JP2008103785A - Outline emphasizing circuit, outline emphasizing method, imaging device, and view finder - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To excellently emphasize an outline without emphasizing noise components. <P>SOLUTION: An outline emphasizing circuit has an outline emphasized signal generation section comprising a horizontal HPF 105, a vertical HPF 106, and an arithmetic unit 109, and an outline emphasized signal generation section comprising a right-downside oblique HPF 107, a left-downside HPF 108, and an arithmetic unit 110. Each of the generation section extracts high-frequency components of two mutually orthogonal directions from an input video signal Vin and subtract and subtracts a high-frequency component smaller in absolute value from the high-frequency component larger in absolute value to obtain an outline emphasized signals Shv or Srl. An arithmetic unit 111 outputs an outline emphasized signal larger in absolute value between the outline emphasized signals Shv and Srl as an outline emphasized signal so as to be used finally. The outline emphasized signals Shv and Srl both have small values at isolation points of noise and the signal So becomes small. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a contour enhancement circuit, a contour enhancement method, an imaging device, and a viewfinder. Specifically, the present invention extracts the high frequency components in the first direction and the second direction orthogonal to each other from the input video signal, and subtracts the smaller absolute value from the larger absolute value to obtain the contour enhancement signal. Thus, the present invention relates to a contour emphasizing circuit and the like that can perform contour enhancement satisfactorily without enhancing noise components.

  2. Description of the Related Art Conventionally, an imaging apparatus (camera) performs contour enhancement processing for enhancing a contour portion of an image on a video signal output from an imaging unit. For example, as described in Patent Document 1 and the like, the contour enhancement process is performed by extracting horizontal and vertical high frequency components from an input video signal and adding these high frequency components to the input video signal. Done.

  FIG. 11 shows a configuration example of a conventional contour emphasizing circuit 200.

  The contour emphasis circuit 200 includes an input terminal 201, line memories 202 and 203, a horizontal high-pass filter (horizontal HPF) 205, a vertical high-pass filter (vertical HPF) 206, gain adjusters 207 and 208, and an adder 209. 210, an output terminal 211, and a delay circuit 212.

  The input terminal 201 is a terminal for inputting an input video signal Vin that is a target for performing contour enhancement processing. Each of the line memories 202 and 203 functions as a delay line that delays the video signal by one horizontal period. The horizontal HPF 205 uses, for each pixel, signals of three pixels n (−1,0), n (0,0), and n (1,0) arranged in the horizontal direction around the pixel n (0,0). Thus, a high frequency component in the horizontal direction is extracted. FIG. 12A shows an example of the filter coefficient of the horizontal HPF 205.

  The vertical HPF 206 uses, for each pixel, signals n (0, -1), n (0,0), n (0,1) of three pixels arranged in the vertical direction around the pixel n (0,0). Thus, the high frequency component in the vertical direction is extracted. FIG. 12B shows an example of the filter coefficient of the vertical HPF 206. The gain adjuster 207 adjusts the gain of the high frequency component in the horizontal direction extracted by the horizontal HPF 205 by multiplying by a coefficient between 0 and 1. The gain adjuster 208 adjusts the gain of the high frequency component in the vertical direction extracted by the vertical HPF 206 by multiplying by a coefficient between 0 and 1. These gain adjusters 207 and 208 are for adjusting the degree of edge enhancement, and the degree of enhancement increases as the coefficient approaches 1.

  The adder 209 adds the high frequency component in the horizontal direction whose gain is adjusted by the gain adjuster 207 to the input video signal Vin. The adder 210 adds the high frequency component in the vertical direction whose gain is adjusted by the gain adjuster 208 to the video signal output from the adder 209. The output terminal 211 outputs the video signal output from the adder 210 as the output video signal Vout. The delay circuit 212 adjusts the timing of the input video signal Vin and the horizontal and vertical high frequency components added to the input video signal Vin.

  The operation of the contour emphasis circuit 200 shown in FIG. 11 will be described. The input video signal Vin is input to the input terminal 201. The input video signal Vin is supplied to the adder 209 via the delay circuit 212. The input video signal Vin is delayed by one horizontal period in the line memory 202 and input to the horizontal HPF 205. The horizontal HPF 205 extracts a high frequency component in the horizontal direction for each pixel based on the input video signal. The high frequency component in the horizontal direction is gain-adjusted by the gain adjuster 207 and supplied to the adder 209. The adder 209 adds the high frequency component in the horizontal direction whose gain is adjusted to the input video signal Vin.

  The video signal output from the adder 209 is input to the adder 210. The input video signal Vin is directly input to the vertical HPF 206 and is input to the vertical HPF 206 after being delayed by one horizontal period in the line memory 202. The video signal output from the line memory 202 is further delayed by one horizontal period in the line memory 203 and input to the vertical HPF 206. In the vertical HPF 206, high frequency components in the vertical direction are extracted for each pixel using the input three lines of video signals. The high frequency component in the vertical direction is gain-adjusted by the gain adjuster 208 and supplied to the adder 210.

In the adder 210, the high frequency component in the vertical direction whose gain is adjusted is added to the output video signal of the adder 209. The video signal obtained by adding the horizontal and vertical high frequency components (contour emphasis signal) to the output video signal of the adder 210, that is, the input video signal Vin, is output to the output terminal 211 as the output video signal Vout. The
JP-A-5-252421

  In the contour emphasis circuit 200 shown in FIG. 11, high frequency components in the horizontal direction and vertical direction are extracted from the input video signal Vin, and the high frequency components in each direction are added to the input video signal Vin as contour emphasis signals. In this case, an edge extending in the horizontal direction or the vertical direction can only obtain a high frequency component in one direction, but a bi-directional high frequency component can be obtained at an isolated point of the noise. The problem arises that is emphasized.

  The problem that noise is emphasized rather than the edge occurs similarly in an edge emphasis circuit in which an edge emphasis signal is obtained using a spatial frequency filter such as a Laplacian filter, although not shown.

  An object of the present invention is to make it possible to perform edge enhancement satisfactorily without enhancing noise components.

The concept of this invention is
A first high pass filter for extracting a high frequency component in a first direction from an input video signal;
A second high-pass filter that extracts a high-frequency component in a second direction orthogonal to the first direction from the input video signal;
Using the high frequency component in the first direction extracted by the first high-pass filter and the high frequency component in the second direction extracted by the second high-pass filter, from the one having a larger absolute value An arithmetic unit for subtracting the smaller absolute value to obtain an edge enhancement signal;
An outline emphasis circuit comprising: an addition unit that adds an outline enhancement signal obtained by the arithmetic unit to the input video signal to obtain an output video signal.

  This contour enhancement circuit is provided in, for example, an imaging apparatus (camera) or a viewfinder. In this contour emphasis circuit, high frequency components in two directions orthogonal to each other are extracted from the input video signal. That is, the first high-pass filter extracts a high-frequency component in the first direction from the input video signal. In the second high-pass filter, a high frequency component in a second direction orthogonal to the first direction is extracted from the input video signal. In the computing unit, the contour enhancement signal is obtained by subtracting the smaller absolute value from the larger absolute value. In the adder, the contour emphasis signal is added to the input video signal, and an output video signal subjected to the contour emphasis processing is obtained.

  For example, when the first direction is the horizontal direction and the second direction is the vertical direction, the contour emphasis signal obtained corresponding to the edge extending in the horizontal direction or the vertical direction has a large value. The contour enhancement signal obtained correspondingly has a small value. Therefore, according to the present invention, it is possible to favorably perform contour enhancement without enhancing noise components.

  For example, the calculation unit includes a gain adjuster that adjusts the gain of a high frequency component having a small absolute value. This gain adjuster adjusts the gain of the high frequency component with a small absolute value subtracted from the high frequency component with a large absolute value, and adjusts the effectiveness of the effect of not enhancing the noise component in the contour emphasizing circuit of the present invention. The

  Further, for example, a plurality of the above-described contour enhancement signal generation units may be provided. The first direction and the second direction in the plurality of generation units are made different from each other. In this case, among the contour emphasis signals generated by the plurality of generation units, the one having the largest absolute value is selected and added to the input video signal to obtain an output video signal. By adopting a configuration including a plurality of generation units in this way, the accuracy of contour emphasis can be increased. For example, as described above, the contour emphasis signal obtained at the edge in the oblique direction is a small value only in the contour emphasis signal generator having the first direction as the horizontal direction and the second direction as the vertical direction. The contour enhancement at the oblique edge is not sufficiently performed. For example, in the case where a contour emphasis signal generating unit that further includes a first direction as a lower right diagonal direction and a second direction as a left lower diagonal direction is provided, the generating unit obtains an edge emphasis obtained by an oblique edge. Since the signal has a large value, the contour enhancement at the oblique edge is sufficiently performed.

  According to the present invention, the high frequency components in the first direction and the second direction orthogonal to each other are extracted from the input video signal, and the contour enhancement signal is obtained by subtracting the smaller absolute value from the larger absolute value. Therefore, the edge enhancement can be favorably performed without enhancing the noise component.

  An embodiment of the present invention will be described. FIG. 1 shows a configuration of an imaging apparatus (video camera) 10 as an embodiment. The imaging device 10 includes a lens block (imaging lens system) 11, an imaging element 12, an analog signal processing unit 13, an A / D converter 14, an outline emphasizing unit 15, a digital signal processing unit 16, a transmission unit 17, and an EVF (Electronic View Finder) 18.

  The lens block 11 forms an optical image corresponding to the subject on the imaging surface of the imaging element 12. The imaging element 12 is an imaging element such as a CCD (Charged Coupled Device) or a CMOS (Complementary Metal-Oxide Semiconductor). The analog signal processing unit 13 performs various analog signal processes such as amplification processing, white / black balance adjustment, white / black shading correction, flare correction, and the like on the analog video signal output from the image sensor 12.

  The A / D converter 14 converts the video signal output from the analog signal processing unit 13 from an analog signal to a digital signal. The contour enhancement unit 15 performs contour enhancement processing on the video signal converted into a digital signal by the A / D converter 14. Details of the contour emphasizing unit 15 will be described later.

  The digital signal processing unit 16 performs various digital signal processing such as knee processing, gamma correction processing, and white / clip processing. For example, the transmission unit 17 performs modulation processing on the video signal output from the digital signal processing unit 16 to form an RF signal in a predetermined frequency band, and transmits the RF signal to the camera control unit (not shown). The viewfinder 18 includes a display such as an LCD (Liquid Crystal Display), and displays an image based on a video signal output from the digital signal processing unit 16.

  The operation of the imaging device 10 shown in FIG. 1 will be described. Image light from the subject is irradiated onto the image pickup surface of the image pickup device 12 via the lens block 11, and a subject image is formed on the image pickup surface. The imaging element 12 performs imaging in a state where a subject image is formed on the imaging surface. The video signal output from the image sensor 12 is supplied to the analog signal processing unit 13. The analog signal processing unit 13 performs various analog signal processing such as amplification processing, white / black balance adjustment, white / black shading correction, and flare correction on the video signal.

  The video signal output from the analog signal processing unit 13 is converted from an analog signal to a digital signal by the A / D converter 14 and supplied to the contour enhancement unit 15. The contour emphasizing unit 15 performs contour emphasis processing on the video signal. The video signal output from the contour enhancement unit 13 is supplied to the digital signal processing unit 16. The digital signal processing unit 16 performs various digital signal processing such as knee processing, gamma correction processing, and white / clip processing on the video signal.

  The video signal output from the digital signal processing unit 16 is transmitted to the camera control unit, for example, through the transmission unit 17. The video signal output from the digital signal processing unit 16 is supplied to the EVF 18. In the EVF 18, an image (imaging) based on the video signal supplied from the digital signal processing unit 16 is displayed on the display.

  The contour emphasizing unit 15 will be described. FIG. 2 shows a specific configuration of the contour emphasizing unit 15. The contour emphasizing unit 15 includes an input terminal 101, line memories 102 and 103, a horizontal high-pass filter (horizontal HPF) 105, a vertical high-pass filter (vertical HPF) 106, and a right-down diagonal high-pass filter (right-down diagonal HPF) 107. A left-downward slanting high-pass filter (left-downward sloping HPF) 108, arithmetic units 109 to 111, a gain adjuster 112, an adder 113, an output terminal 114, and a delay circuit 115.

  The input terminal 101 is a terminal for inputting a video signal output from the A / D converter 14, that is, an input video signal Vin to be subjected to contour enhancement processing. Each of the line memories 102 and 103 functions as a delay line that delays the video signal by one horizontal period. Line memories 102 and 103 are provided to simultaneously obtain video signals for three lines.

  The horizontal HPF 105 uses, for each pixel, signals of three pixels n (−1,0), n (0,0), and n (1,0) arranged in the horizontal direction around the pixel n (0,0). Thus, a high frequency component in the horizontal direction is extracted. FIG. 3A shows an example of the filter coefficient of the horizontal HPF 105. The vertical HPF 106 uses, for each pixel, signals n (0, -1), n (0,0), n (0,1) of three pixels arranged in the vertical direction around the pixel n (0,0). Thus, the high frequency component in the vertical direction is extracted. FIG. 3B shows an example of the filter coefficient of the vertical HPF 106. The horizontal direction and the vertical direction constitute a first direction and a second direction orthogonal to each other.

  The right-downward sloping HPF 107 has, for each pixel, three pixel signals n (-1, -1), n (0,0), n (1 , 1) is used to extract the high frequency component in the diagonally downward direction. FIG. 3C shows an example of the filter coefficient of the diagonally downward HPF 107. The left-downward sloping HPF 108 is, for each pixel, three-pixel signals n (1, -1), n (0,0), n (-1) arranged in the left-down slanting direction with the pixel n (0,0) as the center. , 1) is used to extract the high frequency component in the diagonally downward direction. FIG. 3D shows an example of the filter coefficient of the left-declining diagonal HPF 108. The diagonally downward right direction and the diagonally downward left direction also constitute a first direction and a second direction orthogonal to each other.

  The calculation unit 109 uses the horizontal high-frequency component Sh extracted by the horizontal HPF 105 and the vertical high-frequency component Sv extracted by the vertical HPF 106, and subtracts the smaller absolute value from the larger absolute value. To obtain the contour emphasis signal Shv. The horizontal HPF 105, the vertical HPF 106, and the calculation unit 109 constitute one contour enhancement signal generation unit.

  FIG. 4 shows the configuration of the calculation unit 109. The calculation unit 109 includes subtracters 121 and 122, gain adjusters 123 and 124, absolute value conversion units (ABS units) 125 and 126, a comparison unit 127, and a selector 128.

  The gain adjusters 123 and 124 adjust the gains of the high frequency components Sv and Sh by multiplying the coefficients α and β between 0 and 1, for example. The subtractor 121 subtracts the high frequency component αSv in the vertical direction whose gain is adjusted by the gain adjuster 123 from the high frequency component Sh in the horizontal direction. The subtracter 122 subtracts the high frequency component βSh in the horizontal direction whose gain is adjusted by the gain adjuster 124 from the high frequency component Sv in the vertical direction.

  The operation of the calculation unit 109 shown in FIG. 4 will be described. The high frequency component Sh in the horizontal direction extracted by the horizontal HPF 105 (see FIG. 2) is directly supplied to the subtractor 121, and the gain is adjusted by the gain adjuster 124 and then supplied to the subtractor 122. Further, the vertical high-frequency component Sv extracted by the vertical HPF 106 (see FIG. 2) is directly supplied to the subtractor 122, is gain-adjusted by the gain adjuster 123, and is supplied to the subtractor 121.

  The attenuator 121 obtains an output signal (Sh−αSv) by subtracting the vertical high frequency component αSv whose gain is adjusted from the horizontal high frequency component Sh, and this output signal (Sh−αSv) is selected by the selector. 128. The attenuator 122 subtracts the horizontal high frequency component βSh whose gain is adjusted from the vertical high frequency component Sv to obtain an output signal (Sv−βSh). This output signal (Sv−βSh) is a selector. 128.

  The horizontal high-frequency component Sh extracted by the horizontal HPF 105 (see FIG. 2) is supplied to the absolute value conversion unit 125, and the absolute value | Sh | is obtained. This absolute value | Sh | is supplied to the comparison unit 127. The high frequency component Sv in the vertical direction extracted by the vertical HPF 106 (see FIG. 2) is supplied to the absolute value converting unit 126, and the absolute value | Sv | is obtained. This absolute value | Sv | is supplied to the comparison unit 127.

  The comparison unit 127 compares the absolute value | Sh | with the absolute value | Sv |. The comparison result output from the comparison unit 127 is supplied to the selector 128 as a control signal. The selector 128 outputs the output signal (Sh−αSv) of the subtractor 121 as the contour emphasizing signal Shv when | Sh |> | Sv |, and the subtractor 122 when | Sh | <| Sv |. Output signal (Sv−βSh) is output as the contour emphasis signal Shv.

  As described above, the selector 128 uses the high frequency component Sh in the horizontal direction and the high frequency component Sv in the vertical direction, and contour enhancement generated by subtracting the smaller absolute value from the larger absolute value. A signal Shv is obtained. When viewed from the contour emphasis signal Shv, the gain adjusters 123 and 124 adjust the gain of the high frequency component having a small absolute value.

  Returning to FIG. 2, the calculation unit 110 uses the high-frequency component Sr in the right-down diagonal direction extracted by the right-down diagonal HPF 107 and the high-frequency component Sl in the left-down diagonal direction extracted by the left-down diagonal HPF 108. Then, the contour enhancement signal Srl is obtained by subtracting the smaller absolute value from the larger absolute value. Although not described in detail, the calculation unit 110 is also configured similarly to the calculation unit 109 (see FIG. 4) described above. The downward-sloping diagonal HPF 107, the downward-sloping diagonal HPF 108, the arithmetic unit 110mo, and one contour enhancement signal generation unit are configured.

  The calculation unit 111 selectively outputs the contour enhancement signal So having the largest absolute value among the contour enhancement signals Shv and Srl obtained by the calculation units 109 and 110. The calculation unit 111 constitutes a selection unit.

  FIG. 5 shows the configuration of the calculation unit 111. The calculation unit 111 includes absolute value conversion units (ABS units) 131 and 132, a comparison unit 133, and a selector 134. The absolute value converting unit 131 obtains the absolute value | Shv | of the contour emphasis signal Shv obtained by the calculating unit 109. The absolute value conversion unit 132 obtains the absolute value | Srl | of the contour emphasis signal Srl obtained by the calculation unit 110. The comparison unit 133 compares the absolute value | Shv | with the absolute value | Srl | and sends the comparison result to the selector 134.

  The selector 134 outputs the contour emphasis signal Shv from the arithmetic unit 109 or the contour emphasis signal Srl from the arithmetic unit 110 based on the comparison result obtained from the comparison unit 133. That is, the selector 134 outputs the contour emphasis signal Shv as the contour emphasis signal So to be used when | Shv |> | Srl |, and the contour emphasis signal when | Shv | <| Srl |. Srl is output as a contour emphasis signal So to be used.

  The operation of the calculation unit 111 shown in FIG. 5 will be described. The contour emphasis signals Shv and Srl obtained by the arithmetic units 109 and 110 (see FIG. 2) are supplied to the selector 134, respectively.

  The contour emphasis signal Shv obtained by the calculation unit 109 is supplied to the absolute value converting unit 131, and the absolute value | Shv | is obtained. This absolute value | Shv | is supplied to the comparison unit 133. The contour emphasis signal Srl obtained by the calculation unit 110 is supplied to the absolute value conversion unit 132, and the absolute value | Srl | is obtained. The absolute value | Srl | is supplied to the comparison unit 133.

  The comparing unit 133 compares the absolute value | Shv | with the absolute value | Srl |. The comparison result output from the comparison unit 133 is supplied to the selector 134 as a control signal. From the selector 134, when | Shv |> | Srl |, the contour emphasis signal Shv | obtained by the arithmetic unit 109 is outputted as the contour emphasis signal So to be finally used, and | Shv | <| Srl | In some cases, the contour emphasis signal Srl obtained by the calculation unit 110 is output as the contour emphasis signal So to be finally used.

  Returning to FIG. 2, the gain adjuster 112 adjusts the gain of the contour emphasis signal So output from the computing unit 111 by multiplying by a coefficient between 0 and 1. The gain adjuster 112 is for adjusting the degree of edge enhancement, and the degree of enhancement increases as the coefficient approaches 1. The adder 113 adds the contour emphasis signal gain-adjusted by the gain adjuster 112 to the input video signal Vin. The output terminal 114 outputs the video signal output from the adder 113 as the output video signal Vout. The delay circuit 115 adjusts the timing of the input video signal Vin and the contour enhancement signal added to the input video signal Vin.

  The operation of the contour emphasizing unit 15 shown in FIG. 2 will be described. The input video signal Vin is input to the input terminal 101. The input video signal Vin is supplied to the adder 113 via the delay circuit 115. The input video signal Vin is delayed by one horizontal period in the line memory 102 and input to the horizontal HPF 105. In the horizontal HPF 105, a high frequency component Sh in the horizontal direction is extracted for each pixel based on the input video signal.

  In addition, the input video signal Vin is directly input to the vertical HPF 106, the lower right diagonal HPF 107, and the lower left HPF 108, and is also delayed by one horizontal period in the line memory 102 and input to the vertical HPF 106, the lower right diagonal HPF 107, and the lower left HPF 108. Is done. Further, the video signal output from the line memory 102 is further delayed by one horizontal period in the line memory 103 and input to the vertical HPF 106, the right-downward sloping HPF 107, and the left-down HPF 108.

  In the vertical HPF 106, a high frequency component Sv in the vertical direction is extracted for each pixel by using the input three lines of video signals. In the downward-sloping diagonal HPF 107, the high-frequency component Sr in the downward-sloping diagonal direction is extracted for each pixel using the input three lines of video signals. In the lower left diagonal HPF 108, the high frequency components Sl in the lower left diagonal direction are extracted for each pixel using the input three lines of video signals.

  The horizontal high-frequency component Sh extracted by the horizontal HPF 105 and the vertical high-frequency component Sv extracted by the vertical HPF 106 are supplied to the calculation unit 109. The arithmetic unit 109 uses the high frequency components Sh and Sv to subtract the smaller absolute value from the larger absolute value to obtain the contour emphasis signal Shv (see FIG. 4).

  The high-frequency component Sr in the downward-sloping diagonal direction extracted by the downward-sloping diagonal HPF 107 and the high-frequency component Sl in the downward-sloping diagonal direction extracted by the left-down diagonal HPF 108 are supplied to the computing unit 110. In the calculation unit 110, as in the calculation unit 109 described above, the contour enhancement signal Srl is obtained by subtracting the smaller absolute value from the larger absolute value using the high frequency components Sr and Sl (FIG. 4). reference).

  The contour enhancement signal Shv obtained by the computing unit 109 and the contour enhancement signal Srl obtained by the computing unit 110 are supplied to the computing unit 111. From the contour enhancement signals Shv and Srl, the computation unit 111 outputs the contour enhancement signal So having the largest absolute value as the contour enhancement signal So to be finally used. The contour emphasis signal So is gain-adjusted by the gain adjuster 112 and supplied to the adder 113. The adder 113 adds the edge enhancement signal whose gain is adjusted to the input video signal Vin. The output video signal of the adder 113, that is, the video signal that has undergone the contour enhancement process is output to the output terminal 114 as the output video signal Vout.

  2 extracts high frequency components in two directions orthogonal to each other from the input video signal Vin, and subtracts the smaller absolute value from the larger absolute value to obtain the contour emphasizing signals Shv and Srl. The contour enhancement can be performed well without enhancing the noise component.

  For example, either one of the edge emphasis signals Shv and Srl obtained corresponding to the edge extending in the horizontal direction or the vertical direction takes a large value, but the edge emphasis signals Shv and Srl obtained corresponding to the isolated points of the noise are either. Is also a small value. Therefore, by using this contour emphasizing unit 15, contour enhancement can be performed satisfactorily without enhancing noise components.

  Here, a specific example will be described with reference to FIG. In FIG. 6, “Image 1” is an example of an oblique edge to be emphasized, “Image 2” is an example of a vertical edge to be emphasized, “Image 3” is an example of an isolated point of noise that is not to be emphasized, and “Image 4” is This is an example of a plane that you do not want to emphasize. In addition, the following description regarding an image is performed using the alphabet corresponding to each pixel position shown in the upper part of FIG.

  In the case of the four images in FIG. 6, there are two types of orthogonal components: def and beh, aei and ceg. FIG. 7 shows the result of applying a high-pass filter (HPF) with a filter coefficient of (−1, 2, −1) to each of def, beh, aei, and ceg for images 1 to 4.

From the results shown in FIG. 7, the result obtained by subtracting the smaller one of the two orthogonal HPF outputs corresponding to the processing of the arithmetic units 109 and 110 from the larger absolute value is as follows.
Image 1: def-beh = 0 aei-ceg = 8
Image 2: def-beh = 8 aei-ceg = 0
Image 3: def-beh = 0 aei-ceg = 0
Image 4: def-beh = 0 aei-ceg = 0

  If one of the two calculation results having the larger absolute value is selected, the image 1 and the image 2 are 8 and the image 3 and the image 4 are 0. Therefore, it can be seen that the edge to be emphasized is emphasized, but the isolated point of noise that is not desired to be emphasized is not emphasized.

  In addition, the contour emphasizing unit 15 shown in FIG. 2 includes a contour-enhancing signal generation unit composed of a horizontal HPF 105, a vertical HPF 106, and a computing unit 109, and a right-sloping diagonal HPF 107, a left-sloping diagonal HPF 108, and a computing unit 110. An edge emphasis signal generator is provided, and the edge emphasis signal Shv, Srl generated by these two edge emphasis signal generators is used as the edge emphasis signal So to be used finally. The accuracy of contour enhancement can be increased.

  For example, the contour emphasis signal Shv obtained at the edge in the diagonal direction is a small value only by the contour emphasis signal generator composed of the horizontal HPF 105, the vertical HPF 106, and the arithmetic unit 109, so that the contour emphasis at the diagonal edge is sufficient. Will not be done. However, the contour emphasizing unit 15 shown in FIG. 2 is further provided with a contour emphasizing signal generation unit composed of a right-declining oblique HPF 107, a left-declining oblique HPF 108, and a calculation unit 110. Since the obtained contour emphasis signal Srl has a large value, the contour emphasis at the oblique edge is sufficiently performed.

  In the contour emphasizing unit 15 shown in FIG. 2, the arithmetic units 109 and 110 are provided with gain adjusters 123 and 124 for adjusting the gain of the high frequency component having a small absolute value (see FIG. 4). . By adjusting the gain of the high frequency component having a small absolute value subtracted from the high frequency component having a large absolute value by the gain adjusters 123 and 124, the effect of not enhancing the noise component in the contour emphasizing unit 15 is obtained. You can adjust the condition.

  The gain in the gain adjusters 123 and 124 basically becomes less effective as 1 approaches a maximum of 0. Here, when the gain is set to 1 or more, the edge emphasis signal appears in the direction opposite to the level of the pixel signal at the isolated point of the noise, so that the effect of suppressing the noise can be obtained.

  In FIG. 1, the imaging apparatus 10 shows the one in which the contour emphasis unit 15 is arranged between the A / D converter 14 and the digital signal processing unit 16, but like the imaging apparatus 10 </ b> A shown in FIG. 8. It is also conceivable to arrange the contour emphasizing unit 15 inside the electronic viewfinder 18A. 8, parts corresponding to those in FIG. 1 are denoted by the same reference numerals, and detailed description thereof is omitted.

  In this case, the video signal output from the digital signal processing unit 16 and supplied to the electronic viewfinder 18A is subjected to contour enhancement processing by the contour enhancement unit 15 and then supplied to the display. On the display, an image whose outline is emphasized is displayed.

  Also, the contour emphasizing unit 15 shown in FIG. 2 includes a contour emphasizing signal generating unit composed of a horizontal HPF 105, a vertical HPF 106, and a computing unit 109, and a contour composed of a right-declining diagonal HPF 107, a left-declining diagonal HPF 108, and a computing unit 110. Although what is provided with the emphasis signal generation part was shown, what is equipped with the 1 or 3 or more outline emphasis signal generation part can be comprised similarly. In the case of including only one outline emphasis signal generation unit, the calculation unit 111 is not necessary. Further, by providing three or more contour emphasis signal generators, the accuracy of contour emphasis can be further increased.

  In the above-described embodiment, the high-pass filter having the filter coefficient of (−1, 2, −1) is used to extract the high frequency component. However, if the high-pass filter can determine the edge direction, All available. For example, a Sobel filter showing filter coefficients in FIG. 9 and a Prewitt filter showing filter coefficients in FIG. 10 can be used.

  The present invention can favorably perform contour enhancement without enhancing noise components, and can be applied to an imaging apparatus (video camera), a viewfinder, and the like.

It is a block diagram which shows the structure of an imaging device (video camera). It is a block diagram which shows the structure of an outline emphasis part. It is a figure which shows the filter coefficient of the high-pass filter of the horizontal, the vertical, the diagonal to the right, and the diagonal to the left used by the outline emphasis part. It is a block diagram which shows the structure of the calculating part which comprises an outline emphasis part. It is a block diagram which shows the structure of the calculating part which comprises an outline emphasis part. It is a figure which shows the example of an image for demonstrating an effect. It is a figure which shows the extraction result of the high frequency component of each direction in each image. It is a block diagram which shows the other structure of an imaging device (video camera). It is a figure which shows the filter coefficient of a Sobel filter. It is a figure which shows the filter coefficient of a Prewitt filter. It is a block diagram which shows the structure of the conventional outline emphasis circuit. It is a figure which shows the filter coefficient of the horizontal and vertical high-pass filter used with an outline emphasis circuit.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 10,10A ... Imaging device, 11 ... Lens block, 12 ... Image sensor, 13 ... Analog signal processing part, 14 ... A / D converter, 15 ... Outline emphasis part, DESCRIPTION OF SYMBOLS 16 ... Digital signal processing part, 17 ... Transmission part, 18, 18A ... Electronic viewfinder, 101 ... Input terminal, 102, 103 ... Line memory, 105 ... Horizontal high-pass filter, 106... Vertical high-pass filter, 107... Downward sloping high-pass filter, 108... Left-down slanting highpass filter, 109 to 111. 114, output terminal 115, delay circuit 121, 122 subtractor 123, 124 gain adjuster 125, 126 absolute value unit 127 Compare part, 128 ... selector, 131, 132 ... absolute value unit, 133 ... comparing unit, 134 ... selector

Claims (6)

A first high pass filter for extracting a high frequency component in a first direction from an input video signal;
A second high-pass filter that extracts a high-frequency component in a second direction orthogonal to the first direction from the input video signal;
Using the high frequency component in the first direction extracted by the first high-pass filter and the high frequency component in the second direction extracted by the second high-pass filter, from the one having a larger absolute value An arithmetic unit for subtracting the smaller absolute value to obtain an edge enhancement signal;
An outline emphasis circuit comprising: an adder that adds an outline emphasis signal obtained by the arithmetic unit to the input video signal to obtain an output video signal. The contour enhancement circuit according to claim 1, wherein the arithmetic unit includes a gain adjuster that adjusts a gain of a high frequency component having a small absolute value. A plurality of contour emphasis signal generators for generating an edge emphasis signal from the input video signal;
A selector that selectively outputs a contour enhancement signal having the largest absolute value from each of the contour enhancement signals generated by the plurality of contour enhancement signal generators;
An addition unit that adds an edge enhancement signal output from the selection unit to the input video signal to obtain an output video signal;
Each of the contour enhancement signal generation units includes a first high-pass filter that extracts a high frequency component in a first direction from the input video signal, and a second direction that is orthogonal to the first direction from the input video signal. A second high-pass filter that extracts a high-frequency component of the first direction, a high-frequency component in the first direction that is extracted by the first high-pass filter, and the second high-pass filter that is extracted by the second high-pass filter. A high-frequency component of the direction, using the high-frequency component of the second direction, and calculating a contour emphasis signal by subtracting the smaller absolute value from the larger absolute value,
The contour emphasizing circuit, wherein the first direction and the second direction in each contour emphasizing signal generator are different from each other. A first extraction step of extracting a high frequency component in a first direction from the input video signal;
A second extraction step of extracting a high frequency component in a second direction orthogonal to the first direction from the input video signal;
Using the high frequency component in the first direction extracted in the first extraction step and the high frequency component in the second direction extracted in the second extraction step, from the one with the larger absolute value A calculation step for subtracting the smaller absolute value to obtain an edge enhancement signal;
An outline enhancement method comprising: an addition step of adding an outline enhancement signal obtained in the calculation step to the input video signal to obtain an output video signal. An imaging unit for imaging a subject and outputting a video signal corresponding to the subject;
A video signal output from the imaging unit as an input video signal, and a contour emphasizing unit that obtains an output video signal by performing contour emphasis processing on the input video signal,
The contour emphasis part is
A first high pass filter for extracting a high frequency component in a first direction from the input video signal;
A second high-pass filter that extracts a high-frequency component in a second direction orthogonal to the first direction from the input video signal;
Using the high frequency component in the first direction extracted by the first high-pass filter and the high frequency component in the second direction extracted by the second high-pass filter, from the one having a larger absolute value An arithmetic unit for subtracting the smaller absolute value to obtain an edge enhancement signal;
An image pickup apparatus comprising: an adder that adds the contour emphasis signal obtained by the arithmetic unit to the input video signal to obtain the output video signal. A viewfinder having a display and displaying an image based on a video signal output from an imaging unit,
A video signal output from the imaging unit is used as an input video signal, and further includes a contour emphasizing unit that performs contour emphasis processing on the input video signal to obtain an output video signal to be supplied to the display,
The contour emphasis part is
A first high pass filter for extracting a high frequency component in a first direction from the input video signal;
A second high-pass filter that extracts a high-frequency component in a second direction orthogonal to the first direction from the input video signal;
Using the high frequency component in the first direction extracted by the first high-pass filter and the high frequency component in the second direction extracted by the second high-pass filter, from the one having a larger absolute value An arithmetic unit for subtracting the smaller absolute value to obtain an edge enhancement signal;
A viewfinder, comprising: an addition unit that adds the contour enhancement signal obtained by the calculation unit to the input video signal to obtain the output video signal.