JP4363125B2 - Image signal processing device, viewfinder, display device, image signal processing method, recording medium, and program - Google Patents

Description

  The present invention relates to an image signal processing apparatus. The present invention also relates to a viewfinder and a display device to which the technology is applied. The present invention also relates to an image signal processing method and program for realizing the technology. The present invention also relates to a recording medium on which the program is recorded.

Today, images require very high image quality. In particular, on a large screen, a slight focus shift at the time of imaging affects the image quality. For this reason, focusing of the lens is very important.
However, the image to be picked up is not always a picture that is easy to focus on. For example, it is difficult to focus on a pattern with low contrast. This is because it is difficult to confirm a change in contrast even when the focus is shifted.
For this reason, when it is desired to focus on a picture with low contrast, focusing is performed by using an imaging signal after contrast enhancement.
Japanese Patent Application Laid-Open No. 09-139552 JP 09-46576 A JP 2002-135801 A

  By the way, in the conventional apparatus, as shown in FIG. 1A, the contrast is adjusted by changing the gain of the imaging signal. At this time, the contrast adjustment affects the entire screen. For this reason, there is a problem that black crushing or white crushing occurs as shown in FIG.

The present invention has been made in consideration of the above problems, and an object thereof is to selectively emphasize only an image portion of interest and make it easy to see the picture of the image .

The present invention provides a contrast detection unit that calculates a numerical value indicating contrast with a peripheral pixel of each pixel related to an input image signal, and an absolute value of the numerical value indicating contrast is greater than a first threshold value Determines the filter coefficient as 0, and when the absolute value of the numerical value is between the first threshold value and a second threshold value smaller than the first threshold value, the filter coefficient is set to a predetermined value n (a variable). n is a natural number greater than or equal to 2) and is determined by a linear function. When the absolute value of the numerical value is smaller than the second threshold, the filter coefficient calculation unit determines the filter coefficient to 0, and when the filter coefficient is 0, the all-pass characteristic As the filter coefficient approaches 1, the shielding frequency decreases, and when the filter coefficient is 1, the image signal is completely blocked up to direct current, based on the filter coefficient determined by the filter coefficient calculation unit, for each signal related to each pixel. A low-pass filter unit for filtering, a subtracting unit for subtracting the output from the low-pass filter unit from the image signal, an amplifying unit for multiplying the output from the subtracting unit by a predetermined gain, and an output from the amplifying unit And a synthesizing unit for adding to the input image signal .
Note that the proposed image processing technique is applied when the contrast is individually adjusted for each of the luminance signal and the color signal (including the color difference signal), and when the contrast is individually adjusted for each of the primary color signals. Can do.

One embodiment of the present invention is a filter that selectively extracts only a low-contrast portion of an image signal. Only the low contrast component of the image signal appears in the output of this filter. Therefore, only the extracted low contrast component can be selectively amplified. By using this filter, for example, the following image processing apparatus can be realized.

FIG. 3 illustrates an image processing device which is one embodiment of the present invention. This image processing apparatus mainly has two functional units. One of the functions is the aforementioned filter. That is, the filter 1 selectively extracts only a low contrast portion of the image signal S1. One of the other functions is a synthesizer 2 that synthesizes and outputs the output signal S2 of the filter 1 and the image signal S1 that is the input signal of the filter 1. For the synthesizer 2, for example, an adder may be used.

  Here, the contrast is the contrast (range) of the image. The low contrast portion refers to an image having a small contrast with the surrounding image. This portion includes not only a single pixel but also a plurality of pixels. The combiner 2 outputs a combined output S3 obtained by combining the filter output S2 with the image signal S1. The signal amplitude of the low contrast component included in the combined output S3 is twice that of the image signal S1 that is the input signal. Of course, the signal component not extracted by the filter 1 remains as it is. Since only the low-contrast signal is emphasized, black crushing and white crushing do not occur.

FIG. 4 shows an image processing apparatus which is another embodiment of the present invention. This image processing apparatus mainly has three functional units. Two of the functional units are the same as in the second embodiment. The remaining functional unit is an amplifier 3 that amplifies and outputs the output signal S2 of the filter 1. The output signal S4 of the amplifier 3 is an input to the synthesizer 2. In the invention according to this aspect, the low contrast component S2 can be further enhanced. Therefore, the contrast on the screen increases, and it becomes easy to confirm a low-contrast portion that was difficult to confirm in the original image. Here, the amplification degree can be a fixed value, or can be variably controlled by a control signal.

FIG. 5 shows an image processing apparatus which is another embodiment of the present invention. This image processing apparatus mainly has four functional units. Three of the functional units are the same as in the third embodiment. However, an amplifier 3 having an amplification factor of n-1 times (n is a real number larger than 1) is used. In this case, a signal component obtained by amplifying the low contrast component S2 by n times appears in the combined output S3. Of course, also in this case, other signal components are output without being amplified.

  The remaining one functional unit is an attenuator 4 that attenuates and outputs the output signal S3 of the combiner 2 so as to cancel the amplification by the amplifier 3. The attenuator 4 multiplies the output signal S3 of the synthesizer 2 by 1 / m (m is a real number of 2 or more). In addition, n and m can set arbitrary values independently. Preferably, m is the same as n, or approximately the same value as n. When n and m have substantially the same value, the low contrast component of the image signal S1 appears in the output signal S5 of the attenuator 4 with almost the same contrast as the original image, while the other contrast components are compressed to 1 / m. appear.

  In this form, the contrast is selectively lowered only in a portion where the contrast of the image signal S1 is too high. Also by this method, the pattern of the original image can be converted into an image that is easy to see without losing the pattern of the original image. Needless to say, if the output signal S5 of the attenuator 4 is amplified, the contrast can be enhanced as a whole without causing blackout or whiteout.

  Even in this embodiment, the amplification degree of the amplifier 3 can be variably controlled. However, in that case, it is desirable to control the attenuation in conjunction with the control of the amplification. For example, when the amplification degree is increased, the attenuation degree is similarly increased. On the contrary, when the amplification degree is lowered, the attenuation degree is similarly controlled to be lowered.

Na us, filter 1 in the first embodiment to the fourth embodiment, it is desirable to be composed of two functional units. FIG. 6 shows a configuration example of the filter 1. One of the functional units inputs / outputs so that the input signal S1 appears as it is as the output signal S6 as it is in the high contrast portion, while the signal component appearing in the output signal S6 approaches the direct current component as it decreases in the low contrast portion. This is a low-pass filter 1A whose characteristics change. Another functional unit is a subtractor 1B that subtracts the output signal S6 of the low-pass filter 1A from the input signal S1 of the low-pass filter 1A.

  In the case of this filter 1, the image signal S1 having a high contrast component is canceled by the output signal S6 input to the subtracter 1B. On the other hand, the low-contrast component image signal S1 is output from the subtractor 1B without being canceled because the output signal S6 is substantially a direct current component. That is, a low contrast component is extracted.

  It is preferable that the change in input / output characteristics is small in the low-contrast portion and rapidly increases as the contrast increases. For example, it is preferable that the pass band for the low contrast component hardly changes in the vicinity of the direct current, but the pass band rapidly expands to the high band as the contrast increases. By providing such input / output characteristics, the ability to select the signal components to be extracted can be enhanced.

  The characteristics relating to the pass band may be continuously changed, or may be selected from a plurality of sets of pass bands. For example, the correspondence between the contrast and the pass band may be stored in a lookup table (storage medium), and a low pass filter having an appropriate pass band may be selectively used.

  It is more desirable for the low-pass filter to change its input / output characteristics so that the input signal appears as it is in the output signal when the contrast is lowered to near the noise level. When such a form is employed, noise and a signal component having such a low contrast that it cannot be distinguished from noise can be canceled by the subtractor 1B. Therefore, it is possible to prevent noise different from the original pattern from being emphasized.

  In this embodiment, it is desirable to have a coefficient generator that generates filter coefficients corresponding to the contrast of the image signal and adaptively varies the input / output characteristics of the low-pass filter 1A. In this case, the circuit scale can be reduced because only one low-pass filter 1A is required. The coefficient generation unit may be a lookup table (storage medium) that stores the correspondence between contrast and filter coefficients. Further, a similar relationship may be calculated by an arithmetic circuit, or a similar relationship may be calculated by software.

Na us, the contrast of the image signal in the first embodiment to the fourth embodiment, it is desirable to be determined on the basis of the signal level difference between peripheral pixels including the target pixel. By using information between peripheral pixels, the contrast detection accuracy can be increased.

  The contrast of the image signal is desirably determined based on a signal level difference between two adjacent pixels. Accuracy that can withstand practical use can also be realized by information between two pixels. In this case, the circuit required for detecting the contrast is simple, and the circuit scale can be reduced. Incidentally, the two pixels may be pixels adjacent in the horizontal direction on the screen or pixels adjacent in the vertical direction on the screen.

According to the present invention, without impairing the visibility of the whole image, it can be selectively emphasized only the image portion component of interest.

Embodiments of the present invention will be described below. It should be noted that characteristics not particularly shown or described herein may be selected from those known in the art.
In the following description, a case in which the preferred embodiment is realized as hardware will be described, but it can also be realized by a computer program equivalent to such hardware.

When the present invention is implemented as a computer program, the program is stored in a computer-readable storage medium.
Examples of the storage medium include a magnetic storage medium such as a magnetic disk (flexible disk or hard disk) or magnetic tape, an optical storage medium such as an optical disk, an optical tape, or a machine-readable barcode, and a random access memory (RAM). In addition to semiconductor storage devices such as read only memory (ROM), other physical devices or media used to store computer programs are included.

  Further, when the present invention is implemented in hardware, it may be implemented as an integrated circuit such as an application specific integrated circuit (ASIC) or other means known in the art. The present invention can be implemented not only as a digital circuit but also as an analog circuit.

(A) System Application Example Here, a case where the present invention is applied to a viewfinder will be described. As will be described later, the application example of the present invention is not limited to the viewfinder.
FIG. 11 shows a system configuration example of the viewfinder. This viewfinder is mainly composed of five circuit blocks. The contrast enhancement circuit 20, the peaking circuit 30, the contrast brightness circuit 40, the display device drive circuit 50, and the display device 60 are provided.

  The contrast enhancement circuit 20 in this embodiment is a circuit block for selectively enhancing low contrast components. This circuit block corresponds to one embodiment of the above-described invention. Note that an imaging signal is input to the contrast enhancement circuit 20 from a solid-state imaging device or other imaging device. The detailed configuration of the contrast enhancement circuit 20 will be described later.

  The peaking circuit 30 is a circuit block that expands only a high-frequency component suitable for focusing in the pulse width direction. This circuit block also corresponds to one embodiment of the above-described invention. The peaking circuit 30 receives the output signal of the previous contrast enhancement circuit 20. The detailed configuration of the peaking circuit 30 will also be described later.

The contrast / brightness circuit 40 is a circuit block for individually adjusting contrast and brightness. A known circuit block may be used. Therefore, the contrast adjustment here is adjustment for the entire band.
The display device drive circuit 50 is a drive block for the display device 60. Incidentally, the display device 60 generally uses a liquid crystal panel. In general, the eyepiece used for confirming the subject is called a viewfinder, but the display device here may be called a larger liquid crystal monitor. Further, the display device driving circuit 50 and the display device 60 may constitute an integrated unit, and may be externally attached to an imaging unit on which the contrast enhancement circuit 20 and the peaking circuit 30 are mounted.

As described above, in the case of the viewfinder shown in FIG. 11 equipped with two embodiments of the present invention, a viewfinder that is easy to focus not only on the low contrast portion but also on the high frequency portion is provided. Can be realized.
In the case of FIG. 11, in order to explain an application example of the present invention, both of the embodiment circuits of the present invention are mounted, but the system configuration in which only one embodiment circuit of the present invention is mounted You can also

For example, a known circuit may be used for the peaking circuit 30 among the contrast enhancement circuit 20 and the peaking circuit 30. In this case, it is possible to realize a viewfinder that can be easily focused on a low contrast portion.
Further, the contrast emphasis circuit 20 having a selective emphasis function is not used, and only the peaking circuit 30 which is a circuit of one embodiment of the present invention may be mounted. In this case, it is possible to realize a viewfinder that can be easily focused on the high frequency portion.

(B) Contrast Enhancement Circuit Hereinafter, an embodiment of the contrast enhancement circuit 20 will be described. The contrast enhancement circuit 20 uses an infinite impulse response type (hereinafter referred to as “IIR (Infinite Impulse Response)”) low-pass filter, and a finite impulse response type (hereinafter referred to as “FIR (Finite)”).
Impulse Response) type. ) Using a low-pass filter. Examples of each case will be described below.

(B-1) First Example FIG. 12 shows a first example of the contrast enhancement circuit 20. The contrast enhancement circuit 20 uses an IIR low-pass filter. The contrast enhancement circuit 20 belongs to the type of the image processing apparatus shown in FIG.

  The contrast enhancement circuit 20 is composed of six functional units. A function unit (21A, 21B) for detecting the contrast of the image signal Sin, a function unit (21E) for generating a filter coefficient in accordance with the detected contrast, and a low pass for filtering the low-frequency component of the input image signal Sin. A filter unit (21C, 21G, 21I, 21K), a subtracting unit (21D, 21F) for subtracting the output of the low-pass filter unit from the image signal Sin, and an amplifying unit (21H) used to adjust the enhancement degree, There are six synthesis units (21J) that synthesize the selectively emphasized low contrast component with the original image signal Sin.

  The functional units (21A, 21B) for detecting contrast are composed of a flip-flop 21A and a subtractor 21B. A pixel clock is input to the flip-flop 21A. Pixel data input as an image signal Sin by the flip-flop 21A is delayed by one pixel. In this embodiment, the subtractor 21B calculates a signal level difference between adjacent pixels in the horizontal direction.

In this function unit, the smaller the absolute value of the value obtained by the subtractor 21B is, the lower the contrast is determined to be.
If a delay element such as a flip-flop is further added, the contrast level can be distinguished over a wide range of up, down, left, and right. This is suitable when high detection accuracy is required.

  The low-pass filter unit (21C, 21G, 21I, 21K) includes a subtractor 21C, a multiplier 21G, an adder 21I, and a flip-flop 21K. This low-pass filter unit constitutes a first-order IIR low-pass filter. This filter characteristic can be varied by a filter coefficient given to the multiplier 21G. In this embodiment, when the filter coefficient is zero, all-pass characteristics are obtained. On the other hand, the cutoff frequency decreases as the filter coefficient approaches 1, and when the filter coefficient is 1, the direct current is completely blocked.

  The functional unit (21E) that generates the filter coefficient is configured by a coefficient generator 21E. In this embodiment, a look-up table is used as the coefficient generator 21E. FIG. 13 shows an example of input / output characteristics realized by a lookup table. The thick solid line portion in FIG. 13 is a curve representing the input / output characteristics. The output filter coefficient is closer to 1 as the absolute value of the difference value (input here) is smaller. Further, the filter coefficients become closer to zero as the absolute value of the difference value increases, and are all fixed to zero in a range where the difference value is larger than a certain value.

When such input / output characteristics are set, the above-described characteristics can be exhibited at the output of the low-pass filter section (output of the flip-flop 21K). That is, it is possible to realize a characteristic in which an input signal is passed through a portion with high contrast and a portion with low contrast is filtered to a low frequency.
The input / output characteristic curve has an n-order function curve (n is a natural number greater than or equal to 2) having only one extreme value (maximum value) on the output coordinate axis within the setting range of the difference value that defines variable control of the filter coefficient. ) Is desirable. The input / output characteristic curve is preferably a symmetrical type having the same filter value as long as the absolute value of the difference value is the same.

  If such a curve is employed, the change amount of the filter coefficient can be suppressed to be smaller than the change amount of the difference value in a range where the difference value is small. That is, in a certain range where the difference value is small, the output of the low-pass filter can be made almost zero. This means that the input pixel data can be output as it is from the subtractor 21F. That is, the low contrast component is extracted.

  On the other hand, if such a curve is adopted, for a range where the difference value is larger than a certain degree, the change amount of the filter coefficient can be made very large relative to the change amount of the difference value, and the filter coefficient can be brought close to zero rapidly. it can. That is, when the difference value becomes larger than a certain value, the cutoff frequency can be rapidly increased. This means that the input pixel data appearing in the low-pass filter changes rapidly. That is, when the difference value increases to some extent, the input pixel data can be canceled by the subtractor 21F.

  The input / output characteristics are not limited to those shown in FIG. For example, it may be set in a triangular shape having a vertex when the difference value is zero. That is, the filter coefficient can be set to change at a constant rate in accordance with the change amount of the difference value. Thus, the shape of the input / output characteristic curve can be set appropriately depending on the system to be applied and the image to be processed.

  In FIG. 13, a thin broken line is drawn near the difference value of zero. The input / output characteristics indicated by the broken lines are effective when the screen becomes difficult to see as a result of emphasizing the noise component at the same time as the fine pattern. This input / output characteristic is set in a region having a very low contrast so as to be indistinguishable from noise or a pattern. In a region where this characteristic is set, the filter coefficient suddenly becomes zero as the contrast is smaller. That is, the characteristics of the low-pass filter in this region are all-pass characteristics. This means that a component indistinguishable from noise is canceled out by the subtractor 21F.

  FIG. 14 shows a specific example of the input / output characteristic curve. In this example, the image signal Sin is quantized with 8 bits. In this example, the filter coefficient is set to zero in the range where the difference value is −1 to +1 and the absolute value is 16 or more. With this setting, noise having an amplitude in the range of −1 to +1 is positively removed. Note that the amplitude of noise that can be removed can be freely controlled by adjusting the range in which the filter coefficient is set to zero.

  For reference, FIG. 15 shows the relationship between the filter coefficient and the step response. From the figure, it can be seen that the response time when the filter coefficient is zero is very short. This is because the input signal is passed. Also, when the filter coefficient is 0.5, the response time is short because there are many input signals that are passed through. When the filter coefficient corresponding to the maximum value in FIG. 14 is 0.9375, the response time is very long. This is because the signal is almost cut off.

  The subtraction unit (21D, 21F) includes a flip-flop 21D and a subtractor 21F. The flip-flop 21D is for delay adjustment. The subtractor 21F is used to subtract the output of the low-pass filter unit from the output of the flip-flop 21D. In the output of the subtractor 21F, zero appears where the contrast is high and where the contrast is very low when the filter coefficient is zero for noise removal as described above. On the other hand, in the output of the subtractor 21F, the corresponding signal component appears up to a lower frequency as the contrast is lower, except for the above-described range.

  The amplifying unit (21H) includes a multiplier 21H. The multiplier 21H multiplies the output of the subtractor 21F by an appropriate gain Scc. The gain Scc for contrast control may be fixed. In this embodiment, an arbitrary gain Scc generated by the electronic volume is provided. By adjusting the gain Scc, only the extracted low contrast component can be amplified to an arbitrary contrast.

  The synthesizer (21J) includes an adder 21J. The adder 21J adds the selectively amplified contrast component to the original image (output of the flip-flop 21D) and outputs it. Needless to say, the image portion that is not selectively emphasized remains the original image. That is, nothing is added where the contrast ratio is originally high. Therefore, problems such as white crushing and black crushing as in the conventional circuit do not occur.

  As described above, when the contrast enhancement circuit 20 is configured as shown in FIG. 12, a low contrast component not including a noise component is extracted by a filter (a low-pass filter and a subtractor 21F) whose filter characteristics change according to the contrast. Can do. Then, the extracted low contrast component can be amplified with a desired gain and added to the original image signal.

(B-2) Second Embodiment FIG. 16 shows a second embodiment of the contrast enhancement circuit 20. Here, the contrast enhancement circuit 20 using an FIR type low-pass filter will be described. Since the IIR type low-pass filter is simply replaced with the FIR type low-pass filter, the basic configuration is the same as that of the first embodiment. For this reason, in FIG. 16, the same reference numerals are given to the portions corresponding to FIG. In the figure, the flip-flop 21D is not drawn, but this is because the time is adjusted using the delay stage of the FIR type low-pass filter 22.

  FIG. 17 shows a circuit configuration example of the FIR type low-pass filter 22. This low-pass filter 22 is a type of filter in which filter coefficients are arranged symmetrically with respect to the center tap. Therefore, a configuration is adopted in which 23 adders 22B are arranged for 48 delay devices 22A. Twenty-four multipliers 22C for multiplying the corresponding filter coefficients are arranged at the outputs of the adders 22B and the outputs of the center tap. Then, each multiplication result is added by one adder 22 </ b> D and used as the output of the low-pass filter 22.

  Note that the coefficient generator 21E in this embodiment outputs filter coefficients that are optimal for the contrast components detected by making a set of 24 filter coefficients. In this embodiment, as shown in FIG. 18, 15 types of filter coefficient groups are prepared. FIG. 18 shows a case where the image signal Sin is quantized with 8 bits. In the figure, the set of filter coefficients represented by “filter 0” corresponds to the case where the filter coefficient is zero in FIG. In this “filter 0”, a set of filter coefficients is set so that only the output of the center tap is output to the adder 22D. Also in the case of the second embodiment, in order to provide a characteristic for removing noise, “filter 0” is assigned when the absolute value of the difference is zero and one.

  Further, the filter coefficient pairs represented by “filter 1”, “filter 2”, “filter 3”... “Filter 15” have the same relationship as the filter coefficients in FIG. It is in. FIG. 18 shows the same relationship as when the absolute value of the difference is 2, the filter coefficient is reduced along the predetermined characteristic curve as the difference value increases with the filter coefficient of FIG. 14 as the maximum value. Incidentally, also in this embodiment, when the absolute value of the difference is 17 or more, “filter 0” is assigned to give an all-pass characteristic. Even with such a setting, the contrast enhancement circuit 20 that operates in the same manner as in the first embodiment can be realized.

(B-3) Third Example FIG. 19 shows a third example of the contrast enhancement circuit 20. The contrast enhancement circuit 20 uses an IIR low-pass filter. This contrast enhancement circuit 20 belongs to the type of the image processing apparatus shown in FIG.

  In this embodiment, the basic configuration is the same as that of the first embodiment. For this reason, in FIG. 19, the same reference numerals are given to the portions corresponding to FIG. There are two differences between the third embodiment and the first embodiment. One is that an amplification unit 23 of n-1 times (n is a real number larger than 1) is provided in the subsequent stage of the multiplier 21H. The other is that a 1 / n-times attenuation unit 24 is provided at the subsequent stage of the adder 21J.

  In the case of this embodiment, although it depends on the value of the contrast control gain Scc, the low contrast portion of the original image amplified n times by the multiplier 23 and the adder 21J is 1 / n times by the attenuator 24. Therefore, it is output with the same contrast as that at the time of input. On the other hand, the contrast of the portion where the contrast is high in the original image is reduced to 1 / n times by the attenuator 24. This means that the contrast of the entire image can be averaged. Therefore, since the high contrast portion is included, the low contrast portion of the original image that is difficult to see can be easily confirmed on the screen.

  Needless to say, only the high contrast portion of the original image is changed, and the information of the original image is stored as it is for the pattern (pattern). This embodiment is effective when used in a signal processing circuit of a surveillance camera or when post-processing a captured image. For example, it is suitable for special effects and other image editing.

(B-4) Fourth Embodiment FIG. 20 shows a fourth embodiment of the contrast enhancement circuit 20. This contrast enhancement circuit 20 also uses an IIR low-pass filter, and belongs to the type of the image processing apparatus shown in FIG. In FIG. 20, the same reference numerals are given to the portions corresponding to FIG. 12.

  The fourth embodiment is characterized in that the contrast detection function (21A, 21B) of the first embodiment is realized by a subtractor 21C and a flip-flop 21K of a low-pass filter section. For this reason, the flip-flop 21A and the subtractor 21B which are necessary in the first embodiment are not required, and the circuit configuration is simplified correspondingly. The operation contents will be briefly described below.

  In the fourth embodiment, the subtractor 21C calculates the difference between the current output (the output of the flip-flop 21D) and the current input, and provides the difference value to the coefficient generator 21E. The coefficient generator 21E generates a filter coefficient corresponding to the detected contrast according to the input / output characteristics shown in FIG. 14, and supplies the generated filter coefficient to the multiplier 21G.

  In the same way as the first embodiment, it is determined that the contrast value is higher as the difference value is larger, and the contrast is smaller as the difference value is smaller. As in the first embodiment, the lower the contrast, the lower the filter output filtered to the lower frequency.

By the way, strictly speaking, it is inevitable that the contrast detection output (output of the subtractor 21C) obtained by such a configuration has a part different from the contrast of the actual input image.
However, even if such an error occurs, in the case of the fourth embodiment, since the error is automatically canceled, there is no practical problem. This automatic cancellation function is realized by a combination of the characteristics of the IIR low-pass filter and the input / output characteristics of the coefficient generator 21E.

In an IIR low-pass filter, if an error component is included in an input signal, the error component is accumulated infinitely. For this reason, the difference between the low-pass filter output (the output of the flip-flop 21K) and the input gradually increases due to the large error.
However, since the coefficient generator 21E acts to decrease the filter coefficient when the difference increases, the coefficient generator 21E is automatically reset when the accumulated error increases to some extent. That is, since the coefficient generator 21E brings the filter coefficient close to zero, the error circulation amount is automatically suppressed.
As described above, also in this embodiment, the same accuracy as in the other examples described above can be realized.

(C) Peaking Circuit An example of the peaking circuit 30 will be described below. FIG. 21 shows an embodiment of the peaking circuit 30. The peaking circuit 30 includes a band pass filter 31, a pulse width expansion unit 32, an amplification unit 33, and a synthesis unit 34.

  The bandpass filter 31 is a functional unit that selectively extracts a band component of interest from the image signal. As the band pass filter 31, a filter having a fixed pass band may be used, or a filter that can arbitrarily set the pass band may be used. A plurality of band pass filters 31 having different pass bands may be prepared. In any case, a known bandpass filter can be used.

  In consideration of the use for focusing, it is generally desirable that the passband be as high as possible. In practice, the frequency that is most easily seen on the screen is selected in consideration of the lens characteristics, the signal processing band of the imaging device, the S / N ratio, and the like. In general, it is easier to see when the bandwidth is set narrow.

  The pulse width expansion unit 32 is a functional unit that expands the band component filtered by the bandpass filter 31 in the pulse width direction while keeping the amplitude constant. That is, only the pulse width is selectively increased without saturation of the amplitude. Thereby, it is possible to easily confirm the high frequency component on the screen. This pulse width extending portion 32 corresponds to one embodiment of the above-described invention. A detailed configuration will be described later.

The amplifying unit 33 is a functional unit for amplifying the output of the pulse width expanding unit 32 with an arbitrary gain. The gain is set as a gain Spg for peak gain control. For example, it comprises a multiplier.
The synthesizing unit 34 is a functional unit for synthesizing the amplified image signal and the original image signal. The synthesizer 34 is composed of an adder, for example. Thus, the synthesizing unit 34 outputs an image signal in which the pulse width is expanded by a specific band component (the line is thickened) without changing the approximate pattern of the original image.

  Hereinafter, an embodiment of the pulse width expanding unit 32 which is a main configuration of the peaking circuit 30 will be described.

(C-1) First Example FIG. 22 shows a first example of the pulse width expanding unit 32. The pulse width expansion unit 32 uses the characteristics of amplitude modulation. Specifically, as in the radio superheterodyne system, an appropriate carrier wave is amplitude-modulated with a signal wave to generate a sideband wave, which is filtered to convert a high frequency to a low frequency. In this way, the pulse width can be widened by shifting the band component to the low frequency side.

  The pulse width extending unit 32 is mainly composed of four functional units. There are four oversampling units (32A, 32B), amplitude modulation units (32C, 32D), a lower sideband extraction unit (32E), and a downsampling unit (32F).

  The oversampling units (32A, 32B) are functional units that oversample and output the image signal Sin. The oversampling unit includes a zero insertion unit 32A and a low-pass filter 32B.

  The zero insertion unit 32A inserts n−1 zeros into the image signal Sin filtered by the preceding bandpass filter 31 to obtain an n-times oversampling result. Note that n is a natural number of 1 or more. Assuming that the center frequency of the pass band of the bandpass filter 31 is f, the image signal Sin is represented by the spectrum of FIG.

  When oversampling is performed by the zero insertion unit 32A, a plurality of aliasing components are generated. FIG. 23B shows the spectrum. Here, the band is limited to ½ of the original sampling frequency fs by the low-pass filter 32B so as not to cause excessive aliasing. FIG. 23C shows the output spectrum of the low-pass filter 32B.

  The amplitude modulators (32C, 32D) modulate the amplitude of the carrier wave generated by the oscillator 32C with the output of the low-pass filter 32B. Amplitude modulation is performed by the multiplier 32D. Here, when the frequency of the carrier wave is expressed as F, two sidebands of center frequencies F−f and F + f are generated on both sides of the frequency F of the carrier wave. FIG. 23D shows the spectrum.

  At this time, if the frequency F or F + f exceeds nfs / 2, aliasing from nfs occurs and unexpected interference occurs. For this reason, the magnification n is selected so as not to cause a problem. Incidentally, when there is no problem at n = 1, the oversampling units (32A, 32B) and the downsampling unit (32F) can be omitted.

  The lower sideband extraction unit (32E) includes a bandpass filter 32E that extracts only the sideband centered on the frequency F-f. FIG. 23E shows the output spectrum of the bandpass filter 32E. The downsampling unit (32F) downsamples the output of the bandpass filter 32E and returns it to the original sampling frequency. As a result, the center frequency f of the image signal Sin is converted to a lower F-f.

  As described above, by appropriately selecting the frequency F of the carrier wave, it is possible to extract the vicinity of the frequency f suitable for focusing and convert it to the vicinity of the frequency Ff that is easy to see on the display device.

(C-2) Second Example FIG. 24 shows a second example of the pulse width expanding unit 32. The pulse width expanding unit 32 employs a method of widening the pulse width by limiting the waveform amplified in the amplitude direction with the amplitude before amplification.
The pulse width extending unit 32 is composed of four functional units. There are four oversampling units (32G, 32H), maximum amplitude detecting units (32I, 32J, 32K), amplifying units (32L), and amplitude limiting units (32M).

  The oversampling units (32G, 32H) are functional units that oversample and output the image signal Sin. The configuration of this oversampling unit is the same as that of the oversampling unit used in the first embodiment. That is, it includes a zero insertion unit 32G and a low-pass filter 32H. Detailed description of each part is omitted because it is redundant.

Note that the reason for oversampling in the present embodiment is to detect a correct peak value in the maximum amplitude detector (32I, 32J, 32K) in the subsequent stage. The reason will be described with reference to FIG.
In general, peaking uses a narrow pulse waveform. For this reason, the sampling data is discrete as shown by the black circles in FIG. 25A, and the maximum value is smaller than the true peak value. Therefore, a method of interpolating sampling points by oversampling is adopted.

  As shown in FIG. 25 (B), it is possible to obtain data that is very close to the true peak value only by oversampling twice. Therefore, the peak value can be obtained with high accuracy by using the data after oversampling. In this embodiment, the magnification is doubled, but a higher magnification can be adopted. The accuracy can be improved as the magnification is increased.

The oversampled image signal Sin is input to the maximum amplitude detector (32I, 32J, 32K), and its peak value is detected. The maximum amplitude detector (32I, 32J, 32K) includes an absolute value circuit 32I, a plurality of shift registers 32J constituting a delay device, and a maximum value circuit 32K.
The absolute value circuit 32I is a circuit that extracts the absolute value of the image signal.

The maximum value circuit 32K is a circuit that detects the maximum value from the sample values corresponding to the number of stages of the shift register 32J. The maximum value circuit 32K outputs the detected maximum value as the peak value of the image signal (pulse wave) extracted by the band pass filter 31.
Here, the number of stages of the shift register 32J that determines the search range of the peak value is within the range covered by the impulse response of the band-pass filter 31 in the preceding stage (that is, the number of taps of the band-pass filter 31).
The amplitude of the original image signal is detected and stored by the above two functional units.

  The amplifying unit (32L) amplifies the image signal Sin in the amplitude direction (more precisely, in the pulse width direction). This amplifying unit includes a multiplier 32L. The multiplier 32L amplifies the image signal Sin in accordance with the pulse width control gain Spc. FIG. 26A shows the waveform of the image signal Sin, and FIG. 26B shows the signal waveform after amplification. As the gain is increased, the pulse width after waveform shaping can be increased. The amplified image signal is output to the amplitude limiter (32M).

The amplitude limiter (32M) is composed of a limiter circuit 32M. The peak value of the input pulse is input to the limiter circuit 32M from the aforementioned maximum value circuit 32K. The limiter circuit 32M limits the amplitude of the input signal to be equal to or less than the peak value.
As a result, the limiter 32M outputs a trapezoidal pulse wave obtained by cutting off the upper part of the signal waveform as shown in FIG.
A delay circuit (not shown) is arranged in front of the limiter circuit 32M in order to match the delay between the output of the multiplier 32L and the output of the maximum value circuit 32K.

  The present invention can be applied to, for example, a broadcast camera unit, a movie camera unit, a consumer camera unit, an industrial camera unit, a surveillance camera unit, and other signal processing circuits. Note that these camera units may be integrated with a recording unit for recording image signals on a recording medium. As the recording medium, for example, a magnetic recording medium (including tape and disk), a magneto-optical recording medium, an optical disk, a semiconductor memory, and the like can be used.

  The present invention can also be applied to a signal processing circuit of an electronic camera, for example. Note that the present invention can also be applied to various electronic devices with an electronic camera. For example, the present invention can be applied to a mobile phone with an electronic camera and an information terminal. Moreover, it is applicable also as these attached units.

  The present invention can also be applied to an editing apparatus that performs special effects and other image processing on the image signal output from the image capturing unit. The present invention can also be applied to a circuit board or a circuit unit mounted on such an editing apparatus.

  The present invention can also be applied to image editing software that runs on a computer. That is, it can be used not only for focusing of a camera lens but also for applications such as emphasizing a low-contrast portion while retaining features of an original image such as a pattern or a picture, or for converting a thin line into a thick line.

It is a figure which shows the contrast adjustment function of a conventional circuit, and its problem. It is a figure which shows the reason for using a pattern with a high spatial frequency for focusing. It is a figure which shows one example of the form of the image processing apparatus to which the selective enhancement technique of a low contrast component is applied. It is a figure which shows one example of the form of the image processing apparatus to which the selective enhancement technique of a low contrast component is applied. It is a figure which shows one example of the form of the image processing apparatus to which the selective enhancement technique of a low contrast component is applied. It is a figure which shows the structural example of the filter which has the selection function of a low contrast component. It is a figure which shows one of the examples of the image processing apparatus to which the selective emphasis technique of a pulse width is applied. It is a figure which shows the example of a form of a pulse width expansion | extension part. It is a figure which shows the other example of a pulse width expansion part. It is a figure which shows the other example of a pulse width expansion part. It is a figure which shows the structural example of a viewfinder. It is a figure which shows the 1st Example of a contrast emphasis circuit. It is a figure which shows the input-output characteristic of a coefficient generator. It is a figure which shows the specific example of an input / output characteristic. It is a figure which shows the relationship between a filter coefficient and a step response. It is a figure which shows the 2nd Example of a contrast emphasis circuit. It is a figure which shows the structural example of a FIR type low-pass filter. It is a figure which shows the relationship between a difference value and the group of a filter coefficient. It is a figure which shows the 3rd Example of a contrast emphasis circuit. It is a figure which shows the 4th Example of a contrast emphasis circuit. It is a figure which shows the structural example of a peaking circuit. It is a figure which shows the 1st Example of a pulse width expansion | extension part. It is a figure which shows the signal waveform of each part of the pulse width expansion | extension part shown in FIG. It is a figure which shows the 2nd Example of a pulse width expansion | extension part. It is a figure which shows the interpolation of the sampling point by oversampling. It is a figure which shows the signal waveform of each part of the pulse width expansion | extension part shown in FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1,11 Filter 2,13 Synthesizer 3 Amplifier 4 Attenuator 12, 32 Pulse width expansion part 12A Amplitude modulation part 12C Oversampling part 12D Downsampling part 12E Absolute value part 12F Maximum value detection part 12G Amplification part 12H Amplitude restriction part 20 Contrast enhancement circuit 21E Coefficient generator 30 Peaking circuit

Claims (6)

A contrast detection unit that calculates a numerical value indicating contrast with a peripheral pixel of each pixel for each pixel related to the input image signal;
When the absolute value of the numerical value indicating contrast is larger than the first threshold, the filter coefficient is determined to be 0, and the absolute value of the numerical value is the first threshold and the second threshold smaller than the first threshold; The filter coefficient is determined by a predetermined n (n is a natural number greater than or equal to 2) degree function with the numerical value as a variable, and when the absolute value of the numerical value is smaller than the second threshold, the filter A filter coefficient calculator for determining the coefficient to be 0,
When the filter coefficient is 0, all-pass characteristics are obtained, the closer the filter coefficient is to 1, the lower the shielding frequency is, and when the filter coefficient is 1, the filter coefficient is determined to be completely cut off to the direct current. A low pass filter for filtering the image signal for each signal associated with each pixel based on a filter coefficient;
A subtracting unit for subtracting the output from the low-pass filter unit from the image signal;
An amplifying unit that multiplies the output from the subtracting unit by a predetermined gain;
A synthesis unit for adding the output from the amplification unit to the input image signal;
An image signal processing apparatus comprising: A contrast detection unit that calculates a numerical value indicating contrast with a peripheral pixel of each pixel for each pixel related to the input image signal;
When the absolute value of the numerical value indicating contrast is larger than the first threshold, the filter coefficient is determined to be 0, and the absolute value of the numerical value is the first threshold and the second threshold smaller than the first threshold; The filter coefficient is determined by a predetermined n (n is a natural number greater than or equal to 2) degree function with the numerical value as a variable, and when the absolute value of the numerical value is smaller than the second threshold, the filter A filter coefficient calculator for determining the coefficient to be 0,
When the filter coefficient is 0, all-pass characteristics are obtained, and as the filter coefficient approaches 1, the shielding frequency decreases, and when the filter coefficient is 1, the filter coefficient is determined to be completely cut off to the direct current. A low pass filter for filtering the image signal for each signal associated with each pixel based on a filter coefficient;
A subtracting unit for subtracting the output from the low-pass filter unit from the image signal;
An amplifying unit that multiplies the output from the subtracting unit by a predetermined gain;
An image signal processing apparatus comprising: a synthesis unit that adds the output from the amplification unit to the input image signal;
A display unit for displaying an output signal of the image signal processing device;
Viewfinder with a. For each pixel related to the input image signal, calculate a numerical value indicating the contrast with the surrounding pixels of each pixel, a contrast detection unit,
When the absolute value of the numerical value indicating contrast is larger than the first threshold, the filter coefficient is determined to be 0, and the absolute value of the numerical value is the first threshold and the second threshold smaller than the first threshold; The filter coefficient is determined by a predetermined n (n is a natural number greater than or equal to 2) degree function with the numerical value as a variable, and when the absolute value of the numerical value is smaller than the second threshold, the filter A filter coefficient calculator for determining the coefficient to be 0,
When the filter coefficient is 0, all-pass characteristics are obtained, the closer the filter coefficient is to 1, the lower the shielding frequency is, and when the filter coefficient is 1, the filter coefficient is determined to be completely cut off to the direct current. A low pass filter for filtering the image signal for each signal associated with each pixel based on a filter coefficient;
A subtracting unit for subtracting the output from the low-pass filter unit from the image signal;
An amplifying unit that multiplies the output from the subtracting unit by a predetermined gain;
An image signal processing apparatus comprising: a synthesis unit that adds the output from the amplification unit to the input image signal;
A display control unit for driving a display unit based on an output signal of the image signal processing device and displaying an image on the display unit;
A display device comprising: A contrast detection step for calculating a numerical value indicating contrast with a peripheral pixel of each pixel for each pixel related to the input image signal;
When the absolute value of the numerical value indicating contrast is larger than the first threshold, the filter coefficient is determined to be 0, and the absolute value of the numerical value is the first threshold and the second threshold smaller than the first threshold; The filter coefficient is determined by a predetermined n (n is a natural number greater than or equal to 2) degree function with the numerical value as a variable, and when the absolute value of the numerical value is smaller than the second threshold, the filter A filter coefficient calculating step for determining a coefficient to be 0;
Based on the filter coefficient determined in the filter coefficient step, all-pass characteristics are obtained when the filter coefficient is 0, the shielding frequency decreases as the filter coefficient approaches 1, and complete until DC when the filter coefficient is 1. A filtering step of filtering the image signal for each signal related to each pixel in a low-pass filter section blocked by
A subtraction step of subtracting the output from the low-pass filter unit from the image signal;
An amplification step of multiplying the signal subtracted in the subtraction step by a predetermined gain set in advance;
A synthesis step of adding the signal multiplied in the amplification step to the input image signal;
An image signal processing method including : A contrast detection function for calculating a numerical value indicating a contrast with a peripheral pixel of each pixel for each pixel related to the input image signal;
When the absolute value of the numerical value indicating contrast is larger than the first threshold, the filter coefficient is determined to be 0, and the absolute value of the numerical value is the first threshold and the second threshold smaller than the first threshold; The filter coefficient is determined by a predetermined n (n is a natural number greater than or equal to 2) degree function with the numerical value as a variable, and when the absolute value of the numerical value is smaller than the second threshold, the filter A filter coefficient calculation function for determining the coefficient to be 0;
Based on the filter coefficient determined by the filter coefficient function, all-pass characteristics are obtained when the filter coefficient is 0, the shielding frequency decreases as the filter coefficient approaches 1, and when the filter coefficient is 1, the direct current is completely achieved. A filtering function for filtering the image signal for each signal related to each pixel, in a low-pass filter section blocked by
A subtraction function for subtracting the output from the low-pass filter unit from the image signal;
An amplification function for multiplying the signal subtracted by the subtraction function by a predetermined gain set in advance;
A computer-readable recording medium recording a program for causing a computer to realize a synthesis function of adding a signal multiplied by the amplification function to the input image signal . A contrast detection function for calculating a numerical value indicating a contrast with a peripheral pixel of each pixel for each pixel related to the input image signal;
When the absolute value of the numerical value indicating contrast is larger than the first threshold, the filter coefficient is determined to be 0, and the absolute value of the numerical value is the first threshold and the second threshold smaller than the first threshold; The filter coefficient is determined by a predetermined n (n is a natural number greater than or equal to 2) degree function with the numerical value as a variable, and when the absolute value of the numerical value is smaller than the second threshold, the filter A filter coefficient calculation function for determining the coefficient to be 0;
Based on the filter coefficient determined by the filter coefficient function, all-pass characteristics are obtained when the filter coefficient is 0, the shielding frequency decreases as the filter coefficient approaches 1, and when the filter coefficient is 1, the direct current is completely achieved. A filtering function for filtering the image signal for each signal related to each pixel, in a low-pass filter section blocked by
A subtraction function for subtracting the output from the low-pass filter unit from the image signal;
An amplification function for multiplying the signal subtracted by the subtraction function by a predetermined gain set in advance;
A program for causing a computer to realize a synthesis function for adding a signal multiplied by the amplification function to the input image signal .

Images