High-frequency emphasis amount control device

The high-frequency enhancement amount control device addresses the challenge of manual adjustment by using discrete Fourier transforms and feedback control to automatically set the enhancement amount, achieving optimal image sharpness.

JP7800096B2Active Publication Date: 2026-01-16JVC KENWOOD CORP
View PDF 8 Cites 0 Cited by

Patent Information

Application Number
JP2021200758
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2021-12-10
Publication Date
2026-01-16
Estimated Expiration
2041-12-10

AI Technical Summary

Technical Problem

Existing image display devices require manual user input to adjust high-frequency component enhancement, leading to potential over- or under-emphasis, making it difficult to set an appropriate enhancement amount.

Method used

A high-frequency enhancement amount control device that automatically adjusts the enhancement of high-frequency components using discrete Fourier transforms, weighting filters, and feedback control loops to maintain a target emphasis ratio.

Benefits of technology

Automatically enhances high-frequency components with an appropriate amount, ensuring accurate image sharpness without user intervention.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007800096000001
    Figure 0007800096000001
  • Figure 0007800096000002
    Figure 0007800096000002
  • Figure 0007800096000003
    Figure 0007800096000003
Patent Text Reader

Abstract

To provide a high frequency enhancement amount control device capable of automatically enhancing high frequency components of an image with an appropriate enhancement amount.SOLUTION: A first discrete Fourier transform unit 11 performs two-dimensional discrete Fourier transform on an input image signal to generate a first frequency spectrum. A second discrete Fourier transform unit 21 performs two-dimensional discrete Fourier transform on an output image signal output from a high frequency enhancement circuit 30 to generate a second frequency spectrum. First and second high frequency component index generation units (sum calculation units 13 and 23) generate first and second high frequency component index indicating total amounts of high frequency components on the basis of the first and second frequency spectra, respectively. A high frequency component enhancement ratio calculation unit 4 calculates a high frequency component enhancement ratio which is the ratio between the first high frequency component index and the second high frequency component index. An enhancement amount control unit 5 generates an enhancement amount control value on the basis of the high frequency component enhancement ratio and the target enhancement ratio, and supplies the high frequency enhancement circuit 30 with the enhancement amount control value.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Description

[Technical Field]

[0001] The present invention relates to a high-frequency emphasis amount control device. [Background technology]

[0002] In order to improve the sharpness of an image displayed on an image display device, a high-frequency emphasis circuit that emphasizes the high-frequency components of the image is used (see Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2001-16480 Summary of the Invention [Problem to be solved by the invention]

[0004] The appropriate amount of enhancement when a high-frequency component of an image is enhanced by a high-frequency enhancement circuit varies depending on the content of the image. A user of an image display device manually sets the amount of enhancement of the high-frequency component while viewing the image displayed on the image display device. Setting the amount of enhancement manually is cumbersome for the user. Furthermore, when a user manually sets the amount of enhancement, the high-frequency component may be over-emphasized or under-emphasized, making it difficult to set an appropriate amount of enhancement. Therefore, there is a need for a high-frequency enhancement amount control device that can automatically enhance high-frequency components with an appropriate amount of enhancement.

[0005] An object of the present invention is to provide a high-frequency enhancement amount control device that can automatically enhance the high-frequency components of an image with an appropriate enhancement amount. [Means for solving the problem]

[0006] The present invention includes a first discrete Fourier transform unit that performs a two-dimensional discrete Fourier transform on an input image signal to generate a first frequency spectrum, a second discrete Fourier transform unit that performs a two-dimensional discrete Fourier transform on an output image signal output from a high-frequency emphasis circuit to generate a second frequency spectrum, a first high-frequency component index generation unit that generates a first high-frequency component index indicating the total amount of high-frequency components based on the first frequency spectrum, a second high-frequency component index generation unit that generates a second high-frequency component index indicating the total amount of high-frequency components based on the second frequency spectrum, and a ratio between the first high-frequency component index and the second high-frequency component index. Provided is a high-frequency emphasis amount control device including: a high-frequency component emphasis ratio calculation unit that calculates a high-frequency component emphasis ratio; and an emphasis amount control unit that supplies to the high-frequency emphasis circuit an emphasis amount control value that controls to increase the amount of emphasis of the high-frequency components of the input image signal by the high-frequency emphasis circuit if the high-frequency component emphasis ratio is smaller than a target emphasis ratio, to decrease the amount of emphasis of the high-frequency components of the input image signal by the high-frequency emphasis circuit if the high-frequency component emphasis ratio is larger than the target emphasis ratio, and to keep the amount of emphasis of the high-frequency components of the input image signal by the high-frequency emphasis circuit unchanged if the high-frequency component emphasis ratio matches the target emphasis ratio. [Effects of the Invention]

[0007] According to the high-frequency emphasis amount control device of the present invention, the high-frequency components of an image can be automatically emphasized with an appropriate amount of emphasis. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is a block diagram showing a high-frequency emphasis amount control device according to a first embodiment; [Figure 2] FIG. 13 is a conceptual diagram showing weighting filters that are multiplied by frequency spectra by the weighting calculation units 12 and 22 in FIGS. 1, 9, and 12. [Figure 3A] FIG. 1 shows a frame displaying a circular zone plate. [Figure 3B] FIG. 3B is a diagram showing a partial region extracted from the center of FIG. 3A. [Figure 3C]3B is a diagram showing a frequency spectrum obtained by performing a two-dimensional discrete Fourier transform on the partial region of FIG. 3B by the discrete Fourier transform unit 11 in FIG. [Figure 3D] 2 is a diagram showing a weighting filter that is multiplied by a frequency spectrum by a weighting calculation unit 12 in FIG. 1. FIG. [Figure 3E] 3C is a diagram showing a frequency spectrum map obtained by weighting calculation unit 12 in FIG. 1 multiplying the frequency spectrum shown in FIG. 3C by the weighting filter shown in FIG. 3D. [Figure 4A] FIG. 1 shows a frame with a reduced circular zone plate placed in the center. [Figure 4B] FIG. 4B is a diagram showing a partial region extracted from the center of FIG. 4A. [Figure 4C] 4B is a diagram showing a frequency spectrum obtained by performing a two-dimensional discrete Fourier transform on the partial region of FIG. 4B by the discrete Fourier transform unit 11 in FIG. [Figure 4D] 2 is a diagram showing a weighting filter that is multiplied by a frequency spectrum by a weighting calculation unit 12 in FIG. 1. FIG. [Figure 4E] 4C by the weighting filter shown in FIG. 4D. FIG. [Figure 5] FIG. 13 is a block diagram showing a schematic configuration example of the high frequency emphasis circuit 30 in FIGS. 1, 9, and 12. [Figure 6] 13 is a characteristic diagram showing a time constant when the high-frequency emphasis circuit 30 in FIG. 1, FIG. 9, and FIG. 12 changes the emphasis amount control value. FIG. [Figure 7] 13 is a characteristic diagram showing a hysteresis characteristic when the high-frequency emphasis circuit 30 in FIG. 1, FIG. 9, and FIG. 12 changes the emphasis amount control value. FIG. [Figure 8] FIG. 1 is a diagram showing an example of a medical image displayed on a medical monitor by a DICOM viewer. [Figure 9] FIG. 10 is a block diagram showing a high-frequency emphasis amount control device according to a second embodiment. [Figure 10A] FIG. 10 is a diagram showing an example of a frequency spectrum map output from the weighting calculation unit 12 in FIG. [Figure 10B] 10 is a diagram showing an example of a frequency spectrum map from which horizontal and vertical frequency components have been removed by the horizontal / vertical frequency component removing unit 14 in FIG. 9. FIG. [Figure 11A] 10 is a diagram showing an example of a frequency spectrum map output from a weighting calculation unit 22 in FIG. 9. FIG. [Figure 11B] 10 is a diagram showing an example of a frequency spectrum map from which horizontal and vertical frequency components have been removed by the horizontal / vertical frequency component removing unit 24 in FIG. 9. FIG. [Figure 12] FIG. 10 is a block diagram showing a high-frequency emphasis amount control device according to a third embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, the high-frequency emphasis amount control device of each embodiment will be described with reference to the accompanying drawings.

[0010] First Embodiment 1 shows a high-frequency enhancement amount control device 101 according to a first embodiment. The high-frequency enhancement amount control device 101 includes a discrete Fourier transform unit 11 (first discrete Fourier transform unit), a weighting calculation unit 12 (first weighting calculation unit), and a sum calculation unit 13 (first sum calculation unit). The high-frequency enhancement amount control device 101 also includes a discrete Fourier transform unit 21 (second discrete Fourier transform unit), a weighting calculation unit 22 (second weighting calculation unit), a sum calculation unit 23 (second sum calculation unit), a high-frequency component enhancement ratio calculation unit 4, and an enhancement amount control unit 5. The high-frequency enhancement amount control device 101 controls the amount of enhancement used when a high-frequency enhancement circuit 30 enhances the high-frequency components of an input image signal.

[0011] The high-frequency emphasis amount control device 101 and the high-frequency emphasis circuit 30 shown in Fig. 1 can be mounted on any image display device. As an example, the image display device is a medical monitor on which a doctor diagnoses a patient by viewing images of the human body captured by an imaging device such as an X-ray device, a computed tomography (CT) device, or a magnetic resonance imaging (MRI) device. The captured images used by a doctor for diagnosis are referred to as medical images.

[0012] Medical images are stored in a PACS (Picture Archiving and Communication System) server, and workstations read the images from the PACS server and display them on a medical monitor. Medical images usually conform to the DICOM (Digital Imaging and Communications in Medicine) standard and are displayed on the medical monitor using software called a DICOM viewer.

[0013] 1, a discrete Fourier transform unit 11 performs a two-dimensional discrete Fourier transform on an input image signal, which is a medical image to be displayed on a medical monitor. A discrete Fourier transform unit 21 performs a two-dimensional discrete Fourier transform on an output image signal output from a high-frequency emphasis circuit 30. Typically, the discrete Fourier transform units 11 and 21 perform the discrete Fourier transform on the input image signal and the output image signal, respectively, using a fast Fourier transform (FFT) algorithm.

[0014] Weighting calculation units 12 and 22 generate frequency spectrum maps for evaluation by multiplying the two-dimensional frequency spectra output from discrete Fourier transform units 11 and 21 by weighting filters that extract high-frequency components. The frequency spectrum output from discrete Fourier transform unit 11 is the first frequency spectrum, and the frequency spectrum output from discrete Fourier transform unit 21 is the second frequency spectrum. The frequency spectrum map generated by weighting calculation unit 12 is the first frequency spectrum map, and the frequency spectrum map generated by weighting calculation unit 22 is the second frequency spectrum map.

[0015] Fig. 2 conceptually shows the weighting filters that are multiplied by the frequency spectra by the weighting calculation units 12 and 22. As shown in Fig. 2, it is preferable to use an inverted cone-shaped filter as the weighting filter, which has a value of 0 at the center of frame F of the input image signal and output image signal and whose value increases with increasing distance from the center.

[0016] When the weighting calculation units 12 and 22 multiply the frequency spectrum by the inverted cone filter shown in FIG. 2, the low-frequency components contained in the frequency spectrum are removed and the high-frequency components are increased, thereby obtaining a frequency spectrum map that has been processed so that the high-frequency components are dominant.

[0017] The frequency spectrum maps generated by the weighting calculation units 12 and 22 will be described with reference to Figures 3A to 3E and Figures 4A to 4E. Figure 3A shows frame F displaying a circular zone plate (CZP) as an example of an image with relatively few high-frequency components. Figure 4A shows frame F in which a circular zone plate is reduced and placed in the center as an example of an image with relatively many high-frequency components. Frame F shown in Figures 3A and 4A is an example of a case where the image has a full HD resolution of 1920 pixels in the horizontal direction and 1080 pixels in the vertical direction.

[0018] Figure 3B shows a partial region extracted from the center of frame F shown in Figure 3A, covering a range of approximately 800 pixels horizontally and 800 pixels vertically. Figure 3C shows a frequency spectrum resulting from a two-dimensional discrete Fourier transform performed by the discrete Fourier transform unit 11 on the partial region of Figure 3B. The center of the frequency spectrum shown in Figure 3C represents low-frequency components, while the periphery represents high-frequency components. In the periphery, the higher the frequency, the larger the value, appearing whitish.

[0019] Fig. 3D shows a weighting filter that the weighting calculation unit 12 multiplies by the frequency spectrum. Fig. 3D corresponds to the weighting filter shown in Fig. 2. In Fig. 3D, the darker the area, the closer the value is to 0, and the lighter the area, the larger the value. Fig. 3E shows a frequency spectrum map obtained by the weighting calculation unit 12 multiplying the frequency spectrum shown in Fig. 3C by the weighting filter shown in Fig. 3D. The low-frequency components in the center of the frequency spectrum have been removed, and the values ​​are smaller and displayed darker.

[0020] Figure 4B shows a partial region extracted from the center of frame F shown in Figure 4A, covering a range of approximately 800 pixels horizontally and 800 pixels vertically. Figure 4C shows a frequency spectrum resulting from a two-dimensional discrete Fourier transform of the partial region of Figure 4B by the discrete Fourier transform unit 11. As can be seen by comparing Figures 3C and 4C, the partial region shown in Figure 4B has more high-frequency components than the partial region shown in Figure 3B, and therefore the frequency spectrum shown in Figure 4C appears whitish with larger values ​​overall. Figure 4D shows the same weighting filter as Figure 3D.

[0021] Fig. 4E shows a frequency spectrum map obtained by multiplying the frequency spectrum shown in Fig. 4C by the weighting filter shown in Fig. 4D by the weighting calculation unit 12. As can be seen from a comparison of Fig. 3E and Fig. 4E, the frequency spectrum map shown in Fig. 4E has larger values ​​and is displayed more whitish than the frequency spectrum map shown in Fig. 3E.

[0022] In this way, the more high-frequency components contained in the input image signal, the larger the value of each pixel in the frequency spectrum map generated by the weighting calculation unit 12. Similarly, the more high-frequency components contained in the output image signal, the larger the value of each pixel in the frequency spectrum map generated by the weighting calculation unit 22.

[0023] Returning to Fig. 1, summation calculation units 13 and 23 respectively calculate the sum of the values ​​of each pixel in the frequency spectrum map output from the weighting calculation units 12 and 22 within frame F. The summation calculation unit 13 is a first summation calculation unit, and the summation calculation unit 23 is a second summation calculation unit. The sum of the values ​​of each pixel in the frequency spectrum map calculated by the summation calculation units 13 and 23 is a high-frequency component index indicating the total amount of high-frequency components in the frequency spectrum map.

[0024] The sum calculation unit 13 functions as a first high-frequency component index generation unit that generates a high-frequency component index (first high-frequency component index) based on the frequency spectrum output from the discrete Fourier transform unit 11. The sum calculation unit 23 functions as a second high-frequency component index generation unit that generates a high-frequency component index (second high-frequency component index) based on the frequency spectrum output from the discrete Fourier transform unit 21.

[0025] Although it is not essential to provide the weighting calculation units 12 and 22, it is preferable to provide them. When the weighting calculation units 12 and 22 are provided, the sum calculation units 13 and 23 generate high-frequency component indices based on frequency spectrum maps processed so that high-frequency components are dominant, making it easier to determine the rate at which the high-frequency emphasis circuit 30 should emphasize the high-frequency components of the input image signal.

[0026] The high-frequency component emphasis ratio calculation unit 4 calculates a high-frequency component emphasis ratio, which is the ratio between the high-frequency component index output from the sum calculation unit 13 and the high-frequency component index output from the sum calculation unit 23. The high-frequency component emphasis ratio calculation unit 4 preferably calculates the high-frequency component emphasis ratio by dividing the high-frequency component index output from the sum calculation unit 23 by the high-frequency component index output from the sum calculation unit 13. The high-frequency component emphasis ratio indicates the proportion of the high-frequency component that has increased as a result of the high-frequency component of the input image signal being emphasized by the high-frequency emphasis circuit 30, relative to the high-frequency component of the input image signal. The high-frequency component emphasis ratio is supplied to the emphasis amount control unit 5.

[0027] A target emphasis ratio set by the user is input to the emphasis amount control unit 5. The target emphasis ratio is, for example, 1.1. A target emphasis ratio of 1.1 means that the high-frequency components of the input image signal are emphasized by 1.1 times by the high-frequency emphasis circuit 30. The emphasis amount control unit 5 generates an emphasis amount control value according to the result of comparison between the high-frequency component emphasis ratio supplied from the high-frequency component emphasis ratio calculation unit 4 and the target emphasis ratio, and supplies the generated value to the high-frequency emphasis circuit 30.

[0028] Taking a target emphasis ratio of 1.1 as an example, if the high-frequency component emphasis ratio is smaller than the target emphasis ratio of 1.1, the emphasis amount control unit 5 supplies an emphasis amount control value to the high-frequency component emphasis circuit 30 to increase the amount of emphasis of the high-frequency component by the high-frequency component emphasis circuit 30. The emphasis amount control value to increase the amount of emphasis of the high-frequency component may be, for example, a positive value that increases as the degree of increase in the emphasis amount increases.

[0029] If the high-frequency component emphasis ratio is greater than the target emphasis ratio of 1.1, the emphasis amount control unit 5 supplies an emphasis amount control value to the high-frequency component emphasis circuit 30 to reduce the amount of emphasis of the high-frequency component by the high-frequency component emphasis circuit 30. The emphasis amount control value to reduce the amount of emphasis of the high-frequency component may be, for example, a negative value that decreases as the degree of reduction in the emphasis amount increases.

[0030] If the high-frequency component emphasis ratio matches the target emphasis ratio of 1.1, the emphasis amount control unit 5 supplies an emphasis amount control value to the high-frequency component emphasis circuit 30, which controls the amount of emphasis of the high-frequency component by the high-frequency component emphasis circuit 30 so as not to change the amount. The emphasis amount control value which controls the amount of emphasis of the high-frequency component so as not to change the amount of emphasis can be set to 0, for example.

[0031] The high-frequency emphasis circuit 30 controls the amount of emphasis of the high-frequency components of the input image signal in accordance with the emphasis amount control value supplied from the emphasis amount control unit 5. The output image signal, in which the high-frequency components have been emphasized by the high-frequency emphasis circuit 30, is input to the discrete Fourier transform unit 21. Therefore, the high-frequency emphasis amount control device 101 includes a feedback control loop that automatically controls the amount of emphasis of the high-frequency components by the high-frequency emphasis circuit 30 to an emphasis amount corresponding to a target emphasis ratio set in the emphasis amount control unit 5 by the user.

[0032] The output image signal, in which the amount of emphasis of the high-frequency components has been automatically controlled by the high-frequency emphasis circuit 30 to an amount of emphasis corresponding to the target emphasis ratio, is supplied to a liquid crystal panel (not shown) and displayed. The medical monitor may be equipped with an organic EL panel instead of a liquid crystal panel, or any other display panel.

[0033] 5 shows a schematic configuration example of the high-frequency emphasis circuit 30. The high-frequency emphasis circuit 30 includes a high-pass filter (hereinafter referred to as HPF) 31, a coefficient multiplier 32, a delay unit 33, and an adder 34. The HPF 31 extracts high-frequency components from an input image signal. The coefficient multiplier 32 multiplies the extracted high-frequency components by a coefficient to emphasize the high-frequency components. The delay unit 33 delays the input image signal by a time equivalent to the processing performed by the HPF 31 and the coefficient multiplier 32. The adder 34 adds the high-frequency components output from the coefficient multiplier 32 to the input image signal delayed by the delay unit 33 to generate an output image signal.

[0034] When the enhancement amount control unit 5 supplies the coefficient multiplier 32 with an enhancement amount control value that controls to increase the enhancement amount of the high-frequency components, the coefficient multiplier 32 increases the value of the coefficient by which the high-frequency components are multiplied. When the enhancement amount control unit 5 supplies the coefficient multiplier 32 with an enhancement amount control value that controls to decrease the enhancement amount of the high-frequency components, the coefficient multiplier 32 decreases the value of the coefficient. When the enhancement amount control unit 5 supplies the coefficient multiplier 32 with an enhancement amount control value that controls to not change the enhancement amount of the high-frequency components, the coefficient multiplier 32 does not change the value of the coefficient.

[0035] 5 is merely an example, and the high-frequency emphasis circuit 30 may have any configuration as long as it is configured to emphasize the high-frequency components of the input image signal. The high-frequency emphasis circuit 30 may be a circuit known as an edge emphasis circuit, a contour correction circuit, or an enhancer.

[0036] According to the high-frequency enhancement amount control device 101 described above, the high-frequency components of an image can be automatically enhanced with an appropriate enhancement amount.

[0037] In the high-frequency emphasis amount control device 101, when the emphasis amount control unit 5 changes the emphasis amount control value in a short period of time, a phenomenon called hunting may occur, in which the emphasis amount control value frequently fluctuates around a target emphasis amount control value (target emphasis amount control value). To avoid hunting, it is preferable that the emphasis amount control unit 5 is configured to output the target emphasis amount control value over a predetermined period of time by providing a time constant to a feedback control loop that generates and outputs the target emphasis amount control value.

[0038] FIG. 6 shows how the enhancement amount control value changes when the enhancement amount control unit 5 changes the enhancement amount control value to a predetermined target enhancement amount control value, starting with an enhancement amount control value of 0, as shown by the solid line. If the time constant is set to a small value, the target enhancement amount control value is reached in a short time. If the time constant is set to a medium value, as shown by the dashed line, the target enhancement amount control value is reached in a longer time than when the time constant is set to a small value. If the time constant is set to a large value, as shown by the dashed line, the target enhancement amount control value is reached in a longer time. The enhancement amount control unit 5 sets a predetermined time constant to change the enhancement amount control value.

[0039] The enhancement amount control unit 5 may provide a hysteresis characteristic to the feedback control loop that generates and outputs the enhancement amount control value. Fig. 7 shows an example of the hysteresis characteristic. By providing the feedback control loop with a hysteresis characteristic, it is possible to prevent hunting from occurring.

[0040] 7, when the high-frequency component emphasis ratio initially increases from 0 in a positive direction, the emphasis amount control unit 5 gradually increases the emphasis amount control value, as indicated by the dashed line. If the high-frequency component emphasis ratio decreases after the emphasis amount control value has reached its maximum value, the emphasis amount control unit 5 maintains the maximum emphasis amount control value for a while, and as the high-frequency component emphasis ratio further decreases, gradually decreases the emphasis amount control value. If the high-frequency component emphasis ratio increases after the emphasis amount control value has reached its minimum value, the emphasis amount control unit 5 maintains the minimum emphasis amount control value for a while, and as the high-frequency component emphasis ratio further increases, gradually increases the emphasis amount control value.

[0041] In this way, in order to avoid the occurrence of hunting, it is preferable that the enhancement amount control unit 5 changes the enhancement amount control value with a predetermined time constant or changes the enhancement amount control value in accordance with a predetermined hysteresis characteristic.

[0042] Second Embodiment FIG. 8 shows an example of a medical image displayed by a DICOM viewer on a medical monitor. When a medical image is displayed on a medical monitor using a DICOM viewer, a graphical user interface image (hereinafter referred to as a GUI image) such as a window frame and a menu is also displayed on the medical monitor. Because the GUI image is a geometrical figure, it contains sharp edges and many high-frequency components. Therefore, if a medical image such as that shown in FIG. 8 is used as an input image signal and the high-frequency components of the input image signal are enhanced by the high-frequency enhancement circuit 30 to generate an output image signal, the frequency spectrum map will contain many high-frequency components originating from the GUI image.

[0043] Even if the original medical image does not contain significant high-frequency components, if the high-frequency component index becomes a large value, the high-frequency component emphasis ratio may not represent the ratio of the high-frequency components of the original medical image, and the emphasis amount control value may not be an appropriate value. The high-frequency emphasis amount control device 102 of the second embodiment shown in Figure 9 is configured to generate an appropriate emphasis amount control value even if the input image signal contains high-frequency components caused by a GUI image. In Figure 9, the same parts as in Figure 1 are assigned the same reference numerals, and their description may be omitted.

[0044] GUI images are often composed of a combination of horizontal and vertical lines, but medical images are natural images and contain almost no horizontal or vertical lines. Therefore, the horizontal and vertical frequency components contained in the frequency spectrum map are mostly unnecessary high-frequency components originating from the GUI image.

[0045] 9, horizontal and vertical frequency component removal units 14 and 24 (first and second horizontal and vertical frequency component removal units) respectively remove horizontal and vertical frequency components from the frequency spectrum maps output from the weighting calculation units 12 and 22. The horizontal and vertical frequency component removal units 14 and 24 multiply the frequency spectrum maps by cross-shaped filters that remove horizontal and vertical frequency components.

[0046] FIG. 10A shows a frequency spectrum map output from the weighting calculation unit 12. FIG. 10B shows a frequency spectrum map output from the horizontal and vertical frequency component removal unit 14. The horizontal and vertical frequency components have been removed by the cross-shaped filter of the horizontal and vertical frequency component removal unit 14. FIG. 11A shows a frequency spectrum map output from the weighting calculation unit 22. FIG. 11B shows a frequency spectrum map output from the horizontal and vertical frequency component removal unit 24. The horizontal and vertical frequency components have been removed by the cross-shaped filter of the horizontal and vertical frequency component removal unit 24.

[0047] 9, the horizontal and vertical frequency component removal units 14 and 24 are provided after the weighting calculation units 12 and 22, but they may also be provided before the weighting calculation units 12 and 22. Furthermore, the functions of the horizontal and vertical frequency component removal units 14 and 24 may be provided in the weighting calculation units 12 and 22. Specifically, the weighting calculation units 12 and 22 may multiply the frequency spectra output from the discrete Fourier transform units 11 and 21 by a filter that is obtained by multiplying the inverted cone filter and the cross filter shown in FIG.

[0048] The high-frequency enhancement amount control device 102 described above can automatically enhance the high-frequency components of an image with an appropriate enhancement amount. When an input image signal includes a geometric figure including horizontal and vertical lines, such as a GUI image, other than an image whose high-frequency components should be enhanced, the high-frequency enhancement amount control device 102 can remove unnecessary high-frequency components caused by the geometric figure and enhance the high-frequency components of the image with an appropriate enhancement amount.

[0049] In the high-frequency emphasis amount control device 102 as well, the emphasis amount control section 5 preferably changes the emphasis amount control value with a predetermined time constant, and preferably changes the emphasis amount control value in accordance with a predetermined hysteresis characteristic.

[0050] <Third embodiment> The high-frequency emphasis amount control device 103 of the third embodiment shown in Fig. 12 is configured to generate an appropriate enhancement amount control value even if the input image signal contains high-frequency components caused by a GUI image, using a method different from that of the high-frequency emphasis amount control device 102 of the second embodiment. In Fig. 12, the same parts as in Fig. 1 are assigned the same reference numerals, and their description may be omitted.

[0051] As shown in Fig. 8, GUI images are often placed at the top and bottom or left and right edges of a frame, and medical images are often placed in the center of the frame. Therefore, as shown in Fig. 12, the high-frequency emphasis amount control device 103 includes trimming units 10 and 20 (first and second trimming units) before the discrete Fourier transform units 11 and 21.

[0052] The trimming unit 10 trims each frame of the input image signal by removing the top, bottom, left, and right edges and cropping out the center. The trimming unit 20 trims each frame of the output image signal by removing the top, bottom, left, and right edges and cropping out the center. The trimming units 10 and 20 trim the same area for the input image signal and the output image signal, respectively. The ranges that the trimming units 10 and 20 trim as the center in the horizontal and vertical directions from the center of each frame can be set appropriately.

[0053] The discrete Fourier transform units 11 and 21 perform a two-dimensional discrete Fourier transform on the image signals of the central parts of each frame trimmed by the trimming units 10 and 20. Therefore, the frequency spectra output from the discrete Fourier transform units 11 and 21 contain almost no unnecessary high-frequency components caused by the GUI image (geometrical figure).

[0054] The high-frequency enhancement amount control device 103 can automatically enhance the high-frequency components of an image with an appropriate enhancement amount, remove unnecessary high-frequency components caused by geometric shapes, and enhance the high-frequency components of the image with an appropriate enhancement amount.

[0055] In the high-frequency emphasis amount control device 103 as well, the emphasis amount control section 5 preferably changes the emphasis amount control value with a predetermined time constant, and preferably changes the emphasis amount control value in accordance with a predetermined hysteresis characteristic.

[0056] The present invention is not limited to the first to third embodiments described above, and various modifications are possible without departing from the gist of the present invention. [Explanation of symbols]

[0057] 4 High-frequency component emphasis ratio calculation section 5 Enhancement amount control section 10,20 Trimming section 11,21 Discrete Fourier Transform Section 12,22 Weighting calculation unit 13,23 Sum calculation section 14,24 Horizontal and vertical frequency component removal section 30 High-frequency emphasis circuit 101, 102, 103 High frequency emphasis amount control device

Claims

1. a first discrete Fourier transform unit that performs a two-dimensional discrete Fourier transform on the input image signal to generate a first frequency spectrum; a second discrete Fourier transform unit that performs a two-dimensional discrete Fourier transform on the output image signal output from the high-frequency emphasis circuit to generate a second frequency spectrum; a first high frequency component index generating unit that generates a first high frequency component index indicating a total amount of high frequency components based on the first frequency spectrum; a second high frequency component index generating unit that generates a second high frequency component index indicating a total amount of high frequency components based on the second frequency spectrum; a high-frequency component emphasis ratio calculation unit that calculates a high-frequency component emphasis ratio, which is a ratio between the first high-frequency component index and the second high-frequency component index; an enhancement amount control unit that supplies to the high-frequency emphasis circuit an enhancement amount control value that controls the high-frequency component enhancement circuit to increase an enhancement amount of the high-frequency component of the input image signal if the high-frequency component enhancement ratio is smaller than a target enhancement ratio, to decrease an enhancement amount of the high-frequency component of the input image signal if the high-frequency component enhancement ratio is larger than the target enhancement ratio, and to keep the enhancement amount of the high-frequency component of the input image signal by the high-frequency emphasis circuit unchanged if the high-frequency component enhancement ratio matches the target enhancement ratio; A high-frequency emphasis amount control device comprising:

2. a first weighting calculation unit that processes the first frequency spectrum to remove low-frequency components and increase high-frequency components included in the first frequency spectrum to generate a first frequency spectrum map; a second weighting calculation unit that processes the second frequency spectrum to remove low-frequency components and increase high-frequency components included in the second frequency spectrum to generate a second frequency spectrum map; Furthermore, the first high frequency component index generation unit is a first sum calculation unit that calculates, within each frame, a sum of values ​​of each pixel in the first frequency spectrum map as the first high frequency component index; The second high frequency component index generating unit is a second sum calculation unit that calculates, within each frame, a sum of values ​​of each pixel of the second frequency spectrum map as the second high frequency component index. The high-frequency emphasis amount control device according to claim 1 .

3. a first horizontal and vertical frequency component removal unit that removes horizontal and vertical frequency components from the first frequency spectrum map; a second horizontal and vertical frequency component removal unit that removes horizontal and vertical frequency components from the second frequency spectrum map; The high-frequency emphasis amount control device according to claim 2 , further comprising:

4. a first trimming unit that trims the input image signal so as to cut out a central portion of each frame; a second trimming unit that trims the output image signal by cutting out a central portion of each frame; The high-frequency emphasis amount control device according to claim 1 or 2, further comprising:

5. 5. The high-frequency emphasis amount control device according to claim 1, wherein the emphasis amount control unit changes the emphasis amount control value with a predetermined time constant or changes the emphasis amount control value in accordance with a predetermined hysteresis characteristic.

Citation Information

Patent Citations

  • Image processing method for microscope

    JP1997097332A

  • Image processing method and image processor thereof

    JP2000004398A

  • Image processor, image processing method and medium

    JP2001016480A

  • Image photographing apparatus

    JP2006050494A

  • Image emphasizing apparatus, its method and camera

    JP2007336064A