a plurality of parameter sets for each color. For example, the parameter database may have a parameter set for each different region of the chromaticity coordinate space on the chromaticity diagram. At this time, the color discriminating unit discriminates the chromaticity coordinate as a color, and the parameter selecting unit selects a parameter set corresponding to the chromaticity coordinate discriminated by the color discriminating unit from the parameter database. Alternatively, the color discriminating means discriminates colors such as “red” and “red purple”, and the parameter database may store parameter sets corresponding to the respective colors.

  Each parameter set may be created using a color chart having a plurality of colors in each region on the chromaticity coordinate space corresponding to each parameter set as a color patch.

  Furthermore, the color discriminating unit may be any unit that discriminates the color of the image data of the endoscopic image. For example, the color discriminating unit may discriminate the color for each pixel of the endoscopic image. At this time, the parameter selection unit selects a parameter set for each pixel, and the spectral image generation unit generates a spectral estimation image by performing matrix calculation for each pixel using the parameter set selected by the parameter selection unit.

  Alternatively, the color discriminating unit may divide the endoscope image into predetermined regions and discriminate colors for each region based on the sum of the colors of the pixels in the predetermined region. At this time, the parameter selection unit selects a parameter set for each predetermined region, and the spectral image generation unit generates a spectral estimation image by performing matrix calculation for each pixel using the parameter set selected by the parameter selection unit. . The predetermined area may be, for example, a region of interest, or the endoscopic image is divided into block areas of predetermined pixels (for example, 8 × 8 pixels), and the color is determined for each block area. May be. Further, the color can be discriminated by any method, for example, by determining the color of the region of interest based on the summation of each pixel such as the most frequently used color or the average value of the colors in each pixel in the region of interest.

  According to the endoscopic image processing apparatus, method, and program of the present invention, the color of the image data of the endoscopic image is discriminated, and the discriminated color is selected from the parameter database storing a plurality of different parameter sets for each color. By selecting a corresponding parameter set and performing a matrix operation using the selected parameter set to generate a spectral estimation image, a spectral estimation image using the optimal matrix parameters for each color of the image data of the endoscopic image Compared to the case where a spectral estimation image is generated using a single matrix parameter set, a spectral estimation image that accurately represents the spectral reflectance of the subject can be generated. Can do.

  The parameter database has a parameter set for each different region of the chromaticity coordinate space on the chromaticity diagram, and the color discrimination means discriminates the chromaticity coordinate on the chromaticity diagram as a color. When the means selects from the parameter database a parameter set corresponding to the chromaticity coordinates determined by the color determination means, accurate color determination can be performed and a spectral estimation image can be generated using the optimum parameter set. Therefore, it is possible to prevent the occurrence of errors and generate a spectral estimation image with high accuracy.

  Furthermore, if each parameter set is created by using a color chart having a plurality of colors in each region on the chromaticity coordinate space corresponding to each parameter set as a color patch, the matrix parameter is determined by nonlinear interpolation. Even when the number is increased, the occurrence of errors due to interpolation can be minimized.

  The color discriminating unit discriminates the color from the image data of each pixel of the endoscopic image, the parameter selecting unit selects a parameter set for each pixel, and the spectral image generating unit is selected by the parameter selecting unit. If a spectral estimation image is generated by performing matrix calculation for each pixel using the parameter set, a spectral estimation image can be generated using a parameter set that is optimal for the color of each pixel. Generation of an error can be prevented and a spectral estimation image can be generated with high accuracy.

  Further, the color discriminating means divides the endoscopic image into predetermined areas, and discriminates colors for each divided area based on the sum of colors of each pixel in the area, and the parameter selecting means for each area. When selecting a parameter set, the spectral image generation unit generates a spectral estimation image by performing matrix calculation for each region using the parameter set selected by the parameter selection unit. Since a spectral estimation image can be generated using a simple parameter set, the generation of an error can be prevented and the spectral estimation image can be generated accurately and efficiently.

The block diagram which shows an example of the endoscopic apparatus to which the endoscopic image processing apparatus of this invention was applied. Table showing an example of chromaticity coordinates determined by the color determining means of FIG. Schematic diagram showing the relationship between multiple parameter sets and chromaticity coordinate space Table showing an example of matrix parameters stored in the parameter database of FIG. The flowchart which shows preferable embodiment of the endoscopic image processing method of this invention.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a block diagram showing an example of an endoscope apparatus of the present invention. The endoscope apparatus 1 includes a light source unit 10, a scope 20, and an endoscope image processing apparatus 30. The light source unit 10 irradiates a subject with light for observation by an endoscope, and irradiates white light for normal observation of a xenon lamp or the like. The light source unit 10 is optically connected to the light guide 15 of the scope 20 via the optical fiber 11 and the condenser lens 13, and the white light L1 emitted from the light source unit 10 enters the light guide 15 and enters the observation window. The object is irradiated from 16.

  The scope 20 includes an imaging lens 21, an imaging means 22, a CDS / AGC circuit 23, an A / D converter 24, a CCD driving unit 25, a lens driving unit 26, and the like, and each component is controlled by a scope controller 27. Has been. The imaging lens 21 is composed of, for example, a plurality of lens groups, and the photographing magnification is changed by driving the lens driving unit 26. The imaging means 22 is composed of, for example, a CCD or a CMOS, and obtains an image by photoelectrically converting the subject image formed by the imaging lens 21. As the imaging means 22, for example, a complementary color type having Mg (magenta), Ye (yellow), Cy (cyan), G (green) color filters or a primary color type having RGB color filters on the imaging surface is used. . Note that the operation of the imaging means 22 is controlled by the CCD drive unit 25. When the image pickup means 22 acquires an image (video) signal, a CDS / AGC (correlated double sampling / automatic gain control) circuit 23 samples and amplifies, and an A / D converter 24 outputs from the CDS / AGC circuit 17. The obtained endoscopic image is A / D converted and output to the endoscopic image processing apparatus 30.

  The endoscopic image processing apparatus 30 processes an endoscopic image acquired using the scope 20, and is configured by, for example, a DSP. The endoscopic image processing apparatus 30 includes an image acquisition unit 31, a preprocessing unit 32, a spectral image generation unit 33, an image processing unit 34, and a display control unit 35. The image acquisition unit 31 acquires an endoscopic image P captured by the imaging unit 22 of the scope 20.

  The preprocessing unit 32 performs preprocessing on the endoscopic image P acquired by the image acquisition unit 31. For example, when the endoscopic image P is composed of the YCC color system, an RGB table is used. It converts to a color system, and further has a gamma conversion function, a function of adjusting gradation, and the like.

  The spectral image generating means 33 generates a spectral estimated image SP by performing matrix calculation on the endoscope image P using the matrix parameter M. Note that details of an operation example of the spectral image generation unit 33 are described in JP-A-2003-93336.

なお、式（１）において、ＳＰｒ、ＳＰｇ、ＳＰｂは分光推定画像ＳＰの各ＲＧＢ成分、Ｐｒ、Ｐｇ、Ｐｂは内視鏡画像Ｐの各ＲＧＢ成分、Ｍ_００〜Ｍ_２２からなる３×３行の行列はマトリクス演算を行うためのマトリクスパラメータＭをそれぞれ示している。 Specifically, the spectral image generation means 33 generates a spectral estimation image SP by performing a matrix calculation represented by the following formula (1).

In Equation (1), SPr, SPg, SPb are RGB components of the spectral estimated image SP, Pr, Pg, Pb are RGB components of the endoscopic image P, and 3 × 3 rows comprising M _{00 to} M _22. These matrixes respectively indicate matrix parameters M for performing matrix operations.

  The image processing unit 34 performs enhancement processing or the like on the endoscopic image P and the spectral estimation image SP, and the display control unit 35 performs the endoscopic image P and spectral estimation image SP that have been image-processed by the image processing unit 34. Is displayed on the display device 3 together with character information and the like. In particular, the display control unit 35 has a function of displaying the normal observation image and the spectral estimation image SP when the white light L1 is irradiated as the endoscope image P.

  Here, the above-described spectral image generation unit 33 has a function of performing the matrix calculation of Expression (1) using a matrix parameter M that is different for each color of each pixel. Specifically, the spectral image generation unit 33 includes a color determination unit 40, a parameter selection unit 50, and a calculation unit 60.

  The color discriminating means 40 discriminates the color of the endoscopic image. For example, as shown in FIG. 2, the color represented by the RGB color system is represented by a chromaticity coordinate space (x, y) by a known method. The color discrimination means 40 discriminates the chromaticity coordinates (x, y) based on the component signal values of the R component, G component, and B component of each pixel. .

  In addition, although the case where the color discriminating unit 40 discriminates the color for each pixel of the endoscopic image P and selects the parameter set is illustrated, the endoscopic image P is divided into a plurality of regions, and the color is determined for each region. You may make it discriminate | determine. For example, the color determination unit 40 may divide the endoscopic image into block areas of predetermined pixels (for example, 8 × 8 pixels) and determine the color for each of the divided block areas. The color of each block area is determined based on the total of the average value of colors of a plurality of pixels in the block area, the most frequently used color, or the like. Thus, the spectral estimation image SP can be generated with high accuracy at high speed by reducing the number of parameter set selection processes.

  The parameter selection unit 50 selects a parameter set SMPS corresponding to the color determined by the color determination unit 40 from among the plurality of parameter sets MPS1 to MPS6 stored in the parameter database DB. Here, a plurality of parameter sets MPS1 to MPS6 corresponding to each color are stored in the parameter database DB. FIG. 3 is a schematic diagram showing an example in which a plurality of parameter sets MPS1 to MPS6 exist on the chromaticity coordinate space. In FIG. 3, the center point C is, for example, the chromaticity coordinate of white light, and the color purity increases toward the outside from the point C, and the outermost line indicates the chromaticity coordinate of the spectrum light (the highest color purity). Light). Therefore, the point where the straight line connecting the predetermined point P and the point C intersects the chromaticity coordinates of the spectrum is the dominant wavelength (or complementary dominant wavelength) of the chromaticity coordinates P, and the line segment from the point C to the point P Length indicates purity.

Each parameter set MPS1 to MPS6 includes, for example, matrix parameters M = (M _j0 , M _j1 , M _j2 ) (i = 1 to 61, j is a matrix) for each wavelength range obtained by dividing a wavelength range from 400 nm to 700 nm at 5 nm intervals. In the line of parameter M, j = 0 to 2) is set. FIG. 4 is a table showing an example of the parameter set MPS1, which is one of the parameter sets. Note that FIG. 3 illustrates the case where the image is divided into six regions in the chromaticity coordinate space, but may be divided into two or more regions. Further, as shown in MPS3 to MPS6, the region may be divided for each different main wavelength (complementary color main wavelength), or the region may be divided for each purity even with the same main wavelength as in the parameter sets MPS1 and MPS2. It may be divided.

  Each parameter set MPS1 to MPS6 is created as follows. For example, in the case of the parameter set MPS1, a color chart composed of a plurality of color patches with known spectral reflectances composed of colors on the chromaticity space corresponding to the parameter set MPS1 is prepared. A parameter set MPS1 is created from the relationship between each RGB signal component value and the spectral reflectance of each color patch from an image obtained by photographing this color chart. At this time, in order to create matrix parameters at intervals of 5 nm, nonlinear interpolation processing such as spline interpolation is performed on the matrix parameters detected from each color patch. The parameter sets MPS2 to MPS6 are also created by the method described above.

  The calculation means 60 in FIG. 1 generates a spectral estimation image SP by performing matrix calculation using the parameter set SMPS selected by the parameter selection means 50. Here, as the wavelength to be assigned to the RGB of the spectral estimation image SP in the formula (1), a wavelength set in a wavelength range suitable for the observation site is prepared. For example, any wavelength set is used from an input means such as a mouse or a keyboard. Is entered. Specifically, for example, (λ1, λ2, λ3) = (400, 500, 600) standard set CH1, (λ1, λ2, λ3) = (470, 500, 670) or (475 for rendering blood vessels , 510, 685) and (λ1, λ2, λ3) = (440, 480, 520) or (480, 510, 580) tissue set CH5, oxyhemoglobin for rendering a specific tissue, (Λ1, λ2, λ3) = (400, 430, 475) hemoglobin set CH6 for depicting the difference from deoxyhemoglobin, (λ1, λ2, λ3) = (for depicting the difference between blood and carotene) 415, 450, 500) blood-carotene set CH7, (λ1, λ2, λ3) = (420, 550, 600) blood for depicting the difference between blood and cytoplasm - Eight wavelength set cytoplasmic set CH8 are stored. And the calculating means 60 produces | generates the spectrum estimation image SP based on said Formula (1) using the matrix parameter M applicable to the input wavelength set among the selected parameter sets SMPS.

  In this way, by preparing the spectral estimation image SP by preparing different parameter sets MPS1 to MPS6 for each color, the spectral estimation image SP can be generated with high accuracy while minimizing errors. That is, since the parameter set is created using a color chart, nonlinear interpolation processing is required to create matrix parameters for wavelength regions other than the color patch. In this non-linear interpolation processing, when a wavelength range of 400 to 700 nm is created with one color set as in the prior art, an error may increase in a specific wavelength range.

  On the other hand, data interpolation is performed by dividing the chromaticity coordinate space into a plurality of regions (see FIG. 3), preparing a color chart for each color space region, and creating a plurality of parameter sets MPS1 to MPS6 from the plurality of color charts. Even if processing is performed, the error can be reduced. As a result, the spectral estimation image SP and the spectral characteristics and errors of the actual subject are reduced using the matrix parameter, and the spectral estimation image SP with high accuracy can be generated.

  FIG. 5 is a flowchart showing a preferred embodiment of the endoscopic image processing method of the present invention. The endoscopic image processing method will be described with reference to FIGS. First, by performing imaging while the scope 20 is inserted into the body cavity, an endoscopic image P is obtained (step ST1), and the endoscopic image P is preprocessed by the preprocessing means 32. . Thereafter, the chromaticity coordinates (x, y) are determined as colors based on the color of the image data (pixels or predetermined areas) of the endoscopic image P in the color determining means 40 (step ST2). Thereafter, the parameter selection unit 50 selects one of the parameter sets MPS1 to MPS6 used for matrix calculation based on the determined chromaticity coordinates (x, y) (step ST3). Then, a spectral estimation image SP is generated using the parameter set SMPS selected by the computing means 60 (step ST4), and is displayed and output on the display device 3 (step ST5).

  According to the above embodiment, the color of the endoscopic image P is discriminated, and the parameter set SMPS corresponding to the discriminated color is selected from the parameter database DB storing a plurality of parameter sets MPS1 to MPS6 different for each color. Then, by performing matrix calculation using the selected parameter set SMPS, the spectral estimation image SP can be generated using the optimum matrix parameter M for each color, so that one conventional matrix parameter set is used. As compared with the case where the spectral estimation image SP is generated, the generation of errors can be prevented and the spectral estimation image SP can be generated with high accuracy.

  Further, as shown in FIG. 3, the parameter database DB stores parameter sets MPS1 to MPS6 for different regions in the chromaticity coordinate space on the chromaticity diagram, and the color determining means 40 uses the chromaticity coordinates ( x, y) and the parameter selection means 50 selects the parameter set SMPS corresponding to the chromaticity coordinates (x, y) determined by the color determination means 40 from the parameter database DB. Since the spectral estimation image SP can be generated using an optimal parameter set in accordance with the accurate color purity and lightness, the spectral estimation image SP can be generated with high accuracy by preventing the occurrence of errors.

  Further, the color discriminating means 40 discriminates the color for each pixel of the endoscope image P, the parameter selecting means 50 selects the parameter set SMPS for each pixel, and the computing means 60 is the parameter selecting means. If the spectral estimation image SP is generated by performing matrix calculation for each pixel using the parameter set SMPS selected by 50, the spectral estimation image SP is obtained using the parameter set SMPS optimum for the color of each pixel. Therefore, the spectral estimation image SP can be generated with high accuracy by preventing the occurrence of errors.

  In addition, the color determining unit 40 determines a color for each predetermined region BR in the endoscopic image P based on the total of the colors of the pixels in the predetermined region, and the parameter selecting unit 50 determines the predetermined region. A parameter set SMPS is selected for each BR, and the calculation means 60 generates a spectral estimation image SP by performing matrix calculation for each predetermined region BR using the parameter set selected by the parameter selection means 50. In this case, since the spectral estimation image can be generated using the optimum parameter set SMPS for each region BR, it is possible to generate the spectral estimation image with high accuracy and accuracy by preventing the occurrence of errors.

  The embodiment of the present invention is not limited to the above embodiment. For example, FIG. 2 illustrates the case where five parameter sets MPS1 to MPS6 are stored in the parameter database DB, but two or more may be used. In particular, in an endoscopic image P obtained by imaging the body cavity, there are many cases where the chromaticity coordinates (x, y) of the parameter set MPS1 are obtained. You may make it set a set.

  Further, the parameter database illustrates a case where a plurality of parameter sets MPS1 to MPS6 are stored in the chromaticity coordinate space. For example, the parameter set for each color such as “red” and “red” shown in FIG. May be stored. At this time, the color discriminating unit discriminates which of the above colors is applicable.

  Moreover, although the case where it divides | segments for every area | region and divides | segments into a block area | region when discriminating a color is illustrated, in endoscopic image P which image | photographed the inside of a body cavity, specific chromaticity coordinates (x, y) The region of interest ROI may be extracted from the input means such as a mouse or automatically identified, and the color of the region ROI may be determined by the above-described method. The color may be determined by the above-described method with one entire endoscopic image as one region. Even in this case, by using the parameter sets MPS1 to MPS6 that match the color tendency of the region, it is possible to generate a spectral estimation image SP that accurately represents the spectral reflectance of the subject.

  1 illustrates the case where the configuration of the endoscopic image processing device 30 is configured by a DSP or the like in the endoscopic device 1 of FIG. 1, but an endoscopic image processing program read into the auxiliary storage device is illustrated. The endoscopic image processing program may be realized by being executed on a computer (for example, a personal computer). The endoscope image processing program is stored in an information storage medium such as a CD-ROM or a network such as the Internet. Will be distributed and installed on the computer.

DESCRIPTION OF SYMBOLS 1 Endoscope apparatus 20 Scope 30 Endoscope image processing apparatus 31 Image acquisition means 32 Preprocessing means 33 Spectral image generation means 34 Image processing means 35 Display control means 40 Color discrimination means 50 Parameter selection means 60 Calculation means DB Parameter database M Matrix parameters MPS1 to MPS6 Parameter set P Endoscopic image SP Spectral estimation image

Claims (7)

An endoscope image processing apparatus that generates a spectral estimation image representing a spectral reflectance of a subject by performing a matrix operation on an endoscopic image using a parameter set to which matrix parameters are assigned for each wavelength,
Color discriminating means for discriminating the color of the image data of the endoscopic image;
A parameter database storing a plurality of different parameter sets for each color;
Parameter selection means for selecting the parameter set corresponding to the color determined by the color determination means from the parameter database;
An endoscope image processing apparatus comprising: a calculation unit that generates a spectral estimation image by performing matrix calculation using the parameter set selected by the parameter selection unit. The parameter database has the parameter set for each different region of the chromaticity coordinate space on the chromaticity diagram;
The color discrimination means discriminates the chromaticity coordinates on the chromaticity diagram as the color;
2. The endoscopic image processing apparatus according to claim 1, wherein the parameter selection unit selects the parameter set corresponding to the chromaticity coordinates determined by the color determination unit from the parameter database.   The respective parameter sets are each created using a color chart having a plurality of colors in each region on the chromaticity coordinate space corresponding to the respective parameter sets as color patches. The endoscopic image processing apparatus according to 2.   The color discriminating unit discriminates a color from image data of each pixel of the endoscopic image, the parameter selecting unit selects the parameter set for each pixel, and the spectral image generating unit The endoscope according to any one of claims 1 to 3, wherein the spectral estimation image is generated by performing a matrix operation for each pixel using the parameter set selected by the parameter selection unit. Image processing device.   The color determining unit divides the endoscopic image into predetermined regions, and determines the color for each of the divided regions based on the total of the colors of the pixels in the region, and the parameter selecting unit Selects the parameter set for each region, and the spectral image generation unit performs the matrix calculation for each region using the parameter set selected by the parameter selection unit, thereby obtaining the spectral estimation image. The endoscope image processing apparatus according to claim 1, wherein the endoscope image processing apparatus generates the endoscope image processing apparatus. An endoscopic image processing method for generating a spectral estimation image representing a spectral reflectance of a subject by performing a matrix operation on an endoscopic image using a parameter set to which a matrix parameter is assigned for each wavelength,
Determining the color of the image data of the endoscopic image;
From the parameter database storing a plurality of different parameter sets for each color, select the parameter set corresponding to the determined color,
An endoscopic image processing method characterized in that a spectral estimation image is generated by performing a matrix operation using the selected parameter set. In order to cause the computer to generate a spectral estimation image representing the spectral reflectance of the subject by performing a matrix operation on the endoscopic image using a parameter set to which matrix parameters are assigned for each wavelength. An endoscopic image processing program,
Determining the color of the image data of the endoscopic image;
From the parameter database storing a plurality of different parameter sets for each color, select the parameter set corresponding to the determined color,
An endoscope image processing program for executing generation of a spectral estimation image by performing matrix calculation using the selected parameter set.

Images