JP2011139127A - Image processor and image processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processor which appropriately controls an amount of contour enhancement. <P>SOLUTION: In the image processor, luminance signals respectively corresponding to a plurality of pixels constituting an image and a contour signal indicating luminance of a contour portion included in the image are added to generate a contour enhancement signal, and low frequency signals are extracted from the luminance signals, and a maximum value and a minimum value of the luminance signals and those of the low frequency signals are detected with respect to a plurality of pixels located around a target pixel, and a low frequency edge level is calculated which takes a lower value for a larger difference absolute value between the maximum value of the low frequency signals and the maximum value of the luminance signals or between the minimum value of the low frequency signals and the minimum value of the luminance signals and takes a higher value for a smaller difference absolute value between them, and a Shoot amount of the target pixel calculated on the basis of the contour enhancement signal and the maximum value or the minimum value of the luminance signals is more reduced toward zero for a lower low frequency edge level, and the contour enhancement signal is corrected on the basis of the reduced Shoot amount. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to an image processing apparatus and an image processing method for realizing appropriate edge enhancement processing on an image signal, and more particularly to an image processing apparatus and an image processing method for suppressing Overshoot and Undershot that occur in an edge portion in a low frequency range. .

  When it is desired to increase the sharpness of an image signal, an edge enhancement process is generally performed on the image signal. Outline enhancement is usually realized by inputting an image luminance signal to an outline extraction filter, extracting only the edge portion of the image signal, and adding the luminance signal of the edge portion to the original luminance signal. However, when such a contour emphasizing method is used, Overshoot and Undershot are generated in the edge portion of the low frequency region included in the image signal, and the edge portion is generated in the Edge portion. In particular, when the number of Taps of the contour extraction filter is large, the border width generated by contour enhancement is widened, resulting in a very unnatural image.

  In order to solve such a problem, for example, in Patent Document 1 below, the degree of waveform complexity (flatness) of an image signal is determined, and Overshoot and Undershot are suppressed in a portion where the degree of waveform complexity is low. A method for correcting a contour emphasis signal is disclosed. In addition, in Patent Document 2 below, in addition to the process of correcting the edge enhancement signal based on the flatness of the waveform of the image signal as in Patent Document 1, the average signal of the pixels located around the edge portion is disclosed. A correction method for suppressing Overshot and Undershot based on the difference between the two is disclosed.

JP 2004-266775 A JP 2009-71714 A

  Certainly, by applying the correction method described in Patent Document 1, it is possible to suppress the occurrence of bordering at the edge portion of the low frequency region caused by the contour enhancement. However, since the correction method described in Patent Document 1 discriminates the edge portion of the low frequency region based on the complexity of the waveform of the image signal, the edge portion is a frequency at which overshoot and undershoot are to be suppressed. Appropriate judgment is not made as to whether it belongs to the belt. As a result, when the technique of Patent Document 1 is applied, Overshoot and Undershot that should not be suppressed are suppressed, and an appropriate contour emphasis effect cannot be obtained.

  Further, the correction method described in Patent Document 2 also determines the edge portion of the low frequency region based on the flatness of the waveform of the image signal, similarly to the correction method described in Patent Document 1. Therefore, as in the case of the above-mentioned Patent Document 1, Overshoot and Undershot that should not be suppressed are suppressed, and an appropriate contour emphasis effect cannot be obtained. That is, when the techniques described in Patent Documents 1 and 2 are applied, as shown in FIG. 1, it is possible to suppress Overshot or Undershot that should be the target of suppression, but want to retain Overshot and Undershot by edge enhancement. The edge portion of the frequency range is also suppressed if the degree of complexity in the vicinity is low, and an appropriate suppression effect cannot be obtained.

  Therefore, the present invention has been made in view of the above problems, and an object of the present invention is to appropriately set the low frequency range Edge of the frequency to be suppressed for Overshoot and Undershot generated in the Edge portion of the low frequency range. It is an object of the present invention to provide a new and improved image processing apparatus and an image processing method capable of obtaining a suitable contour-enhanced image by determining and suppressing the above.

  In order to solve the above-described problem, according to an aspect of the present invention, a luminance signal corresponding to each of a plurality of pixels forming an image and a contour signal indicating the luminance of a contour portion included in the image are added. Contour emphasis means for generating an edge emphasis signal, low frequency signal extraction means for extracting a low frequency signal that is a low frequency band component of the luminance signal, and a plurality of pixels located around the pixel of interest, A maximum / minimum value detecting means for detecting a maximum value and a minimum value, and a maximum value and a minimum value of the low-frequency signal; an absolute value of a difference between the maximum value of the low-frequency signal and the maximum value of the luminance signal; Alternatively, the low-frequency edge degree for calculating a low-frequency edge degree that becomes a lower value as the difference absolute value between the minimum value of the low-frequency signal and the minimum value of the luminance signal is larger and becomes higher as the difference absolute value is smaller. A calculation means; The shot amount of the target pixel calculated based on the contour enhancement signal and the maximum value or the minimum value of the luminance signal is suppressed so as to approach 0 as the low-frequency edge degree is small, and is suppressed. There is provided an image processing apparatus comprising suppression shot amount calculation means for calculating the edge enhancement signal and edge enhancement signal correction means for correcting the edge enhancement signal based on the suppression shot amount calculated by the suppression shot amount calculation means.

  With this configuration, the Overshoot and Undershot of the Edge portion generated by edge enhancement are suppressed from the image signal and the low-frequency signal detected from the image signal using a low-pass filter having a specific low-frequency band as a pass band. Since the degree of the low frequency range edge of the frequency to be determined is determined and the appropriate amount of shot is suppressed according to the degree, it is possible to obtain a more suitable contour-enhanced image. That is, it is possible to appropriately suppress Overshoot and Undershot for the low frequency range Edge to be suppressed without impairing the enhancement effect for the Edge portion to be enhanced. As a result, it is possible to obtain an image signal in which the low-frequency edge portion to be emphasized is emphasized while avoiding unnatural borders due to edge enhancement.

  The low frequency edge degree calculating means may calculate the low frequency edge degree based on an absolute value of a difference between the maximum value of the low frequency signal and the maximum value of the luminance signal when the Shot amount is larger than a predetermined value. And when the amount of shot is smaller than a predetermined value, the low frequency edge degree is calculated based on an absolute difference value between the minimum value of the low frequency signal and the minimum value of the luminance signal. Also good. With this configuration, a more appropriate low frequency range Edge degree is calculated.

  Further, the Shot amount calculation means calculates the Shot amount of the target pixel based on the contour emphasis signal and the maximum value or minimum value of the luminance signal, and when the value of the contour signal is positive. Calculates a shot amount of the target pixel based on the contour emphasis signal and the maximum value of the luminance signal. When the value of the contour signal is negative, the contour emphasis signal and the minimum value of the luminance signal are calculated. Based on this, it may be configured to calculate the Shot amount of the target pixel. With this configuration, a more appropriate amount of shot is calculated.

  Further, the suppression Shot amount calculation means calculates the suppression Shot amount so that the smaller the low frequency edge degree is, the closer to 0, and the higher the low frequency edge degree is, the closer the Shot amount is. It may be configured. With such a configuration, a more appropriately suppressed amount of shot (inhibited shot amount) is calculated.

  In addition, the maximum / minimum value detecting unit may detect the maximum value and the minimum value of the luminance signal or the low frequency signal for a plurality of pixels located within a predetermined range with reference to a target pixel as a correction amount of a shot amount. When detecting the maximum value and the minimum value, the predetermined range for determining the plurality of pixels to be detected may be arbitrarily changed.

  The image processing apparatus may further include: a luminance signal extracting unit that extracts a luminance signal from an image signal of an image including a plurality of pixels; and a contour signal extracting unit that extracts a contour signal indicating an image contour from the luminance signal. . With this configuration, it is possible to obtain an image signal that has been subjected to appropriate edge enhancement processing with respect to the input of the image signal.

  In order to solve the above problem, according to another aspect of the present invention, a luminance signal corresponding to each of a plurality of pixels constituting an image and a contour signal indicating the luminance of a contour portion included in the image are obtained. A contour emphasizing step for generating a contour emphasizing signal by adding, a low frequency signal extracting step for extracting a low frequency signal that is a low frequency band component of the luminance signal, and a plurality of pixels located around the pixel of interest, Maximum / minimum value detection step for detecting the maximum value and minimum value of the luminance signal and the maximum value and minimum value of the low-frequency signal, and the difference between the maximum value of the low-frequency signal and the maximum value of the luminance signal The lower the absolute value, or the lower the absolute value of the difference between the minimum value of the low frequency signal and the minimum value of the luminance signal, the lower the value. Frequency The shot amount of the target pixel calculated based on the degree of degree calculation step, the contour emphasis signal, and the maximum value or the minimum value of the luminance signal is made closer to 0 as the low frequency edge degree is smaller. A suppression shot amount calculation step for suppressing and calculating a suppression shot amount; and a contour enhancement signal correction step for correcting the contour enhancement signal based on the suppression shot amount calculated in the suppression shot amount calculation step. An image processing method is provided.

  With this configuration, the Overshoot and Undershot of the Edge portion generated by edge enhancement are suppressed from the image signal and the low-frequency signal detected from the image signal using a low-pass filter having a specific low-frequency band as a pass band. Since the degree of the low frequency range Edge of the frequency to be determined is determined and appropriate suppression according to the degree is performed, it is possible to obtain a more suitable contour-enhanced image. That is, it is possible to appropriately suppress Overshoot and Undershot for the low frequency range Edge to be suppressed without impairing the enhancement effect for the Edge portion to be enhanced. As a result, it is possible to obtain an image signal in which the low-frequency edge portion to be emphasized is emphasized while avoiding unnatural borders due to edge enhancement.

  In order to solve the above problems, according to another aspect of the present invention, there is provided a program for causing a computer to realize the functions of the units included in the image processing apparatus. Furthermore, in order to solve the above problems, according to another aspect of the present invention, a computer-readable recording medium on which the program is recorded is provided. And in order to solve the said subject, according to another viewpoint of this invention, the imaging device and display apparatus which mount said image processing apparatus are provided.

  As described above, according to the present invention, a suitable contour-enhanced image can be obtained by appropriately determining and suppressing the low frequency range edge of the frequency to be suppressed with respect to the overshoot and the undershot generated in the edge portion of the low frequency range. It becomes possible to obtain.

It is explanatory drawing which shows schematically the suppression method of a general outline emphasis signal. It is explanatory drawing which shows roughly the suppression method of the outline emphasis signal which concerns on one Embodiment of this invention. FIG. 3 is an explanatory diagram illustrating a configuration example of an imaging apparatus according to the embodiment. It is explanatory drawing which shows the function structural example of the outline emphasis processing part (image processing apparatus) which concerns on the same embodiment. It is explanatory drawing which shows the expression method of the attention pixel which concerns on the embodiment. It is explanatory drawing which shows an example of the range of the surrounding pixel considered when calculating the maximum value / minimum value regarding the attention pixel which concerns on the embodiment. It is explanatory drawing which shows an example of the range of the surrounding pixel considered when calculating the maximum value / minimum value regarding the attention pixel which concerns on the embodiment. It is explanatory drawing which shows an example of the determination method of Overshot / Undershot which concerns on the embodiment. It is explanatory drawing which shows the structural example of the low frequency range Edge degree which concerns on the same embodiment. It is explanatory drawing which shows the structural example of the low frequency range Edge degree which concerns on the same embodiment. It is explanatory drawing which shows the example of a characteristic of the Shot suppression coefficient k which concerns on the same embodiment. It is explanatory drawing which shows the structural example of the correction | amendment outline emphasis signal which concerns on the same embodiment. It is explanatory drawing which shows the structural example of the correction | amendment outline emphasis signal which concerns on the same embodiment. It is explanatory drawing which shows the structural example of the correction | amendment outline emphasis signal which concerns on the same embodiment. It is explanatory drawing which shows the flow of the outline emphasis process which concerns on the same embodiment. It is explanatory drawing which shows the flow of the outline emphasis process which concerns on the same embodiment.

  Exemplary embodiments of the present invention will be described below in detail with reference to the accompanying drawings. In addition, in this specification and drawing, about the component which has the substantially same function structure, duplication description is abbreviate | omitted by attaching | subjecting the same code | symbol.

[About the flow of explanation]
Here, the flow of explanation regarding the embodiment of the present invention described below will be briefly described. First, an outline of a method for suppressing an edge enhancement signal according to the present embodiment will be described with reference to FIGS. 1 and 2. Next, referring to FIG. 3, the configuration of the imaging device 10 is illustrated as an example of a device to which the contour enhancement signal suppression method according to the present embodiment can be applied. Next, the functional configuration of the contour emphasis processing unit 42 that can execute the contour emphasis processing to which the method of suppressing the contour emphasis signal according to the present embodiment is applied will be described in detail with reference to FIGS. Next, the flow of the contour enhancement process according to the present embodiment will be described with reference to FIGS. 15 and 16.

<Embodiment>
Hereinafter, an embodiment of the present invention will be described. The present embodiment relates to an image processing apparatus and an image processing method for realizing appropriate edge enhancement processing for an image signal, and in particular, an image processing apparatus and an image processing method for suppressing Overshoot and Undershot that occur in an edge portion in a low frequency range. About.

[1: Overview]
First, an outline of an image signal contour enhancement method according to the present embodiment will be described with reference to FIGS. 1 and 2. FIG. 1 is an explanatory view schematically showing a conventional contour enhancement method. FIG. 2 is an explanatory diagram schematically showing the contour emphasizing method according to the present embodiment.

  1 and 2 show the signal intensity distributions of the luminance signal and the contour enhancement signal, respectively. Note that the luminance signal is a signal representing the luminance at each pixel of the image signal composed of a plurality of pixels. The luminance enhancement signal is a luminance signal after contour enhancement. However, the edge emphasis signal in FIG. 1 (right figure) is an edge emphasis signal in which undershoot and overshoot are suppressed by the conventional edge emphasis method. On the other hand, the edge emphasis signal in FIG. 2 (right figure) is an edge emphasis signal in which undershoot and overshoot are suppressed by the edge emphasis method according to the present embodiment.

  As described in Patent Documents 1 and 2, when the conventional edge enhancement method suppresses Undershot or Overshoot, the suppression target is determined based on the flatness of the waveform of the image signal. Therefore, as shown in FIG. 1, although the Undershot and Overshot (Undershot in the example of FIG. 1) to be suppressed are effectively suppressed, the Undershot and Overshot (Overshot in the example of FIG. 1) should not be suppressed. There is a problem of being suppressed.

  On the other hand, as will be described later, the edge emphasis method according to the present embodiment determines the suppression target based on the degree of the low frequency range edge when suppressing the Undershot and Overshot. Therefore, as shown in FIG. 2, the Undershot and Overshot that should be suppressed (Undershot in the example of FIG. 2) are effectively suppressed, but the Undershot and Overshot that are not to be suppressed (Overshot in the example of FIG. 2) are suppressed. Therefore, it is possible to obtain an appropriate edge enhancement signal.

  As described above, the contour emphasizing method according to the present embodiment is a technique for preventing undershoot and overshot that should not be originally suppressed when suppressing undershot and overshot so that an unnatural border is not generated. About.

  Hereinafter, the contour enhancement method according to the present embodiment will be described in detail with a specific example. In this article, the configuration of the imaging apparatus 10 is illustrated as an application example of the contour enhancement method according to the present embodiment. However, the application range of the contour enhancement method according to the present embodiment is not limited to the imaging apparatus 10. For example, the present invention can be applied to image display on a display device mounted on an electronic device such as a personal computer, a mobile phone, a portable information terminal, a car navigation system, various information home appliances, or a projector connected to the electronic device. . Furthermore, the present invention can be applied to image display for a television receiver. As described above, it should be noted that the application target of the contour enhancement method according to the present embodiment is various.

[2: Configuration of the imaging apparatus 10]
Hereinafter, the configuration of the imaging apparatus 10 according to the present embodiment will be described.

(2-1: Overall Configuration of Imaging Device 10)
First, the overall configuration of the imaging apparatus 10 according to the present embodiment will be described with reference to FIG. FIG. 3 is an explanatory diagram illustrating the overall configuration of the imaging apparatus 10 according to the present embodiment.

  As illustrated in FIG. 3, the imaging device 10 includes a lens 12, a diaphragm 14, a shutter 16, an imaging device 18, an AFE circuit 20 (AFE: Analog Front End), a preprocessing unit 22, and a postprocessing unit. 24, a data compression unit 26, a memory card interface 28, and a memory card 30.

  The post-processing unit 24 includes a white balance processing unit 32, a demosaic processing unit 34, a noise reduction processing unit 36, a gamma correction processing unit 38, a color correction processing unit 40, and an edge enhancement processing unit 42. Is done. As will be described later, the contour enhancement method according to the present embodiment is provided as a function of the contour enhancement processing unit 42.

  The image sensor 18 is, for example, a CCD (Charge Coupled Device), a CMOS (Complementary Metal Oxide Semiconductor), or the like. Further, the memory card 30 may be a recording unit built in the imaging apparatus 10 or an external recording unit that is detachable.

  The lens 12 constitutes an optical system and condenses the light reflected by the subject on the image sensor 18. However, light incident through the lens 12 is narrowed down by the diaphragm 14. Further, the image sensor 18 receives light only while the shutter 16 is open. The imaging element 18 receives light collected by the lens 12 by a plurality of light receiving elements (for example, a plurality of sets of semiconductor elements having red, blue, and green light receiving sensitivities) and receives the light. The converted light is photoelectrically converted. Further, the image sensor 18 outputs an electrical signal having an intensity corresponding to the intensity of light received by each light receiving element. The electric signal output from each light receiving element is input to the AFE circuit 20, and an image signal is formed based on the intensity of the electric signal.

  The image signal formed by the AFE circuit 20 is input to the preprocessing unit 22. The preprocessing unit 22 performs preprocessing such as OB (Optical Black) processing and defect pixel correction on the image signal input from the AFE circuit 20. The OB process is a process for adjusting the black level of the image signal. The image signal preprocessed by the preprocessing unit 22 is input to the postprocessing unit 24. The post-processing unit 24 performs digital image processing such as white balance adjustment processing, demosaic processing, noise removal processing, gamma correction processing, color correction processing, and edge enhancement processing on the image signal input from the pre-processing unit 22. .

  The white balance adjustment process is executed by the white balance processing unit 32. The demosaic process is executed by the demosaic processing unit 34. The noise reduction processing is executed by the noise reduction processing unit 36. The gamma correction processing is executed by the gamma correction processing unit 38. The color correction process is executed by the color correction processing unit 40. The contour emphasis process is executed by an edge emphasis processing unit 42 described later. Note that the demosaic process is a process (so-called demosaicing) in which, for example, a pixel for which color information is insufficient, the color information of the pixel is complemented based on the information of surrounding pixels.

  The image signal that has been subjected to the digital image processing by the post-processing unit 24 is input to the data compression unit 26. The data compression unit 26 converts the image signal into encoded data such as BMP, GIF, JPEG, PICT, or PING and outputs the encoded data. The encoded data output by the data compression unit 26 is recorded on the memory card 30 via the memory card interface 28. Instead of the memory card 30, a recording medium such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive) may be used.

  The overall configuration of the imaging device 10 has been described above. As described above, the contour emphasis method according to the present embodiment is realized as a function of the contour emphasis processing unit 42. Therefore, in the following, the functional configuration of the contour enhancement processing unit 42 will be described in more detail.

(2-2: Details of Outline Enhancement Processing Unit 42)
Hereinafter, the functional configuration of the contour enhancement processing unit 42 to which the contour enhancement method according to the present embodiment is applied will be described in more detail with reference to FIG. FIG. 4 is an explanatory diagram showing a detailed functional configuration of the contour enhancement processing unit 42 to which the contour enhancement method according to the present embodiment is applied. Note that the contour enhancement processing unit 42 functions as an image processing device that executes the contour enhancement processing according to the present embodiment.

  As shown in FIG. 4, the contour enhancement processing unit 42 includes a luminance signal extraction unit 422, a contour signal extraction unit 424, a contour enhancement signal calculation unit 426, a luminance signal maximum value detection unit 428, and a luminance signal minimum value detection. Means 430 and Shot amount calculation means 432 are provided. Further, the contour enhancement processing unit 42 includes a low frequency signal extraction unit 434, a low frequency signal maximum value detection unit 436, a low frequency signal minimum value detection unit 438, a low frequency range Edge degree calculation unit 440, and a shot amount suppression. Means 442 and corrected outline emphasis signal calculating means 444.

In the following description, the following expressions are used.
Y: Luminance signal E: Contour signal YEn: Contour emphasis signal YnMax: Luminance signal maximum value YnMin: Luminance signal minimum value L: Low frequency signal LnMax: Low frequency signal maximum value LnMin: Low frequency signal minimum value α: Low frequency range Edge Degree Sn: Shot amount SSn: Suppressed shot amount CYEn: Correction outline emphasis signal

  When the contour enhancement process is started, the image signal to be processed is input to the luminance signal extraction unit 422. The luminance signal extraction unit 422 extracts a luminance signal Y indicating the luminance for each pixel from the input image signal. The luminance signal Y extracted by the luminance signal extraction unit 422 is input to the contour signal extraction unit 424, the luminance signal maximum value detection unit 428, the luminance signal minimum value detection unit 430, and the low frequency signal extraction unit 434.

(Calculation of outline emphasis signal Yn)
The contour signal extraction unit 424 generates a contour signal E from the luminance signal Y input from the luminance signal extraction unit 422 based on the following formula (1). However, in the following equation (1), E (i, j) represents the value of the contour signal E corresponding to the pixel at the position (i, j). Y (i + k, j + l) indicates the value of the luminance signal Y corresponding to the pixel at the position (i + k, j + l). W1 (k, l) is a weighting coefficient of the spatial filter (hereinafter, contour extraction filter).

…（１）

... (1)

  Here, the position (i, j) is the target position, and as shown in FIG. 5, the value of the luminance signal Y at the pixel at the target position is the luminance signal Yn, and the value of the contour signal E at the pixel at the target position is the contour signal En. Is written. In the example of FIG. 5, the position (2, 2) is the target position. Note that the attention position here refers to the position of a pixel to be processed in the contour enhancement process.

  The contour signal E generated by the contour signal extraction unit 424 is input to the contour enhancement signal calculation unit 426. Further, the luminance signal Y is also input to the contour enhancement signal calculation means 426. Based on the input contour signal E and luminance signal Y, the contour enhancement signal calculation means 426 adds the contour signal En and the luminance signal Yn as shown in the following equation (2) to generate the contour enhancement signal YEn. The contour enhancement signal YEn generated by the contour enhancement signal calculation unit 426 is input to the Shot amount calculation unit 432 and the corrected contour enhancement signal calculation unit 444.

…（２）

... (2)

(Calculation of luminance signal maximum value YnMax and luminance signal minimum value YnMin)
The luminance signal Y extracted by the luminance signal extraction unit 422 is also input to the luminance signal maximum value detection unit 428 and the luminance signal minimum value detection unit 430. The luminance signal maximum value detecting means 428 detects a maximum value YnMax (luminance signal maximum value YnMax) of a luminance signal of a pixel located around the target position (within a predetermined range) from the input luminance signal Y. However, the predetermined range setting method can be arbitrarily changed.

  For example, when the predetermined range is set to a one-dimensional range, the luminance signal maximum value detecting unit 428 determines the luminance signal maximum value YnMax from the luminance signal Y of the pixels in the one-dimensional range as shown in FIG. Is detected. Further, when the predetermined range is set to a two-dimensional range, the luminance signal maximum value detecting unit 428, as shown in FIG. 7, determines the luminance signal maximum value YnMax from the luminance signal Y of the pixels within the two-dimensional range. Is detected. The luminance signal maximum value YnMax detected by the luminance signal maximum value detecting unit 428 in this way is input to the Shot amount calculating unit 432, the low frequency range edge degree calculating unit 440, and the corrected contour enhancement signal calculating unit 444.

  Also, the luminance signal minimum value detecting means 430 is the luminance signal minimum value YnMin (luminance signal minimum) of the pixels located around the attention position (within a predetermined range) among the luminance signals Y input from the luminance signal extraction means 422. The value YnMin) is detected. However, the predetermined range setting method can be arbitrarily changed.

  For example, when the predetermined range is set to a one-dimensional range, the luminance signal minimum value detecting unit 430, as shown in FIG. 6, determines the luminance signal minimum value YnMin from the luminance signal Y of the pixels in the one-dimensional range. Is detected. When the predetermined range is set to a two-dimensional range, the luminance signal minimum value detecting unit 430, as shown in FIG. 7, determines the luminance signal minimum value YnMin from the luminance signal Y of the pixels in the two-dimensional range. Is detected. The luminance signal minimum value YnMin detected by the luminance signal minimum value detecting unit 430 in this way is input to the Shot amount calculating unit 432, the low frequency range Edge degree calculating unit 440, and the corrected contour enhancement signal calculating unit 444.

(Calculation of Shot amount Sn)
As described above, the edge amount calculation unit 432 receives the edge enhancement signal YEn, the luminance signal maximum value YnMax, and the luminance signal minimum value YnMin. The shot amount calculation means 432 uses the luminance signal maximum value YnMax and the luminance signal minimum value YnMin as the thresholds of Overshoot and Undershot at the target position, respectively, and calculates the Shot amount Sn based on the following formula (3) or formula (4). calculate.

  For example, when the contour signal En is larger than 0, the shot amount calculation unit 432 calculates the shot amount Sn based on the following equation (3). When the contour signal En is smaller than 0, the shot amount calculation unit 432 calculates the shot amount Sn based on the following equation (4). Further, when the contour signal En is 0, the shot amount calculation unit 432 sets the shot amount Sn = 0. The shot amount Sn calculated by the shot amount calculation unit 432 in this way is input to the shot amount suppression unit 442.

…（３）

…（４）

... (3)

... (4)

(Calculation of low frequency signal maximum value LnMax, low frequency signal minimum value LnMin)
The luminance signal Y extracted by the luminance signal extracting unit 422 is also input to the low frequency signal extracting unit 434. The low frequency signal extraction means 434 calculates the low frequency signal L from the input luminance signal Y based on the following equation (5).

  However, in the following formula (5), L (i, j) represents the value of the low frequency signal L corresponding to the pixel at the position (i, j). Y (i + k, j + l) indicates the value of the luminance signal Y corresponding to the pixel at the position (i + k, j + l). W2 (k, l) is a weighting coefficient of the spatial filter (hereinafter, low-pass filter). The weighting factor w2 is set so that a specific frequency band that is a suppression target of Overshoot and Undershot passes.

…（５）

... (5)

  The low frequency signal L calculated by the low frequency signal extraction unit 434 is input to the low frequency signal maximum value detection unit 436 and the low frequency signal minimum value detection unit 438. The low frequency signal maximum value detecting means 436 calculates the maximum value LnMax (low frequency signal maximum value LnMax) of the low frequency signal of the pixel located around the target position (within a predetermined range) of the input low frequency signal L. To detect. However, the predetermined range setting method can be arbitrarily changed.

  For example, when the predetermined range is set to a one-dimensional range, the low-frequency signal maximum value detection unit 436 generates a low-frequency signal from the low-frequency signal L of the pixels in the one-dimensional range as shown in FIG. The maximum value LnMax is detected. When the predetermined range is set to a two-dimensional range, the low-frequency signal maximum value detection unit 436 generates a low-frequency signal from the low-frequency signal L of the pixels in the two-dimensional range as shown in FIG. The maximum value LnMax is detected. The low frequency signal maximum value LnMax detected by the low frequency signal maximum value detection unit 436 in this way is input to the low frequency range Edge degree calculation unit 440.

  The low frequency signal minimum value detecting means 438 also includes a low frequency signal minimum value LnMin (low frequency signal minimum value LnMin) of a pixel located around the target position (within a predetermined range) of the input low frequency signal L. ) Is detected. However, the predetermined range setting method can be arbitrarily changed.

  For example, when the predetermined range is set to a one-dimensional range, the low-frequency signal minimum value detection unit 438 performs the low-frequency signal from the low-frequency signal L of the pixels in the one-dimensional range as shown in FIG. The minimum value LnMin is detected. Further, when the predetermined range is set to a two-dimensional range, the low frequency signal minimum value detecting means 438 performs the low frequency signal from the low frequency signal L of the pixels within the two dimensional range as shown in FIG. The minimum value LnMin is detected. The low frequency signal minimum value LnMin detected by the low frequency signal minimum value detecting means 438 in this way is input to the low frequency range Edge degree calculating means 440.

(Calculation of low-frequency edge degree α)
As described above, the luminance signal maximum value YnMax, the luminance signal minimum value YnMin, the low frequency signal maximum value LnMax, and the low frequency signal minimum value LnMin are input to the low frequency range Edge degree calculation means 440. The low frequency range edge degree calculating means 440 is a difference absolute value between the luminance signal maximum value YnMax and the low frequency signal maximum value LnMax (hereinafter, maximum value difference), or the luminance signal minimum value YnMin and the low frequency signal minimum value. The low frequency range Edge degree α is calculated based on the absolute value of the difference from LnMin (hereinafter, the minimum value difference).

  As shown in FIG. 9, the smaller the maximum value difference or the minimum value difference, the higher the degree of edge in the low frequency range. On the other hand, as shown in FIG. 10, the greater the maximum value difference or the minimum value difference, the lower the degree of edge in the low frequency range. Therefore, the low frequency range edge degree calculation means 440 calculates the low frequency range edge degree α based on the following equation (6). In the following formula (6), VMax is the maximum value that the luminance signal Y can take. The low frequency range edge degree α calculated by the low frequency range edge degree calculating means 440 in this way is input to the shot amount suppressing means 442.

…（６）

(6)

(Calculation of suppression shot amount SSn)
As described above, the shot amount Sn and the low frequency range edge degree α are input to the shot amount suppression unit 442. The shot amount suppression means 442 calculates a suppressed shot amount SSn in which the shot amount Sn is suppressed according to the input low frequency range edge degree α. However, when the Shot amount Sn is 0, the Shot amount suppression unit 442 determines that Overshoot and Undershot do not occur, and does not suppress the Shot amount Sn. In this case, the Shot amount suppression unit 442 inputs the original Shot amount Sn to the corrected contour emphasis signal calculation unit 444 as the suppression Shot amount SSn.

  On the other hand, if the Shot amount Sn is greater than 0, the Shot amount suppressing means 442 determines that Overshoot has occurred at the position of interest (see FIG. 8). Also, if the Shot amount Sn is smaller than 0, the Shot amount suppressing means 442 determines that Undershoot has occurred at the position of interest (see FIG. 8). Then, the shot amount suppression unit 442 suppresses the shot amount Sn so as to approach 0 as the low frequency range edge degree α decreases, for example, based on the following equation (7).

…（７）

... (7)

  However, in said Formula (7), k is a suppression coefficient which determines the suppression degree of Shot amount Sn. As shown in FIG. 11, the suppression coefficient k is determined based on the low frequency range Edge degree α. In the example of FIG. 11, the suppression coefficient k increases in proportion to the low frequency range Edge degree α, but the low frequency range Edge degree α and the suppression coefficient k do not necessarily have a proportional relationship. For example, a relationship in which the suppression coefficient k increases in the nth order (n ≧ 2) of the low frequency range Edge degree α may be used. Further, as illustrated in FIG. 11, when the low frequency range Edge degree α exceeds a predetermined value, the suppression coefficient k becomes a constant value 1.0.

  The suppression shot amount SSn calculated by the shot amount suppression unit 442 in this way is input to the corrected contour emphasis signal calculation unit 444. The shot amount suppression unit 442 corrects the clip value as the suppression shot amount SSn when the suppression shot amount SSn exceeds the clip value so that the suppression shot amount SSn is equal to or less than a predetermined value (clip value). The calculation unit 444 may be configured to be input.

(Calculation of corrected contour emphasis signal CYEn)
As described above, the contour enhancement signal YEn, the luminance signal maximum value YnMax, the luminance signal minimum value YnMin, and the suppression shot amount SSn are input to the corrected contour enhancement signal calculation unit 444. The corrected contour emphasizing signal calculation means 444 is based on the input contour emphasizing signal YEn, luminance signal maximum value YnMax, luminance signal minimum value YnMin, and suppression shot amount SSn, and the corrected contour emphasizing signal CYEn according to the following equation (8). Is calculated.

  However, when the corrected Shot amount SSn is 0 (when the Shot amount Sn is 0), the corrected contour emphasis signal calculation unit 444 outputs the contour emphasis signal YEn as the corrected contour emphasis signal CYEn. On the other hand, when the corrected Shot amount SSn is not 0, the corrected outline emphasis signal calculating unit 444 outputs a corrected outline emphasis signal CYEn calculated based on the following equation (8). The corrected contour emphasis signal CYEn is a contour emphasis signal in which overshoot and undershoot are suppressed in consideration of the degree of edge in the low frequency range.

…（８）

(8)

(Summary)
As described above, by calculating the suppression Shot amount SSn based on the low frequency range Edge degree α and executing the edge enhancement process using the suppression Shot amount SSn, as shown in FIGS. An appropriate edge enhancement effect can be obtained. Referring to FIGS. 12 to 14, a corrected contour emphasizing signal is obtained in which Overshot and Undershot that should be strongly suppressed are strongly suppressed, and Overshot and Undershot that should not be suppressed so much are weakly suppressed. Such an adaptive suppression effect cannot be obtained when a method for suppressing Overshoot and Undershot according to the flatness of the waveform of the image signal is applied.

[3: Outline enhancement processing flow]
Next, the flow of the contour enhancement process according to the present embodiment will be described with reference to FIGS. 15 and 16. 15 and 16 are explanatory diagrams showing the flow of the contour emphasis process according to the present embodiment. Note that the contour enhancement processing shown in FIGS. 15 and 16 is realized by the function of the contour enhancement processing unit 42 described above.

  As shown in FIG. 15, first, an image signal is input to the luminance signal extraction means 422 (S102). Next, the luminance signal extraction unit 422 calculates the luminance signal Y from the input image signal (S104). The luminance signal Y calculated here is input to the contour signal extraction unit 424, the luminance signal maximum value detection unit 428, the luminance signal minimum value detection unit 430, and the low frequency signal extraction unit 434.

  Next, the contour signal extraction unit 424 extracts the contour signal E from the input luminance signal Y (S106). The contour signal E extracted here is input to the contour enhancement signal calculation means 426. Next, the low frequency signal extraction means 434 extracts the low frequency signal L from the input luminance signal (S108). The low frequency signal L extracted here is input to the low frequency signal maximum value detecting means 436 and the low frequency signal minimum value detecting means 438.

  Next, as shown in FIG. 16, a processing loop for all target pixels is started (S110). This processing loop repeats the processing from steps S112 to S128 described later while moving the position of the target pixel with respect to the positions of all the pixels included in the image signal. That is, the process from step S112 to S128 is executed on a pixel basis for all pixels of the pixel signal.

  For example, assuming that the index indicating the position of the target pixel is n and the number of pixels included in the image signal is N, the processing steps surrounded by steps S110 and S130 are repeatedly executed while incrementing the index n from 1 to N. The Hereinafter, a case where the index indicating the position of the target pixel is n will be described.

  In this processing loop, first, the contour enhancement signal calculation unit 426 calculates a contour enhancement signal YEn based on the input contour signal E (S112). The contour enhancement signal YEn calculated here is input to the Shot amount calculation unit 432 and the corrected contour enhancement signal calculation unit 444.

  Next, the luminance signal maximum value detection unit 428 detects the luminance signal maximum value YnMax from the input luminance signal Y (S114). Further, the luminance signal minimum value detecting means 430 detects the luminance signal minimum value YnMin from the inputted luminance signal Y (S114). The detected luminance signal maximum value YnMax and luminance signal minimum value YnMin are input to the Shot amount calculating unit 432, the low frequency range edge degree calculating unit 440, and the corrected contour enhancement signal calculating unit 444.

  Next, the Shot amount calculation unit 432 calculates the Shot amount Sn based on the input contour emphasis signal YEn, the luminance signal maximum value YnMax, and the luminance signal minimum value YnMin (S116). The calculated Shot amount Sn is input to the Shot amount suppressing means 442. Next, the Shot amount suppressing means 442 determines whether or not the inputted Shot amount Sn is 0 (S118). If the Shot amount Sn = 0, the process proceeds to Step S120. On the other hand, if the amount of shot Sn is not 0, the process proceeds to step S122.

  When the process proceeds to step S120, the shot amount suppression unit 442 does not suppress the shot amount Sn, and the corrected outline emphasis signal calculation unit 444 outputs the outline emphasis signal YEn without correction (S120). On the other hand, when the process proceeds to step S122, the low-frequency signal maximum value detecting unit 436 detects the low-frequency signal maximum value LnMax from the input low-frequency signal L (S122). The low frequency signal minimum value detecting means 438 detects the low frequency signal minimum value LnMin from the input low frequency signal L (S122). The detected low frequency signal maximum value LnMax and the low frequency signal minimum value LnMin are input to the low frequency range edge degree calculating means 440.

  Next, the low frequency range edge degree calculation means 440 calculates the low frequency range edge degree α based on the input luminance signal maximum value YnMax, luminance signal minimum value YnMin, low frequency signal maximum value LnMax, and low frequency signal minimum value LnMin. Is calculated (S124). The low frequency range edge degree α calculated here is input to the shot amount suppressing means 442.

  Next, the shot amount suppressing means 442 calculates the suppressed shot amount SSn based on the input shot amount Sn and the low frequency range edge degree α (S126). The suppression shot amount SSn calculated here is input to the corrected contour emphasis signal calculation means 444. Next, the corrected contour emphasizing signal calculation unit 444 calculates a corrected contour emphasizing signal CYEn based on the input luminance signal maximum value YnMax, luminance signal minimum value YnMin, and suppression shot amount SSn (S128). Then, the corrected contour emphasis signal CYEn calculated here is output.

  When the process of step S120 or step S128 is completed, the process proceeds to step S130. At this time, the contour emphasis signal (the contour emphasis signal YEn or the corrected contour emphasis signal CYEn) at the target position corresponding to the index n is output. Therefore, when the process proceeds to step S130, the process returns to step S110, the index is incremented by 1, and the process steps from step S112 to S130 are executed again. When the contour emphasis signal is output for all the pixels included in the image signal, a series of processing ends.

  The flow of the contour enhancement process according to the present embodiment has been described above. As described above, the effects of the edge portion to be emphasized are determined in order to determine the degree of the low frequency range edge by considering the frequency to be suppressed and the overshoot and the undershot generated in the low frequency range edge. It is possible to appropriately suppress the low frequency range Edge to be suppressed without impairing.

  As mentioned above, although preferred embodiment of this invention was described referring an accompanying drawing, it cannot be overemphasized that this invention is not limited to the example which concerns. It will be apparent to those skilled in the art that various changes and modifications can be made within the scope of the claims, and these are naturally within the technical scope of the present invention. Understood.

DESCRIPTION OF SYMBOLS 10 Imaging device 12 Lens 14 Aperture 16 Shutter 18 Image sensor 20 AFE circuit 22 Pre-processing part 24 Post-processing part 26 Data compression part 28 Memory card interface 30 Memory card 32 White balance processing part 34 Demosaic processing part 36 Noise reduction processing part 38 Gamma Correction processing unit 40 Color correction processing unit 42 Contour emphasis processing unit 422 Luminance signal extraction means 424 Contour signal extraction means 426 Contour emphasis signal calculation means 428 Luminance signal maximum value detection means 430 Luminance signal minimum value detection means 432 Shot amount calculation means 434 Low Frequency signal extraction means 436 Low frequency signal maximum value detection means 438 Low frequency signal minimum value detection means 440 Low frequency range Edge degree calculation means 442 Shot amount suppression means 444 Correction contour emphasis signal calculation means

Claims (7)

Contour enhancement means for generating a contour enhancement signal by adding a luminance signal corresponding to each of a plurality of pixels constituting an image and a contour signal indicating the luminance of a contour portion included in the image;
Low frequency signal extraction means for extracting a low frequency signal that is a low frequency band component of the luminance signal;
Maximum / minimum value detection means for detecting the maximum value and minimum value of the luminance signal and the maximum value and minimum value of the low-frequency signal for a plurality of pixels located around the pixel of interest;
The larger the absolute value of the difference between the maximum value of the low frequency signal and the maximum value of the luminance signal, or the absolute value of the difference between the minimum value of the low frequency signal and the minimum value of the luminance signal, the lower the value. A low frequency edge degree calculating means for calculating a low frequency edge degree that is higher as the absolute value is smaller;
The shot amount of the target pixel calculated based on the contour emphasis signal and the maximum value or the minimum value of the luminance signal is suppressed so as to approach zero as the low frequency edge degree is smaller, and is suppressed. A suppression Shot amount calculating means for calculating
Contour enhancement signal correction means for correcting the contour enhancement signal based on the suppression shot amount calculated by the suppression shot amount calculation means;
An image processing apparatus comprising: The low-frequency edge degree calculating means calculates the low-frequency edge degree based on an absolute difference between the maximum value of the low-frequency signal and the maximum value of the luminance signal when the amount of shot is larger than a predetermined value. The low-frequency edge degree is calculated based on an absolute difference value between a minimum value of the low-frequency signal and a minimum value of the luminance signal when the amount of shot is smaller than a predetermined value. The image processing apparatus according to 1. The Shot amount calculating means calculates the Shot amount of the target pixel based on the contour emphasis signal and the maximum value or the minimum value of the luminance signal, and when the value of the contour signal is positive, A Shot amount of the target pixel is calculated based on a contour emphasis signal and the maximum value of the luminance signal, and when the value of the contour signal is negative, based on the contour emphasis signal and the minimum value of the luminance signal. The image processing apparatus according to claim 2, wherein a Shot amount of the target pixel is calculated. The suppression Shot amount calculation means calculates the suppression Shot amount so that the smaller the low frequency edge degree is, the closer to 0 is, and the lower the low frequency edge degree is, the closer the Shot amount is. The image processing apparatus according to claim 3. The maximum / minimum value detecting means is configured to detect the maximum value and minimum value of the luminance signal or the maximum value of the low-frequency signal for a plurality of pixels located within a predetermined range with reference to a target pixel as a correction amount of a shot amount. The image processing apparatus according to claim 4, wherein when the value and the minimum value are detected, a predetermined range for determining the plurality of pixels to be detected is arbitrarily changed. A luminance signal extracting means for extracting a luminance signal from an image signal of an image composed of a plurality of pixels;
Contour signal extraction means for extracting a contour signal indicating the contour of an image from the luminance signal;
The image processing apparatus according to claim 5, further comprising: A contour emphasizing step for generating a contour emphasizing signal by adding a luminance signal corresponding to each of a plurality of pixels constituting the image and a contour signal indicating the luminance of a contour part included in the image;
A low frequency signal extraction step of extracting a low frequency signal which is a low frequency band component of the luminance signal;
A maximum / minimum value detecting step for detecting a maximum value and a minimum value of the luminance signal and a maximum value and a minimum value of the low-frequency signal for a plurality of pixels located around the pixel of interest;
The larger the absolute value of the difference between the maximum value of the low frequency signal and the maximum value of the luminance signal, or the absolute value of the difference between the minimum value of the low frequency signal and the minimum value of the luminance signal, the lower the value. A low frequency edge degree calculating step for calculating a low frequency edge degree that becomes a higher value as the absolute value is smaller;
The shot amount of the target pixel calculated based on the contour emphasis signal and the maximum value or the minimum value of the luminance signal is suppressed so as to approach zero as the low frequency edge degree is smaller, and is suppressed. A suppression shot amount calculating step for calculating
A contour enhancement signal correcting step of correcting the contour enhancement signal based on the suppression shot amount calculated in the suppression shot amount calculation step;
An image processing method comprising the steps of:

Images