WO2010070732A1

WO2010070732A1 - Image processing device, display device, image processing method, program therefor, and recording medium containing the program

Info

Publication number: WO2010070732A1
Application number: PCT/JP2008/072846
Authority: WO
Inventors: 賢司奥道
Original assignee: パイオニア株式会社
Priority date: 2008-12-16
Filing date: 2008-12-16
Publication date: 2010-06-24

Abstract

The image processing device of a display device recognizes a focus area (Af) and an unfocus area (Au) in an input image, applies contour emphasizing processing to the focus area (Af), and generates an output image (Po) in which blurring processing is applied to the unfocus area (Au). Thus, the image processing device can effectively emphasize the sense of depth in the output image (Po) by applying the contour emphasizing processing to the focus area (Af) and the blurring processing to the unfocus area (Au) at the same time.

Description

Image processing apparatus, display apparatus, image processing method, program thereof, and recording medium recording the program

The present invention relates to an image processing device, a display device, an image processing method, a program thereof, and a recording medium on which the program is recorded.

2. Description of the Related Art Conventionally, there has been known a configuration in which an object contour enhancement process is performed on an image (see, for example, Patent Document 1 and Patent Document 2).
In the device described in Patent Document 1, the level of the peak point of the video signal is calculated, and the correction value is calculated from the calculated level with the rated video signal level being 100%. And the structure which correct | amends the level of an outline emphasis signal according to this correction value is taken.
According to the technique disclosed in Patent Document 2, in a scene where the brightness and contrast of a subject are large, the center frequency of the aperture signal is increased so that a fine picture can be more easily seen by the fine aperture signal. Further, in a scene where the brightness and contrast of the subject are small, the center frequency of the aperture signal is lowered, and a sharp and clear image is obtained with a thick and clear aperture signal.

JP 2004-193769 A Japanese Patent Laid-Open No. 6-30303

By the way, in the image, there are a focus area where the subject is focused and an unfocus area where the subject is not focused. This means that the distance from the subject in the focus area to the imaging means and the subject in the unfocus area This means that a sense of depth is expressed in the image because the distance from the camera to the imaging means is different. With recent advances in image processing technology, it is considered that there is a need to enhance the sense of depth by processing images expressing such a sense of depth.
However, in the configurations as disclosed in

Patent Documents

1 and 2, there is a possibility that the contour emphasis processing may be performed regardless of whether the region is a focus region or an unfocus region, and there is a problem in that the sense of depth cannot be effectively enhanced. is there.

An object of the present invention is to provide an image processing device, a display device, an image processing method, a program thereof, and a recording medium on which the program is recorded, which can effectively enhance the sense of depth of the image.

An image processing apparatus according to the present invention includes a focus state recognition unit that recognizes a focus area in which an object is in focus in an input image and an unfocus area in which the object is not in focus, and a contour for the recognized focus area And a depth processing unit that performs an enhancement process and generates an output image obtained by performing a blurring process on the unfocus area.

A display device according to the present invention includes the above-described image processing device and a display unit that displays an output image generated by the image processing device.

The image processing method of the present invention is an image processing method for generating an output image obtained by performing predetermined processing on an input image by a calculation unit, and the calculation unit focuses on a subject in the input image. A focus state recognition step for recognizing a focused area and an unfocused area where the subject is not in focus, contour enhancement processing for the recognized focus area, and blurring processing for the unfocused area And a depth processing step for generating the executed output image.

An image processing program according to the present invention is characterized by causing an arithmetic means to execute the above-described image processing method.

The image processing program of the present invention is characterized in that the calculation means functions as the above-described image processing apparatus.

The recording medium on which the image processing program of the present invention is recorded is characterized in that the above-described image processing program is recorded so as to be readable by the arithmetic means.

It is a block diagram which shows schematic structure of the display apparatus which concerns on 1st, 4th embodiment of this invention, a modification, and another modification. It is a schematic diagram of the input image in the first embodiment. It is a schematic diagram of the output image in the first embodiment. FIG. 2 is a block diagram illustrating a schematic configuration of a focus state recognition unit in the first embodiment. It is a schematic diagram which shows the zone | band characteristic calculation block in the input image in the said 1st Embodiment. It is a schematic diagram which shows the relationship between the frequency band characteristic after Hadamard transformation in the said 1st Embodiment, and a high region component and a low region component. It is a block diagram which shows schematic structure of the polarity gain determination part in the said 1st Embodiment. It is a schematic diagram which shows the setting state of the focus degree according to block in each setting unit block in the said 1st Embodiment. It is a schematic diagram which shows the setting state of the difference value of the average value of the focus degree according to block in the said 1st Embodiment, and the focus degree according to block of FIG. It is a schematic diagram which shows the state which took the absolute value of the difference value of FIG. 9 in the said 1st Embodiment. It is a schematic diagram which shows the positional relationship of the process target pixel which is a linear interpolation process target in the said 1st Embodiment, and a center pixel. It is a schematic diagram for demonstrating the calculation method of the data conversion value in the said 1st Embodiment. It is a block diagram which shows schematic structure of the display apparatus which concerns on 2nd Embodiment of this invention. It is a block diagram which shows schematic structure of the focus state recognition part in the said 2nd Embodiment. It is a schematic diagram for demonstrating the calculation method of the horizontal band centroid value and the vertical band centroid value in the said 2nd Embodiment. It is a schematic diagram which shows the setting state of the filter horizontal band and unsharp horizontal band in the said 2nd Embodiment. It is a schematic diagram which shows the adjustment state of the filter horizontal band and unsharp horizontal band with respect to the focus area | region and unfocus area | region in the said 2nd Embodiment. It is a block diagram which shows schematic structure of the display apparatus which concerns on 3rd Embodiment of this invention. It is a block diagram which shows schematic structure of the focus state recognition part in the said 4th Embodiment. FIG. 10 is a schematic diagram showing first to fourth regions in an input image in the fourth embodiment. It is a block diagram which shows schematic structure of the focus state recognition part which concerns on the said modification. It is a block diagram which shows schematic structure of the focus state recognition part which concerns on the said other modification.

Explanation of symbols

DESCRIPTION OF

SYMBOLS

100, 200, 300, 400, 500, 600 ... Display apparatus 110 ...

Display part

120, 220, 320, 420, 520, 620 ...

Image processing apparatus

130, 230, 430, 530, 630 ... Focus state recognition Unit 140 ... time constant processing unit 150 ... inter-block linear interpolation unit 321 ... face detection unit 322 ... gain weighting unit as a face correspondence correction control unit 437 ... screen position weight addition unit as a position correspondence correction control unit Dg ... position correspondence correction Weight added focus degree as information Gk: Weighted gain correspondence value as face correspondence correction information

[First Embodiment]
DESCRIPTION OF EXEMPLARY EMBODIMENTS Hereinafter, a first embodiment according to the invention will be described with reference to the drawings.
In the first embodiment and second to fourth embodiments to be described later, description will be given by exemplifying a configuration that is a display device including the image processing device of the present invention and emphasizes the sense of depth of an image.
FIG. 1 is a block diagram illustrating a schematic configuration of a display device. FIG. 2 is a schematic diagram of an input image. FIG. 3 is a schematic diagram of an output image. FIG. 4 is a block diagram illustrating a schematic configuration of the focus state recognition unit. FIG. 5 is a schematic diagram showing a band characteristic calculation block in the input image. FIG. 6 is a schematic diagram showing the relationship between the frequency band characteristics after Hadamard transform and the high-frequency component and low-frequency component. FIG. 7 is a block diagram illustrating a schematic configuration of the polarity gain determination unit. FIG. 8 is a schematic diagram showing a setting state of the degree of focus for each block in each setting unit block. FIG. 9 is a schematic diagram showing a setting state of a difference value between the average value of the degree of focus for each block and the degree of focus for each block of FIG. FIG. 10 is a schematic diagram showing a state in which the absolute value of the difference value in FIG. 9 is taken. FIG. 11 is a schematic diagram illustrating a positional relationship between a processing target pixel which is a linear interpolation processing target and a central pixel. FIG. 12 is a schematic diagram for explaining a method of calculating a data conversion value.

[Configuration of display device]
As shown in FIG. 1, the display device 100 includes a display unit 110 and an image processing device 120 as a calculation unit.
The display unit 110 displays the image processed by the image processing device 120. Examples of the display unit 110 include a PDP (Plasma Display Panel), a liquid crystal panel, an organic EL (Electro Luminescence) panel, a CRT (Cathode-Ray Tube), an FED (Field Emission Display), and an electrophoretic display panel.

The image processing apparatus 120 processes the input image Pi including the focus area Af where the subject is focused and the unfocus area Au where the subject is not focused, as shown in FIG. The output image Po in which the feeling is emphasized is output to the display unit 110. The image processing apparatus 120 includes a focus state recognition unit 130, a time constant processing unit 140, an inter-block linear interpolation unit 150, and a depth processing unit 160 that are configured from various programs.

As shown in FIG. 4, the focus state recognition unit 130 includes a Hadamard transform unit 131, a high frequency component accumulation unit 132, a low frequency component accumulation unit 133, a focus degree calculation unit 134, a block-by-block accumulation unit 135, A polarity gain determination unit 136.
The Hadamard transform unit 131 acquires the input signal Vi of the input image Pi from the image signal output unit 10. Further, based on this input signal Vi, as shown in FIG. 5, Hadamard transform is performed for each band characteristic calculation block Bf composed of a total of 64 pixels (not shown) arranged in a vertical direction and 8 in the horizontal direction. Processing is performed to calculate a frequency band component F of the input signal Vi as shown in FIG. Here, the square portion (indicated by a black square) at the upper left corner in FIG. 6 is a direct current band, so the component is replaced with 0. Then, the frequency band component F is output to the high frequency component accumulating unit 132 and the low frequency component accumulating unit 133.

The high-frequency component accumulating unit 132 converts a component of the frequency band component F that has a frequency band equal to or higher than a preset threshold to a high-frequency component Fh (a square portion that is hatched with a line extending diagonally to the left in FIG. 6). And the high frequency component accumulation value Sh obtained by accumulating the high frequency component Fh for each band characteristic calculation block Bf is output to the focus degree calculation unit 134.
The low-frequency component accumulating unit 133 recognizes a component of the frequency band component F that is less than the above threshold value as a low-frequency component Fl (a square portion hatched with a line extending diagonally upward to the right in FIG. 6). The low frequency component accumulation value S1 obtained by accumulating the frequency component Fl for each band characteristic calculation block Bf is output to the focus degree calculation unit 134.
The focus degree calculation unit 134 outputs, for each band characteristic calculation block Bf, a value obtained by dividing the high-frequency component accumulated value Sh by the low-frequency component accumulated value S1 to the block-by-block accumulation unit 135 as the focus degree Du.
The block-by-block accumulating unit 135 determines the polarity-by-block focus degree Ds obtained by accumulating the degree of focus Du for each set unit block Bs composed of (a × b) band characteristic calculation blocks Bf as shown in FIG. To the unit 136. The setting unit block Bs is a block obtained by dividing the input image Pi into five pieces in the vertical direction and eight pieces in the horizontal direction. The block-specific focus degree Ds is set to a value of 0 to 255 (8 bits).

The polarity gain determination unit 136 sets polarity information and a gain value for each setting unit block Bs based on the block-specific focus degree Ds. As shown in FIG. 7, the polarity gain determination unit 136 includes an average calculation unit 136A, a difference calculation unit 136B, a polarity determination unit 136C, and a gain corresponding value setting unit 136D.
The average calculation unit 136A calculates an average value of the block-specific focus degrees Ds and outputs the calculation result to the difference calculation unit 136B. For example, as shown in FIG. 8, when the block-by-block focus degree Ds (the numerical value in the square) of the set unit block Bs illustrated by a square is set, the average of the block-by-block focus degrees Ds of all the set unit blocks Bs The value is calculated as 64.
The difference calculation unit 136B calculates a value obtained by subtracting the average value from the average calculation unit 136A from the block-specific focus degree Ds of each setting unit block Bs from the block-by-block accumulation unit 135 as the difference value of each setting unit block Bs. Then, the calculation result is output to the polarity determination unit 136C and the gain corresponding value setting unit 136D. For example, the average value 64 is subtracted from the block-specific focus degree Ds of each setting unit block Bs shown in FIG. 8 to obtain a difference value as shown in FIG.

The polarity determination unit 136C acquires the difference value from the difference calculation unit 136B, and sets the setting unit block Bs having the difference value of 0 or more to the focus area Af (a line extending diagonally right above in FIGS. 9, 10, and 12). And the polarity information Km of “+1” indicating that fact is output to the time constant processing unit 140. Further, it is determined that the set unit block Bs having the difference value less than 0 is the unfocus area Au (a square portion hatched with a line extending diagonally to the upper left in FIGS. 9, 10, and 12). Is output to the time constant processing unit 140.
The gain corresponding value setting unit 136D acquires the difference value from the difference calculation unit 136B and calculates the absolute value of the difference value of each setting unit block Bs as shown in FIG. Further, when the absolute value of the setting unit block Bs is 127 or less, this absolute value is set as a gain corresponding value. When the absolute value is 128 or more, 127 is set as a gain corresponding value, that is, the absolute value represented by 8 bits is limited to 7 bits. Then, the gain corresponding value setting unit 136D outputs the gain corresponding value Gm to the time constant processing unit 140.

Based on the input signal Vi, the time constant processing unit 140 determines whether or not the continuous input image Pi corresponds to a moving image scene change. Here, the scene change means that the relevance between successive input images Pi is low, such as when a moving image scene changes. If it is determined that the scene change is detected, the polarity information Km and the gain corresponding value Gm are output as they are to the inter-block linear interpolation unit 150 as the time constant processing polarity information Kj and the time constant processing gain corresponding value Gj.
On the other hand, when the time constant processing unit 140 determines that the scene change does not occur, the time constant processing unit 140 performs time constant processing on each of the polarity information Km and the gain corresponding value Gm to thereby obtain the time constant processing polarity information Kj and the time constant processing gain. Corresponding value Gj is calculated.
Specifically, time constant processing polarity information Kj and time constant processing gain corresponding value Gj are calculated by performing IIR (Infinite Impulse Response) filter processing in the time direction on polarity information Km and gain corresponding value Gm. . Then, the time constant processing unit 140 outputs the time constant processing polarity information Kj and the time constant processing gain corresponding value Gj to the inter-block linear interpolation unit 150.

The inter-block linear interpolation unit 150 performs linear interpolation processing on each of the time constant processing polarity information Kj and the time constant processing gain corresponding value Gj, and performs linear interpolation polarity information Ks and linear interpolation for each pixel Q shown in FIG. A gain corresponding value Gs is calculated. Note that the size of the pixel Q with respect to the setting unit block Bs is exaggerated for easy understanding of the content.
Specifically, the inter-block linear interpolation unit 150 calculates linear interpolation polarity information Ks and a linear interpolation gain corresponding value Gs of a processing target pixel (hereinafter referred to as a processing target pixel) Qs included in the setting unit block Bs2. In this case, the time constant processing polarity is applied to the central pixel Q1, Q3, and Q4 of the setting unit block Bs1, Bs3, and Bs4 adjacent to the setting unit block Bs2 and the pixel Q2 (hereinafter referred to as the central pixel) Q2 positioned at the center of the setting unit block Bs2. It is assumed that the information Kj and the time constant processing gain corresponding value Gj are set. Further, the data conversion value H is calculated by substituting the time constant processing polarity information Kj and the time constant processing gain corresponding value Gj of each setting unit block Bs into the following equation (1).

(Equation 1)
H = (Kj × 128) + Gj (1)

For example, when the time constant processing polarity information Kj and the time constant processing gain corresponding value Gj of the setting unit blocks Bs1 to Bs4 are set as shown in FIG. 12, the data conversion values H1, H2, and the like of the setting unit blocks Bs1 to Bs4 are set. H3 and H4 are 104, 150, 104 and 162, respectively.
The inter-block linear interpolation unit 150 then calculates the vertical distance m from the processing target pixel Qs to the central pixel Q2, the vertical distance n to the central pixel Q4, the horizontal distance x to the central pixel Q1, and the central pixel Q2. The first linear interpolation value I of the processing target pixel Qs is calculated by substituting the horizontal direction distance y and the data conversion values H1, H2, H3, and H4 into the following equation (2).

(Equation 2)
I = ((((H1 * y + H2 * x) / (x + y)) * n + ((H3 * y + H4 * x) / (x + y)) * m) / (n + m) (2)

For example, if x = 96, y = 32, m = 32, n = 96, and the values shown in FIG. 12 (H1 = 104, H2 = 150, H3 = 104, H4 = 162) are substituted into H1 to H4, linear interpolation is performed. The value I is 140.75.
Further, the inter-block linear interpolation unit 150 calculates the second linear interpolation value by subtracting 128 from the linear interpolation value I.
Then, when the second linear interpolation value is greater than or equal to 0, the inter-block linear interpolation unit 150 indicates that the polarity of the processing target pixel Qs is “1”, that is, the linear interpolation that the processing target pixel Qs is included in the focus area Af. When the polarity information Ks is output to the polarity adjusting unit 162 described later of the depth processing unit 160 and the second linear interpolation value is less than 0, the polarity of the processing target pixel Qs is “−1”, that is, the processing target pixel Qs is Linear interpolation polarity information Ks that is included in the unfocus area Au is output.
Further, the inter-block linear interpolation unit 150 outputs the absolute value of the second linear interpolation value as a linear interpolation gain corresponding value Gs to the gain adjustment unit 163 described later of the depth processing unit 160.
For example, in the case shown in FIGS. 11 and 12, the second linear interpolation value is calculated as 12.75, and the inter-block linear interpolation unit 150 performs linear processing indicating that the polarity of the processing target pixel Qs is “1”. The interpolation polarity information Ks is output to the polarity adjustment unit 162, and the value of 12.75 is output to the gain adjustment unit 163 as the linear interpolation gain corresponding value Gs.

As shown in FIG. 1, the depth processing unit 160 includes a high frequency component detection unit 161, a polarity adjustment unit 162, a gain adjustment unit 163, and an addition unit 164.
The high-frequency component detection unit 161 acquires the input signal Vi from the image signal output unit 10 and performs high-pass filter (HPF) processing such as a Laplacian filter, that is, the subject in the input image Pi. The components constituting the contour and the like are output to the polarity adjustment unit 162.

The polarity adjustment unit 162 acquires the high frequency component Vh from the high frequency component detection unit 161 and the linear interpolation polarity information Ks from the inter-block linear interpolation unit 150. When the linear interpolation polarity information Ks of the pixel Q corresponding to the high frequency component Vh is “1”, that is, when the high frequency component Vh is in the focus area Af, the polarity without changing the polarity (positive / negative) of the high frequency component Vh. The adjusted high frequency component Vk is output to the gain adjusting unit 163. In addition, when the linear interpolation polarity information Ks of the pixel Q corresponding to the high frequency component Vh is “−1”, that is, when the high frequency component Vh is in the unfocused area Au, the polarity of the high frequency component Vh is changed to change the polarity. The component Vk is output to the gain adjustment unit 163.

The gain adjustment unit 163 acquires the linear interpolation gain corresponding value Gs of the predetermined pixel Q from the inter-block linear interpolation unit 150, and sets a value obtained by dividing the linear interpolation gain corresponding value Gs by 64 as a gain value. To do. Then, the gain adjustment high frequency component Vg obtained by multiplying the gain value by the polarity adjustment high frequency component Vk corresponding to the pixel Q from the polarity adjustment unit 162 is output to the addition unit 164.

The addition unit 164 acquires the input signal Vi from the image signal output unit 10 and the gain adjustment high-frequency component Vg from the gain adjustment unit 163. Then, the gain adjustment high-frequency component Vg corresponding to each pixel Q and the high-frequency component Vh corresponding to the gain adjustment high-frequency component Vg in the input signal Vi are added to generate the output signal Vo of the output image Po, and the display unit 110 Output.

With the above processing, in the focus area Af, the polarity of the high frequency component Vh is the same as the polarity of the gain adjustment high frequency component Vg. Therefore, the high frequency component Vh and the gain adjustment high frequency component Vg are added, the high frequency component of the focus area Af in the output image Po is larger than that of the input image Pi, and the outline of the focus area Af in the output image Po is To be emphasized.
In the unfocus area Au, the polarity of the high frequency component Vh is different from the polarity of the gain adjustment high frequency component Vg. For this reason, the gain adjustment high frequency component Vg is subtracted from the high frequency component Vh, the high frequency component of the unfocus area Au in the output image Po becomes smaller than that of the input image Pi, and the contour of the unfocus area Au in the output image Po. Will be blurred.

[Effects of First Embodiment]
As described above, in the first embodiment, the following operational effects can be achieved.

(1) The image processing device 120 of the display device 100 recognizes the focus area Af and the unfocus area Au in the input image Pi, performs an edge emphasis process on the focus area Af, and applies the unfocus area Au to the unfocus area Au. Thus, an output image Po subjected to the blurring process is generated.
For this reason, it is possible to effectively enhance the sense of depth in the output image Po by simultaneously performing the contour enhancement process and the blurring process on the input image Pi.

(2) The image processing device 120 calculates a time constant processing gain corresponding value Gj by performing time constant processing on the gain corresponding value Gm of the input image Pi that is configured and continuously input. Then, based on the time constant processing gain corresponding value Gj, an output image Po subjected to contour enhancement processing and blur processing is generated. That is, the image processing apparatus 120 performs time constant processing on the degree of contour enhancement and the degree of blurring on the input image Pi, and an output image that has been subjected to the contour enhancement processing and blurring processing based on the degree of the time constant processing. Generate Po.
Here, when the difference in gain corresponding value Gm between consecutive input images Pi is large, the degree of contour emphasis or blurring between input images Pi is large when contour emphasis processing or blurring processing is performed based on the gain corresponding value Gm. It may change and a moving image with a sense of incongruity may be displayed.
On the other hand, by performing contour emphasis processing and blurring processing based on the time constant processing gain corresponding value Gj, contour emphasis and blurring between the input images Pi can be performed as compared with processing based on the gain corresponding value Gm. The change in the degree can be reduced, and a moving image with a reduced sense of incongruity can be displayed.

(3) The image processing device 120 does not perform time constant processing when it is determined that the input image Pi corresponds to a scene change, and performs time constant processing when it is determined that the input image Pi does not correspond to a scene change.
Here, if time constant processing is performed at the time of a scene change, a time constant processing gain corresponding value Gj reflecting the gain corresponding value Gm of the immediately preceding input image Pi with low relevance for a predetermined input image Pi. Since contour enhancement processing and blurring processing are performed based on this, there is a possibility that a moving image with a sense of incongruity may be displayed.
On the other hand, by not performing the time constant processing in the case of a scene change, the predetermined input image Pi is contoured without reflecting the gain corresponding value Gm of the immediately preceding input image Pi with low relevance. Emphasis processing and blurring processing can be performed, and moving images with a reduced sense of incongruity can be displayed.

(4) The image processing apparatus 120 calculates a linear interpolation gain corresponding value Gs by performing linear interpolation processing on the time constant processing gain corresponding value Gj of a predetermined adjacent pixel Q of the input image Pi. Then, based on the linear interpolation gain corresponding value Gs, an output image Po that has undergone contour enhancement processing and blurring processing is generated. That is, the image processing apparatus 120 performs linear interpolation processing on the degree of contour enhancement and the degree of blurring on adjacent pixels Q in the input image Pi, and performs contour enhancement processing and processing based on the degree of linear interpolation processing. An output image Po subjected to the blurring process is generated.
Here, when the difference between the time constant processing gain corresponding values Gj with respect to the adjacent pixels Q is large, if contour enhancement processing or blurring processing is performed based on the time constant processing gain correspondence values Gj, contour enhancement is performed at the boundary between the pixels Q. Or the degree of blurring may be discontinuous, and a moving image with a discontinuous boundary may be displayed.
On the other hand, by performing contour enhancement processing and blurring processing based on the linear interpolation gain corresponding value Gs, the contour at the boundary portion between the pixels Q is compared with the case where processing based on the time constant processing gain corresponding value Gj is performed. The continuity of the degree of emphasis and blurring can be increased, and a moving image with a continuous boundary can be displayed.

[Second Embodiment]
Next, a second embodiment according to the present invention will be described with reference to the drawings.
In addition, about the structure same as 1st Embodiment, the same name and code | symbol are attached | subjected and description is abbreviate | omitted or simplified.
FIG. 13 is a block diagram illustrating a schematic configuration of the display device. FIG. 14 is a block diagram illustrating a schematic configuration of the focus state recognition unit. FIG. 15 is a schematic diagram for explaining a method of calculating the horizontal band centroid value and the vertical band centroid value. FIG. 16 is a schematic diagram illustrating a setting state of the filter horizontal band and the unsharp horizontal band. FIG. 17 is a schematic diagram illustrating an adjustment state of the filter horizontal band and the unsharp horizontal band with respect to the focus area and the unfocus area.

[Configuration of display device]
As shown in FIG. 13, the display device 200 includes a display unit 110 and an image processing device 220 as a calculation unit.
The image processing apparatus 220 includes a focus state recognition unit 230 and a depth processing unit 240 as an outline emphasis processing unit and a blur processing unit, which are configured by various programs.
As shown in FIG. 14, the focus state recognition unit 230 includes a Hadamard transform unit 131, a horizontal / vertical band centroid calculation unit 232, a blur band adjustment unit 233, an edge enhancement band adjustment unit 234, and a high frequency component accumulation unit 132. A low-frequency component accumulating unit 133, a focus degree calculating unit 134, a block-by-block accumulating unit 135, and a polarity gain determining unit 136.

The Hadamard transform unit 131 performs a Hadamard transform process on one input image Pi for each band characteristic calculation block Bf, and converts the 64 frequency band components F obtained by this process into a horizontal / vertical band centroid calculation unit 232, The high frequency component accumulation unit 132 and the low frequency component accumulation unit 133 are output.

The horizontal / vertical band center-of-gravity calculation unit 232 normalizes 64 frequency band components by replacing DC band components with 0 in the frequency band components F obtained by Hadamard transform processing.
Specifically, the horizontal / vertical band centroid calculating unit 232 acquires the maximum value Max among the 64 frequency band components F, and multiplies the value obtained by dividing 4096 by the maximum value Max and the frequency band component F. 15 is calculated as a normalized frequency band component (referred to as a normalized frequency band component) as shown in FIG. Here, the numerical value in the square part of FIG. 15 represents the normalized frequency band component. In addition, since the square portion at the upper left corner is a direct current band, the component is replaced with 0.
Further, the horizontal / vertical band centroid calculation unit 232 calculates the horizontal band centroid value Jh and the vertical band centroid value Jv based on the normalized frequency band component, and supplies them to the blur band adjustment unit 233 and the contour enhancement band adjustment unit 234. Output.
Specifically, when calculating the horizontal band centroid value Jh, the horizontal / vertical band centroid calculating unit 232 calculates a vertical accumulated value obtained by accumulating components arranged in the vertical direction, as shown in FIG. Then, 61944 obtained by accumulating these vertical accumulated values is calculated as an overall accumulated value. Further, the vertical multiplication value is calculated by multiplying the vertical cumulative value of the horizontal band Tth (T is 1 to 8) by (T-1), and 145541 obtained by accumulating these vertical multiplication values is used as the vertical multiplication cumulative value. calculate. Then, 2.35 obtained by dividing the vertical multiplication cumulative value by the total cumulative value is calculated as the horizontal band centroid value Jh.
Similarly, the horizontal / vertical band centroid calculating unit 232 calculates 2.49 obtained by dividing the horizontal multiplication accumulated value obtained by accumulating the horizontal multiplication value by the total accumulated value obtained by accumulating the horizontal accumulated value as the vertical band centroid value Jv. To do.

Based on the horizontal band centroid value Jh and the vertical band centroid value Jv, the blur band adjusting unit 233 blurs the unsharp horizontal band Ub1, the blur unsharp vertical band Ub2, and the blur band for each band characteristic calculation block Bf. A filter horizontal band Ub3 and a blurring filter vertical band Ub4 are set.
Specifically, when the level of each horizontal band of the input signal Vi is set as shown in FIG. 16, the blur band adjusting unit 233 sets the horizontal band centroid to the maximum horizontal band frequency (half the sampling frequency) Fmax. A value obtained by multiplying by the value Jh and dividing by 7 is calculated as the horizontal band centroid frequency. Then, frequencies lower than the horizontal band centroid frequency by a predetermined value are set as the blurring unsharp horizontal band Ub1 and the blurring filter horizontal band Ub3.
Similarly, the blur band adjusting unit 233 calculates a vertical band centroid frequency by multiplying the vertical band maximum frequency Fmax by the vertical band centroid value Jv and dividing the result by 7 as a vertical band centroid frequency. The frequency lower by the value is set as the blurring unsharp vertical band Ub2 and the blurring filter vertical band Ub4.
Then, the blurring band adjustment unit 233 sends the blurring unsharp horizontal band Ub1 and blurring unsharp vertical band Ub2 to the blurring filter horizontal band Ub3 and blurring filter vertical to the unsharp processing unit 242 (to be described later) of the depth processing unit 240. The band Ub4 is output to the later-described filter processing unit 241 of the depth processing unit 240, respectively.

The contour emphasis band adjusting unit 234, based on the horizontal band centroid value Jh and the vertical band centroid value Jv, for each band characteristic calculation block Bf, the unsharp horizontal band Ur1 during contour emphasis and the unsharp vertical band Ur2 during contour emphasis. The filter-enhanced filter horizontal band Ur3 and the filter-enhanced filter vertical band Ur4 are set.
Specifically, when the frequency level of the input signal Vi is set as shown in FIG. 16, the contour enhancement band adjustment unit 234 calculates the horizontal band centroid frequency in the same manner as the blur band adjustment unit 233, A frequency higher than the horizontal band centroid frequency by a predetermined value is set as the unsharp horizontal band Ur1 during contour emphasis. The unsharp horizontal band Ur1 at the time of contour emphasis is set so that the horizontal band centroid frequency corresponds to the centroid between the filter horizontal band Ub3 at the time of blurring and the unsharp horizontal band Ur1 at the time of contour emphasis. Further, a frequency higher than the unsharp horizontal band Ur1 at the time of contour emphasis by a predetermined value is set as the filter horizontal band Ur3 at the time of contour emphasis.
Similarly, the contour enhancement band adjustment unit 234 sets a frequency higher than the vertical band centroid frequency by a predetermined value as the unsharp vertical band Ur2 during contour enhancement. The unsharp vertical band Ur2 at the time of contour emphasis is set such that the vertical band centroid frequency is a frequency corresponding to the centroid of the filter vertical band Ub4 at the time of blurring and the unsharp vertical band Ur2 at the time of contour emphasis. Further, a frequency higher than the unsharp vertical band Ur2 at the time of contour enhancement by a predetermined value is set as the filter vertical band Ur4 at the time of contour emphasis.
Then, the contour emphasis band adjusting unit 234 sends the unsharp horizontal band Ur1 during contour emphasis and the unsharp vertical band Ur2 during contour emphasis to the unsharp processing unit 242, the filter emphasis horizontal band Ur3 and the contour emphasis filter vertical band Ur4. Are output to the filter processing unit 241.

The polarity gain determining unit 136 supplies the polarity information Km generated by the polarity determining unit 136C to the filter processing unit 241 and the unsharp processing unit 242, and the gain corresponding value Gm set by the gain corresponding value setting unit 136D to be described later of the depth processing unit 240. Output to the multiplication unit 244.

As illustrated in FIG. 13, the depth processing unit 240 includes a filter processing unit 241, an unsharp processing unit 242, a subtraction unit 243, a multiplication unit 244, and an addition unit 245.
The filter processing unit 241 acquires the input signal Vi. Further, for each band characteristic calculation block Bf of the input signal Vi, a blurring filter horizontal band Ub3, a blurring filter vertical band Ub4, a contour emphasis filter horizontal band Ur3, and a contour emphasis filter vertical band Ur4 are obtained. To do. Further, the polarity information Km for each set unit block Bs of the input signal Vi is acquired.
When the filter processing unit 241 recognizes that the band characteristic calculation block Bf is the focus area Af based on the polarity information Km, as shown in FIG. A filtered post-contour emphasis horizontal component Sx1 from which components above the filter horizontal band Ur3 have been removed, and a post-filter outline emphasis vertical component Sx2 from which components above the filter emphasis filter vertical band Ur4 have been removed from the vertical component of the band characteristic calculation block Bf; Is output to the subtracting unit 243 and the adding unit 245.
Further, when the filter processing unit 241 recognizes that the band characteristic calculation block Bf is the unfocus area Au, as shown in FIG. 17, a component equal to or higher than the blurring filter horizontal band Ub3 is calculated from the horizontal component of the band characteristic calculation block Bf. The filtered post-blurring horizontal component Sx3 and the post-filtering blur vertical component Sx4 obtained by removing components equal to or higher than the blurring filter vertical band Ub4 from the vertical component of the band characteristic calculation block Bf are output to the subtracting unit 243 and the adding unit 245. To do.

The unsharp processing unit 242 includes the unsharp unsharp horizontal band Ub1, the unsharp unsharp vertical band Ub2, and the unsharp unsharp horizontal band for each of the input signal Vi and the band characteristic calculation block Bf of the input signal Vi. Ur1, the unsharp vertical band Ur2 during edge enhancement, and the polarity information Km for each set unit block Bs are acquired.
Then, when the unsharp processing unit 242 recognizes that the band characteristic calculation block Bf is the focus area Af based on the polarity information Km, as shown in FIG. 17, the unsharpness at the time of contour emphasis is obtained from the horizontal component of the band characteristic calculation block Bf. A post-mask contour enhancement horizontal component Sy1 from which components above the horizontal band Ur1 are removed, and a post-mask contour enhancement vertical component Sy2 from which components above the unsharp vertical band Ur2 during contour enhancement are removed from the vertical components of the band characteristic calculation block Bf; Is output to the subtraction unit 243.
Further, when the unsharp processing unit 242 recognizes that the band characteristic calculation block Bf is the unfocus area Au, as shown in FIG. 17, the unsharpening unsharp horizontal band Ub1 or more from the horizontal component of the band characteristic calculation block Bf is obtained. The post-mask blurring horizontal component Sy3 from which the component is removed and the post-mask blurring vertical component Sy4 from which the component equal to or greater than the blurring unsharp vertical band Ub2 is removed from the vertical component of the band characteristic calculation block Bf are output to the subtracting unit 243.

As shown in FIG. 17, the subtraction unit 243 subtracts the post-subtraction contour enhancement horizontal component Sx1 from the unsharp processing unit 242 from the post-filter contour enhancement horizontal component Sx1 from the filter processing unit 241. Sg1 is calculated and output to the multiplication unit 244. Further, the subtracting unit 243 calculates a post-subtraction contour emphasizing vertical component Sg2 obtained by subtracting the post-mask contour emphasizing vertical component Sy2 from the post-filtering contour emphasizing vertical component Sx2, and outputs it to the multiplying unit 244. Here, the post-filter contour enhancement horizontal component Sx1 and the post-mask contour enhancement horizontal component Sy1 are different, and the post-filter contour enhancement vertical component Sx2 and the post-mask contour enhancement vertical component Sy2 are different. Also, the post-subtraction contour emphasis vertical component Sg2 never becomes zero.
The subtractor 243 subtracts the post-mask blur horizontal component Sy3 from the post-filter blur horizontal component Sx3, and subtracts the post-mask blur vertical component Sy4 from the post-filter blur vertical component Sx4. Here, the post-filter blur horizontal component Sx3 and the post-mask blur horizontal component Sy3 are the same, and the post-filter blur vertical component Sx4 and the post-mask blur vertical component Sy4 are the same. To the multiplication unit 244.

The multiplication unit 244 outputs the gain-adjusted horizontal component Sp1 obtained by multiplying the post-subtraction contour emphasizing horizontal component Sg1 and the gain corresponding value Gm of the band characteristic calculation block Bf to the addition unit 245. Also, the gain-adjusted vertical component Sp2 obtained by multiplying the post-subtraction contour emphasizing vertical component Sg2 and the gain corresponding value Gm is output to the adder 245. That is, the multiplying unit 244 outputs a gain-adjusted horizontal component Sp1 and a gain-adjusted vertical component Sp2 corresponding to the band characteristic calculation block Bf of the focus area Af.
Further, since the component corresponding to the band characteristic calculation block Bf of the unfocused area Au from the subtracting unit 243 is 0, the multiplying unit 244 indicates that the result obtained by multiplying this component by the gain corresponding value Gm is 0. Output to.

The adder 245 acquires the components Sx1, Sx2, Sx3, and Sx4 from the filter processor 241. Then, the post-filter contour emphasis horizontal component Sx1 and the gain-adjusted horizontal component Sp1 corresponding to the band characteristic calculation block Bf of the focus area Af are added, and the post-filter contour emphasis vertical component Sx2 and the gain-adjusted vertical component Sp2 are obtained. to add.
Further, the adding unit 245 adds the post-filtering blurred horizontal component Sx3 corresponding to the band characteristic calculation block Bf of the unfocused area Au and 0 which is the output result from the multiplying unit 244, and the post-filtered blurred vertical component Sx4 and 0 And add.
Then, based on the above-described addition result of each band characteristic calculation block Bf, an output signal Vo of the output image Po is generated and output to the display unit 110.
Here, in the focus area Af, the horizontal component Sp1 after gain adjustment and the vertical component Sp2 after gain adjustment are not zero. For this reason, the horizontal and vertical components of the focus area Af in the output image Po are larger than the components Sx1 and Sx2, that is, larger than those of the input image Pi, and the outline of the focus area Af in the output image Po is emphasized. Is done.
Further, in the unfocus area Au, the above-described addition result becomes the post-filter blurring horizontal component Sx3 and the post-filter blurring vertical component Sx4. For this reason, the horizontal and vertical components of the unfocus area Au in the output image Po are smaller than those in the input image Pi, and the unfocus area Au in the output image Po is blurred.

[Effects of Second Embodiment]
As described above, in the second embodiment, the following operational effects can be achieved in addition to (1) of the first embodiment.

(5) It is possible to always perform optimum enhancement processing and blurring processing by emphasizing according to the pattern of the input image Pi or changing the frequency band to be blurred.

[Third Embodiment]
Next, 3rd Embodiment which concerns on this invention is described based on drawing.
In addition, about the structure same as 1st, 2nd embodiment, the same name and code | symbol are attached | subjected and description is abbreviate | omitted or simplified.
FIG. 18 is a block diagram illustrating a schematic configuration of the display device.

[Configuration of display device]
As shown in FIG. 18, an image processing device 320 as a calculation unit of the display device 300 includes a face detection unit 321 and a gain weighting unit 322 as a face correspondence correction control unit in the image processing device 120 of the first embodiment. It has the provided structure.
The face detection unit 321 detects a human or animal face present in the input image Pi of the input signal Vi. Then, face information Gu having information regarding whether or not a face has been detected and information regarding its position when it has been detected is output to gain weighting section 322.

The gain weighting unit 322 acquires the linear interpolation gain corresponding value Gs from the inter-block linear interpolation unit 150 and the face information Gu from the face detection unit 321. Then, based on the face information Gu, it is determined whether or not a face has been detected from the input image Pi. If it is determined that a face has been detected, the first face is set to the linear interpolation gain corresponding value Gs of the pixel Q corresponding to this face. A weighting gain corresponding value Gk as face corresponding correction information multiplied by a correction value (for example, less than 1) is calculated, and a second face larger than the first face correction value is added to the linear interpolation gain corresponding value Gs of the pixels Q other than the face. A weighting gain corresponding value Gk multiplied by a correction value (for example, 1) is calculated. If it is determined that no face is detected, a weighting gain corresponding value Gk obtained by multiplying the linear interpolation gain corresponding value Gs of all the pixels Q by the second face correction value is calculated. That is, the correction is performed so that the value of the linear interpolation gain corresponding value Gs of the pixel Q corresponding to the face is smaller than that in the case where it is not detected. Then, the gain weighting unit 322 outputs the weighting gain corresponding value Gk to the gain adjustment unit 163.
Note that a weighting gain corresponding value Gk obtained by subtracting a predetermined value from the linear interpolation gain corresponding value Gs of only the pixel Q where the face is detected is calculated, or the linear interpolation gain corresponding value Gs of only the pixel Q where the face is not detected is calculated. Even if correction is made to reduce the value of the linear interpolation gain corresponding value Gs of the pixel Q corresponding to the face compared to the case where it is not detected by calculating the weighting gain corresponding value Gk to which a predetermined value is added. Good.

The gain adjustment unit 163 sets a value obtained by dividing the weighted gain corresponding value Gk of the predetermined pixel Q by 64 as a gain value, and a gain adjustment high frequency component based on the gain value and the polarity adjustment high frequency component Vk Vg is output to the adder 164.
As a result of the processing by the adding unit 164, the gain adjustment high-frequency component Vg smaller than the pixel Q not corresponding to the face in the focus area Af is added to the high-frequency component Vh in the pixel Q corresponding to the face in the focus area Af. For this reason, the degree of contour emphasis on the face in the output image Po is lower than that other than the face.
In the pixel Q corresponding to the face in the unfocus area Au, the gain adjustment high-frequency component Vg smaller than the pixel Q not corresponding to the face in the unfocus area Au is subtracted from the high-frequency component Vh, so the face in the output image Po The degree of blurring with respect to is lower than that other than the face.

[Effects of Third Embodiment]
As described above, in the third embodiment, in addition to the same functions and effects as the first embodiment (1) to (4), the following functions and effects can be achieved.

(6) The image processing device 320 makes the degree of the contour enhancement processing for the face in the focus area Af of the input image Pi lower than the degree of the contour enhancement processing for the part other than the face in the focus area Af. Furthermore, the degree of blurring processing for the face in the unfocused area Au is set lower than the degree of blurring processing for the part other than the face in the unfocused area Au.
Here, in general, humans have a higher visibility to a face such as a person than a visibility to a portion other than the face. For this reason, if an excessive outline emphasis process or blurring process is performed on the face, an output image Po with a sense of incongruity will be generated.
On the other hand, by reducing the degree of contour enhancement processing and blurring processing on the face as compared with the portions other than the face, it is possible to enhance the sense of depth while eliminating the uncomfortable feeling.

[Fourth Embodiment]
Next, 4th Embodiment which concerns on this invention is described based on drawing.
In addition, about the structure same as 1st, 2nd embodiment, the same name and code | symbol are attached | subjected and description is abbreviate | omitted or simplified.
FIG. 19 is a block diagram illustrating a schematic configuration of the focus state recognition unit. FIG. 20 is a schematic diagram showing first to fourth regions in the input image.

[Configuration of display device]
As shown in FIGS. 1 and 19, an image processing device 420 as a calculation unit of the display device 400 includes a screen position weight as a position correspondence correction control unit in the focus state recognition unit 130 of the image processing device 120 of the first embodiment. A focus state recognizing unit 430 including an adding unit 437 is provided.

The screen position weight addition unit 437 acquires the focus degree Du for each band characteristic calculation block Bf from the focus degree calculation unit 134. Then, as shown in FIG. 20, a horizontally long elliptical first region R1 located at the center of the input image Pi, a ring-shaped second region R2 surrounding the first region R1, and a ring shape surrounding the second region R2. The third region R3 and the ring-shaped fourth region R4 surrounding the third region R3 are recognized.
Then, the screen position weight addition unit 437 calculates a weight addition focus degree Dg as position correspondence correction information obtained by multiplying the pixel Q of the first region R1 by a first position correction value (for example, 1). Further, the weight added focus degree Dg obtained by multiplying the pixel Q in the second region R2 by the second position correction value smaller than the first position correction value, and the pixel Q in the third region R3 is smaller than the second position correction value. A weight-added focus degree Dg obtained by multiplying the three-position correction value and a weight-added focus degree Dg obtained by multiplying the pixel Q in the fourth region R4 by a fourth position correction value smaller than the third position correction value are calculated.
Then, the screen position weight addition unit 437 outputs the weight addition focus degree Dg to the block accumulation unit 135.

The block accumulation unit 135 outputs the block-specific focus degree Ds obtained by accumulating the weight-added focus degree Dg for each set unit block Bs to the polarity gain determination unit 136. The polarity gain determination unit 136 sets the polarity information Km and the gain corresponding value Gm for each set unit block Bs based on the block-specific focus degree Ds, and outputs it to the time constant processing unit 140. When the polarity adjustment high frequency component Vk of the first region R1 to the fourth region R4 has the same value, the gain adjustment unit 163 outputs a gain adjustment high frequency component Vg that is smaller as the block-specific focus degree Ds is smaller to the addition unit 164. That is, the gain adjustment high frequency component Vg is the smallest in the fourth region R4 and the largest in the first region R1.
When the polarity adjustment high-frequency component Vk in the first region R1 to the fourth region R4 has the same value by the processing in the adding unit 164, the gain adjustment high-frequency component Vg that is smaller in the pixel Q farther from the center of the input image Pi Added to component Vh. For this reason, the degree of contour emphasis in the output image Po decreases as the distance from the center increases.
Also, when the polarity adjustment high frequency component Vk in the first region R1 to the fourth region R4 has the same value, the gain adjustment high frequency component Vg that is smaller as the pixel Q is farther from the center of the input image Pi is subtracted from the high frequency component Vh. The degree of blur in the image Po decreases as the distance from the center increases.

[Effects of Fourth Embodiment]
As described above, in the fourth embodiment, in addition to the same functions and effects as in (1) to (4) of the first embodiment, the following functions and effects can be achieved.

(7) The image processing apparatus 420 decreases the degree of the contour enhancement process as the distance from the center of the input image Pi increases.
Here, in general, the focus area Af is often the center of the input image Pi.
For this reason, by increasing the degree of enhancement processing at the center of the input image Pi, it is possible to prevent erroneous detection of a portion that emphasizes an edge or a portion that is blurred, and the detection accuracy can be increased.

[Modification of Embodiment]
Note that the present invention is not limited to the first and second embodiments described above, and includes modifications shown below within a range in which the object of the present invention can be achieved.

That is, as shown in FIG. 1, display devices 500 and 600 including image processing devices 520 and 620 may be used instead of the image processing device 120 of the first embodiment. The image processing apparatuses 520 and 620 have a configuration in which focus

state recognition units

530 and 630 are provided instead of the focus state recognition unit 130.

As shown in FIG. 21, the focus state recognition unit 530 of the display device 500 has the same configuration as the focus state recognition unit 130 of the first embodiment except that a Fourier transform unit 531 is provided instead of the Hadamard transform unit 131. have.
Based on the input signal Vi of the input image Pi from the image signal output unit 10, the Fourier transform unit 531 performs a Fourier transform process for each band characteristic calculation block Bf as shown in FIG. 5, for example, and the frequency of the input signal Vi. The band component F is calculated. Here, in the case of Hadamard transform processing, a frequency band component F having only positive frequency components is calculated. In the case of Fourier transform processing, a frequency band component F having positive and negative frequency components is calculated. Note that the number of pixels constituting the band characteristic calculation block Bf may be different from that of the first embodiment.
Then, the Fourier transform unit 531 outputs the frequency band component F to the high frequency component accumulation unit 132 and the low frequency component accumulation unit 133.
The high frequency component accumulating unit 132 and the low frequency component accumulating unit 133 specify only the positive frequency component of the frequency band component F. Then, the high frequency component accumulating unit 132 recognizes, as the high frequency component Fh, a component whose frequency band is equal to or higher than a predetermined threshold value among the specified frequency components, and the high frequency component Fh is determined for each band characteristic calculation block Bf. The accumulated high frequency component accumulated value Sh is output to the focus degree calculation unit 134. Similarly to the high frequency component accumulating unit 132, the low frequency component accumulating unit 133 recognizes a component less than a preset threshold value as the low frequency component Fl, and uses the low frequency component accumulated value Sl as a focus degree calculating unit. To 134.
Even if it is such a structure, the effect similar to the said 1st Embodiment can be anticipated.

Further, as shown in FIG. 22, the focus state recognition unit 630 of the display device 600 includes a high frequency edge detection unit 631, a low frequency edge detection unit 632, a focus degree calculation unit 134, a block-by-block accumulation unit 135, A polarity gain determination unit 136.
The high frequency edge detection unit 631 is an HPF (High Pass Filter) or the like, and detects a high frequency component Fh as shown in FIG. Further, the low frequency edge detection unit 632 is an LPF (low pass filter) or the like, and detects a low frequency component Fl as shown in FIG. Here, the high frequency edge detection unit 631 and the low frequency edge detection unit 632 detect the positive and negative high frequency components Fh and the positive and negative low frequency components Fl, respectively.
Then, the high frequency edge detection unit 631 and the low frequency edge detection unit 632 calculate absolute values of the high frequency component Fh and the low frequency component Fl, and output the absolute values to the focus degree calculation unit 134.
Even if it is such a structure, the effect similar to the said 1st Embodiment can be anticipated.

For example, in the first embodiment, at least one of the time constant processing unit 140 and the inter-block linear interpolation unit 150 may not be provided, or the time constant processing based on the scene change may not be performed.
The time constant process and the linear interpolation process may be performed only for the degree of the contour enhancement process or the blur process.
Furthermore, the degree of adjustment for only the outline enhancement process or the blurring process may be performed based on the face detection.

The configuration in which the image processing apparatus of the present invention is applied to a display device has been exemplified.

In addition, each function described above is constructed as a program, but it may be configured by hardware such as a circuit board or an element such as one IC (Integrated Circuit), and can be used in any form. In addition, by adopting a configuration that allows reading from a program or a separate recording medium, handling is easy, and usage can be easily expanded.

In addition, the specific structure and procedure for carrying out the present invention can be appropriately changed to other structures and the like within a range in which the object of the present invention can be achieved.

[Effect of the embodiment]
As described above, in the above-described embodiment, the image processing device 120 of the display device 100 recognizes the focus area Af and the unfocus area Au in the input image Pi, and performs contour enhancement processing on the focus area Af. The output image Po is generated by performing the blurring process on the unfocus area Au.
For this reason, it is possible to effectively enhance the sense of depth in the output image Po by simultaneously performing the contour enhancement process and the blurring process on the input image Pi.

The present invention can be used as an image processing device, a display device, an image processing method, a program thereof, and a recording medium on which the program is recorded.

Claims

A focus state recognition unit for recognizing a focus area in which an object is in focus in an input image and an unfocus area in which the object is not in focus;
A depth processing unit that performs an edge enhancement process on the recognized focus area and generates an output image in which a blur process is performed on the unfocus area;
An image processing apparatus comprising:
The image processing apparatus according to claim 1.
A face detection unit for detecting that the subject includes at least one face of a person and an animal;
When the focus area includes a face, processing for reducing the degree of contour enhancement for the face lower than the degree of contour enhancement for a portion other than the face of the focus area, and the unfocus area includes a face. Face correspondence correction control for generating face correspondence correction information indicating that at least one of the processing of making the degree of blurring of the face lower than the degree of blurring of a part other than the face of the unfocused region is performed. And comprising
The depth processing unit generates an output image in which at least one of the processes based on the face correspondence correction information is performed.
The image processing apparatus according to claim 1.
A position-corresponding correction control unit that generates position-corresponding correction information for processing to increase the degree of contour emphasis as the part of the focus area that is farther from the center of the input image comprises:
The depth processing unit generates an output image that has been subjected to the contour enhancement processing based on the position correspondence correction information.
The image processing apparatus according to any one of claims 1 to 3,
A time constant processing unit that performs time constant processing for at least one of the degree of edge enhancement and the degree of blurring of an input image that is configured and continuously input as a moving image;
The depth processing unit generates an output image that has undergone processing based on the degree of time constant processing.
The image processing apparatus according to claim 4.
The time constant processing unit generates an output image that does not perform the time constant processing when it is determined that the continuously input image corresponds to a scene change of a moving image, and determines that it does not correspond to the scene change. In this case, an image processing apparatus that generates an output image that has been subjected to the time constant processing.
The image processing apparatus according to any one of claims 1 to 5,
A linear interpolation unit that performs linear interpolation processing on at least one of the degree of edge enhancement and the degree of blurring of a predetermined adjacent portion in the input image;
The depth processing unit generates an output image subjected to processing based on the degree of linear interpolation processing.
An image processing apparatus according to any one of claims 1 to 6,
A display unit for displaying an output image generated by the image processing apparatus;
A display device comprising:
An image processing method for generating an output image obtained by performing predetermined processing on an input image by an arithmetic means,
The computing means is
A focus state recognition step for recognizing a focus area in which the subject in the input image is in focus and an unfocus area in which the subject is not in focus;
A depth processing step for generating an output image in which a contour enhancement process is performed on the recognized focus area and a blur process is performed on the unfocus area;
The image processing method characterized by implementing.
An image processing program for causing an arithmetic means to execute the image processing method according to claim 8.
An image processing program for causing a calculation means to function as the image processing apparatus according to any one of claims 1 to 6.
11. A recording medium on which an image processing program is recorded, wherein the image processing program according to claim 9 or 10 is recorded so as to be readable by an arithmetic means.