JP4356604B2 - VIDEO SIGNAL PROCESSING DEVICE, VIDEO SIGNAL PROCESSING METHOD, VIDEO SIGNAL PROCESSING PROGRAM, AND RECORDING MEDIUM CONTAINING VIDEO SIGNAL PROCESSING PROGRAM - Google Patents

Description

  The present invention relates to a video signal processing device, a video signal processing method, a video signal processing program, and a recording medium on which a video signal processing program is recorded, and is applied to, for example, a video camera using an XY address scanning type imaging device. Can do. The present invention detects a flicker frequency spectrum by frequency analysis of a change in signal level, determines a change in the initial phase of this spectrum in a continuous field or frame, and suppresses a flicker component of a video signal. Compared to the above, flicker can be surely reduced by a simpler process.

  Conventionally, various methods for reducing flicker have been proposed for a video signal processing apparatus using an XY address scanning type image sensor such as a CMOS image sensor.

  Here, flicker occurs when shooting is performed with illumination of a blinking light source such as a fluorescent lamp. In the video signal processing apparatus using an XY address scanning type image sensor, as shown in FIG. It is observed as a change in brightness level and hue. Here, in a fluorescent lamp with a commercial power source having a frequency of 50 [Hz], the blinking frequency is 100 [Hz]. Therefore, in an NTSC video signal with a field frequency of 60 [Hz], the number of lines per field is M. One cycle L of such fluctuation is L = M × 60/100 lines. Further, in one field, such a periodic variation occurs 100/60 = 1.66 period.

  For this reason, Japanese Patent Laid-Open No. 2004-222228 discloses a technique for detecting a flicker frequency spectrum by frequency analysis of a change in the signal level of the video signal and correcting the signal level of the video signal based on the amplitude value of the spectrum. A method for reducing flicker has been proposed. According to this method, it is possible to reliably prevent vertical luminance level and hue fluctuations in the vertical direction due to flicker, which are inherent to an XY address scanning type imaging device, without separately providing an optical system or the like for detecting flicker. can do. Here, the flicker frequency means the blinking frequency of the light source.

  In the case of this method, when the amplitude value of the spectrum by the flicker frequency is equal to or less than a predetermined threshold value, it is determined that the image is taken with natural light illumination outdoors or the like, and the imaging result is not corrected. Output.

  However, in practice, in outdoor shooting, a moving object may be shot, and a spectrum based on a flicker frequency may be detected by the moving object. The amplitude value of the spectrum based on the flicker frequency detected in this way varies depending on the exposure time, the speed of the electronic shutter, and the like.

As a result, the technique disclosed in Japanese Patent Application Laid-Open No. 2004-222228 requires various changes to the threshold for determining the flicker frequency spectrum depending on the shooting conditions, and there is a problem that the processing becomes complicated accordingly. . If the threshold value is inappropriately set, the imaging result is erroneously corrected, and the image quality is deteriorated.
JP 2004-222228 A

  The present invention has been made in consideration of the above points, and a video signal processing apparatus, a video signal processing method, and a video signal processing program capable of reliably reducing flicker by simpler processing than in the prior art. And a recording medium on which a video signal processing program is recorded.

  In order to solve such a problem, in the invention of claim 1, applied to a video signal processing device, video signal acquisition means for acquiring a video signal, and an integration period of one or more lines of the video signal as a unit, An integration means for integrating the video signal and outputting an integration result; a difference value of the integration result from the corresponding integration period in an adjacent field or frame; or a normalizing means for normalizing the integration result; Frequency analysis of the processing result of the conversion means, and for each integration period, between the frequency analysis means for detecting the amplitude value and initial phase of the spectrum at the flicker frequency and the corresponding integration period of the adjacent field or frame , A phase change detecting means for detecting a change in the initial phase, a detection result by the phase change detecting means, and the initial phase by the flicker. Correction coefficient calculation means for generating a correction coefficient whose value decreases as the detection result moves away from the logical value based on a difference value from the logical value of the correction, and corrects the signal level of the video signal based on the correction coefficient And a video signal correcting means.

  According to a sixth aspect of the present invention, the video signal is applied to a video signal processing method, and the video signal is acquired in units of an integration period of one or more lines of the video signal. An integration step of integrating and outputting an integration result; a difference value of the integration result from the corresponding integration period in an adjacent field or frame; or a normalization step of normalizing the integration result; and the normalization Frequency analysis of the processing result of the step, and for each integration period, between the frequency analysis step of detecting the amplitude value and initial phase of the spectrum at the flicker frequency and the corresponding integration period of the adjacent field or frame A phase change detection step for detecting a change in the initial phase, a detection result by the phase change detection step, and a flicker. Based on a difference value from the logical value of the change in the initial phase, a correction coefficient calculation step for generating a correction coefficient whose value decreases as the detection result moves away from the logical value, and an image based on the correction coefficient And a video signal correcting step for correcting the signal level of the signal.

  According to a seventh aspect of the present invention, the processing procedure is applied to a video signal processing program for reducing the flicker of the video signal by causing the arithmetic processing means to execute a predetermined processing procedure. A signal acquisition step, an integration step of integrating the video signal and outputting an integration result in units of integration periods of one or more lines of the video signal, and a corresponding integration period in an adjacent field or frame; The normalization step of normalizing the integration result and the processing result of the normalization step are frequency-analyzed, and the amplitude value of the spectrum at the flicker frequency and the initial value for each integration period. Between the step of frequency analysis detecting the phase and the corresponding integration period of the adjacent field or frame. Based on a difference value between a phase change detection step of detecting a phase change, a detection result of the phase change detection step, and a logical value of the initial phase change due to flicker, the detection result is calculated from the logical value. A correction coefficient calculation step for generating a correction coefficient whose value decreases as the distance increases, and a video signal correction step for correcting the signal level of the video signal based on the correction coefficient are provided.

  According to an eighth aspect of the present invention, the processing procedure is applied to a recording medium recorded with a video signal processing program for reducing video signal flicker by causing the arithmetic processing means to execute a predetermined processing procedure. Video signal acquisition step of acquiring a signal, integration step of integrating the video signal and outputting an integration result in units of integration periods of one or more lines of the video signal, and correspondence in adjacent fields or frames A difference value of the integration result with respect to the integration period, or a normalization step for normalizing the integration result, and a frequency analysis of the processing result of the normalization step, and a spectrum at a flicker frequency for each integration period. A step of frequency analysis to detect an amplitude value and an initial phase of the signal, and a corresponding integration period of the adjacent field or frame Based on the difference value between the phase change detection step for detecting the change of the initial phase, the detection result of the phase change detection step, and the logical value of the change of the initial phase due to flicker, A correction coefficient calculation step for generating a correction coefficient whose value decreases as the detection result moves away from the logical value; and a video signal correction step for correcting the signal level of the video signal based on the correction coefficient. .

  According to the configuration of the first aspect, the present invention is applied to a video signal processing apparatus, and a video signal acquisition unit that acquires a video signal, and integrates and integrates the video signal in units of integration periods of one or more lines of the video signal. The integration means for outputting the result, the difference value of the integration result between the corresponding integration periods in adjacent fields or frames, or the normalizing means for normalizing the integration result, and the processing result of the normalization means as the frequency Analyzing and detecting, for each integration period, the change in the initial phase between the frequency analysis means for detecting the amplitude value and initial phase of the spectrum at the flicker frequency, and the corresponding integration period in the adjacent field or frame. And a phase change detecting means for detecting the luminance of the video signal when the initial phase change detected by the phase change detecting means is detected. If changes such bell is by flicker, whereas a value according to the logical value determined by the blinking frequency of the light source, in the case of movement of the subject or the like, a value different from the logical value by the degree. As a result, based on the difference value between the detection result by the phase change detection means and the logical value of the initial phase change by the flicker, a correction coefficient whose value decreases as the detection result moves away from the logical value is generated. If the correction coefficient calculation means and the video signal correction means for correcting the signal level of the video signal based on the correction coefficient are provided, it is not necessary to change the determination criteria depending on the exposure time, the speed of the electronic shutter, etc. In addition, even when a moving object is photographed, the flicker component can be surely suppressed, and thus flicker can be reliably reduced by a simpler process than in the prior art.

  Thus, according to the configurations of claims 6, 7, and 8, a video signal processing method, a video signal processing program, and a video signal processing program capable of reliably reducing flicker by a simpler process than in the past. A recording medium in which a video signal processing program is recorded can be provided.

  According to the present invention, flicker can be surely reduced by a simpler process than in the prior art.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

(1) Configuration of Embodiment FIG. 2 is a block diagram illustrating an imaging apparatus according to Embodiment 1 of the present invention. In this imaging device 1, the lens 2 condenses incident light by the diaphragm and magnification driven by the lens driver 3, and forms an optical image on the imaging surface of the subsequent imaging device 4. Here, the lens driver 3 drives the lens 2 under the control of the system controller 5, thereby changing the magnification and aperture of the lens 2.

  The image pickup device 4 is a CMOS solid-state image pickup device that is an XY address scanning type image pickup device. The image pickup device 4 operates in response to a timing signal output from a timing generator (TG) 6 and displays an image pickup result of an optical image formed on the image pickup surface. Output by the imaging signal S1. In other words, the image pickup device 4 includes pixels having a photodiode (photogate), a transfer gate (shutter transistor), a switching transistor (address transistor), an amplification transistor, a reset transistor (reset gate), and the like arranged in a matrix on a semiconductor substrate. In addition, a vertical scanning circuit, a horizontal scanning circuit, and a video signal output circuit for driving these pixels are formed, and photoelectric conversion results are sequentially output by the pixels in the order of raster scanning. In the image pickup device 4, each pixel is provided with a complementary color or primary color filter, thereby outputting an image pickup signal using a color signal based on the complementary color or the primary color. The timing generator 6 outputs various timing signals to the image pickup device 4 under the control of the system controller 5, whereby the image pickup device 1 sets the electronic shutter speed under the control of the system controller 5.

  The analog signal processing circuit 7 performs correlated double sampling on the imaging signal S1 output from the imaging device 4, and then corrects and outputs the signal level by automatic gain control. The subsequent analog / digital conversion circuit (A / D) performs an analog / digital conversion process on the output signal of the analog / digital signal processing circuit 7 to output image data D1.

  The digital signal processing circuit 9 is formed by, for example, a digital signal processor. The image data D1 output from the analog-digital conversion circuit 8 is subjected to white balance adjustment, gamma adjustment, and the like, and then an NTSC video signal having a field frequency of 60 [Hz]. Output by SV. In this series of processing, the digital signal processing circuit 9 reduces flicker by a flicker reduction circuit 10 which is one of functional blocks. In the imaging apparatus 1, the video signal SV output from the digital signal processing circuit 9 is recorded on a recording medium such as an optical disk, a magnetic disk, or a memory card.

  Accordingly, in this embodiment, the image pickup device 4, the analog signal processing circuit 7, the analog-digital conversion circuit 8, and the digital signal processing circuit 9 take an image of a desired subject and output a video signal SV based on the image pickup result. An image pickup means using a type of image pickup element is configured.

  The system controller 5 is a control unit that controls the operation of the imaging apparatus 1 and controls the overall operation by executing a predetermined processing program in response to the operation of the operator by the user. Specifically, the system controller 5 controls the operation of the lens driver 3 and the timing generator 6 in response to the operation of the operator by the user and in accordance with the luminance level of the imaging result, thereby the magnification of the lens 2 is adjusted. Variable, and further, aperture control processing is executed. Further, the white balance adjustment in the digital signal processing circuit 9 is controlled in accordance with the signal level of each color signal, thereby executing the auto white balance adjustment process.

  When executing this processing program, the system controller 5 executes the flicker reduction program provided in this processing program in accordance with an instruction from the user to control the operation of the flicker reduction circuit 10, whereby the flicker reduction circuit 10 Reduce flicker. A series of programs of the system controller 5 relating to the flicker reduction processing and a processing program in the digital signal processing circuit 9 are provided by being installed in the imaging apparatus 1 in advance, but instead via a network such as the Internet. These programs may be provided by installation, or may be provided by various recording media. As such a recording medium, various recording media such as an optical disk, a magnetic disk, and a memory card can be widely applied.

  FIG. 3 is a functional block diagram showing the flicker reduction circuit 10 together with a related configuration. In the image pickup apparatus 1 according to this embodiment, the processing with the configuration shown in FIG. 3 is executed for each of the luminance signal and color difference signal constituting the video signal SV. The processing related to the flicker reduction circuit 10 may be executed at least for the luminance signal, and may be executed for the color difference signal and each color signal as necessary. The luminance signal may be executed at the stage of the color signal before being synthesized with the luminance signal, and the process at this stage of the color signal is executed at any stage of the color signal by the primary color and the color signal by the complementary color. May be. Incidentally, when these color signals are executed, processing with the configuration shown in FIG. 3 is executed for each color signal. In the following, flickering of a fluorescent lamp by a commercial power source with a frequency of 50 [Hz] will be described.

  The flicker reduction circuit 10 inputs the sequentially input image data D1 to the integration circuit 21, where the integration period is set to one horizontal scanning period and integration processing is performed in units of lines to reduce the influence of the pattern.

  That is, in the imaging result by flicker, the pixel value by flicker is superimposed on the pixel value by the pattern to obtain each pixel value, and thereby the pixel value In ′ (x, y) at an arbitrary pixel (x, y) is It is expressed by the following formula. Here, In (x, y) is a signal component that is a pixel value according to a pattern, Γn (y) × In (x, y) is a flicker component, and Γn (y) is a signal of a video signal. It is a flicker coefficient which shows the ratio of the flicker component with respect to a component.

  Here, since the blinking frequency of the light source is sufficiently longer than the horizontal scanning frequency, the flicker coefficient can be regarded as being constant in the same line of the same field. Accordingly, in the following, the flicker coefficient is Γn (y). Represented by In addition, the flicker coefficient Γn (y) is described in the following form in a form expanded to a Fourier series as appropriate, as shown by the following equation, thereby covering emission characteristics and afterglow characteristics that differ depending on the type of fluorescent lamp. Describe in a simplified format.

  Note that λo is the flicker wavelength on the display screen described above with reference to FIG. 6, and ωo is the normalized angular frequency normalized by the wavelength λo. γm is the amplitude value of each order (m = 1, 2, 3...) of the flicker component. Φm, n indicates each initial phase in the flicker component and is determined by the blinking cycle of the light source and the exposure timing of each pixel. In the case of a fluorescent lamp with a commercial power supply having a frequency of 50 [Hz], the initial phase Φm, n has the same value every three fields, and the difference in Φm, n from the immediately preceding field is expressed by the following equation. Is done.

  Thus, the integration circuit 21 performs the arithmetic processing of the following equation to calculate the integration value Fn (y).

  Note that αn (y) is an integral value for one line of the signal component In (x, y) as shown in the following equation.

  The memory 22 records and holds the integrated value Fn (y) obtained by the integrating circuit 21. Further, in response to the output of the integrated value Fn (y) by the integrating circuit 21, the memory 22 outputs the integrated values Fn_1 (y) and Fn_2 (y) by the corresponding lines in the previous field and the previous field.

  Therefore, when the subject does not move, the integral values Fn (y), Fn_1 (y), and Fn_2 (y) related to the corresponding lines in these continuous fields are only flicker components αn (y) × Γn (y). Thus, by calculating the difference value between these integrated values Fn (y), Fn_1 (y), and Fn_2 (y), the flicker component αn (y) × Γn (y) can be easily obtained. Can only detect. However, in a general subject, the signal component αn (y) changes in each field, which makes it difficult to extract only the flicker component by simply calculating the difference value. Further, even if the signal level of the flicker component is smaller than that of the signal component, it is difficult to extract only the flicker component by simply calculating the difference value.

  Therefore, in the flicker reduction circuit 10, the average value calculation circuit 23 includes the integration value Fn (y) output from the integration circuit 21, the previous field output from the memory 22, and the integration value Fn_1 ( y) and Fn_2 (y) are input, and an average value AVE [of integration values Fn (y), Fn_1 (y), Fn_2 (y) related to the corresponding lines of these three consecutive fields is calculated by the following equation. Fn (y)] is calculated.

  Here, in the case of a fluorescent lamp with a commercial power source having a frequency of 50 [Hz], the initial phase Φm, n has the same value every three fields, and thus the flicker component can be calculated by calculating the average value over three consecutive fields. Can be canceled out, so that the average value of only the signal components can be calculated. As a result, when the subject does not move in three consecutive fields, that is, when the following relational expression is established, only the signal component αn (y) can be detected as represented by the expression (6).

  When the movement of the subject is large, the relational expression (7) is not established, and it becomes difficult to detect only the signal component αn (y) by the calculation process of the expression (6). However, in such a case, by setting the continuous field related to the averaging process to a multiple of 3 fields, the effect of the motion is reduced by the low-pass filter action in the time axis direction to approximately approximate the signal component. Only αn (y) can be detected.

  The difference calculation circuit 24 subtracts the integration value Fn_1 (y) from the corresponding line in the previous field output from the memory 22 from the integration value Fn (y) output from the integration circuit 21, and is expressed by the following equation. The subtraction result Fn (y) −Fn_1 (y) is output.

  The normalization circuit 25 uses the subtraction result Fn (y) −Fn — 1 (y) calculated by the difference calculation circuit 24 as an average value AVE [Fn (y )], And the subtraction result Fn (y) −Fn — 1 (y) is normalized by the average value AVE [Fn (y)]. Accordingly, the normalization circuit 25 removes the influence of the signal component from the subtraction result Fn (y) −Fn_1 (y), and calculates the difference value gn (y) of the flicker coefficient Γn (y) −Γn (y).

  The DFT circuit 26 performs frequency analysis on the difference value gn (y) calculated by the normalization circuit 25 by discrete Fourier transform. Here, when the relational expression (3) is applied to the relational expression (9), the following relational expression can be obtained.

  However, | Am | and θm are the amplitude value and the initial phase of each order spectrum related to the flicker frequency in the normalized difference value gn (y), and are expressed by the following equations, respectively.

  Further, by using the amplitude value | Am | and the initial phase θm expressed by the equations (11) and (12), the following amplitude values γm and the initial values at the flicker frequency used in the equation (2) by the following equations, respectively: The phase Φm, n can be represented.

  Here, the discrete Fourier transform result of the order m is Gn (m), and the discrete Fourier transform [gn (y)] is expressed by the following equation. However, W is expressed by the equation (16), and L is the data length of the DFT operation. In this embodiment, W is the data length corresponding to the number of lines of one wavelength of flicker. In this way, if the data length of the DFT calculation is set to one wavelength of flicker (for L line), a discrete spectrum group that is an integral multiple of the normalized angular frequency ωo can be directly obtained. It can be simplified.

  Using the relational expression of the expression (15), each amplitude value γm and initial phase θm of the flicker frequency in the normalized difference value gn (y) represented by the expressions (11) and (12), Each n can be expressed by the following formula.

  Thereby, from the equations (17), (18), (13), and (14), the next amplitude value γm and the initial phase Φm, n at the flicker frequency can be obtained by the following equations, respectively.

  As a result, the DFT circuit 26 executes the discrete Fourier transform processing expressed by the equation (15) for each line based on the output value of the normalization circuit 25 using continuous L lines, and thereby extracts the spectrum at each order of the flicker frequency. . Further, by executing the arithmetic processing of the equations (19) and (20) using this processing result, the next amplitude value γm and the initial phase Φm, n at the flicker frequency are calculated for each line.

  In frequency analysis in digital signal processing, generally, fast Fourier transform (FFT) is used. However, since the fast Fourier transform requires that the data length be a power of 2, in this embodiment, frequency analysis is performed by discrete Fourier transform processing, and the data processing is simplified accordingly. In this case, the data to be processed and the data of the processing result may be processed, and the frequency analysis may be performed by fast Fourier transform.

  In this embodiment, high-order processing in the discrete Fourier transform is omitted as long as the flicker component can be approximated sufficiently in practice, thereby simplifying the processing.

  Incidentally, the amplitude value γm and the initial phase Φm, n obtained in this way also reflect the movement of the subject. It also varies depending on the exposure time and the speed of the electronic shutter. Thus, in this embodiment, the amplitude value γm and the initial phase Φm, n detected by the DFT circuit 26 are output to the system controller 5 and processed by the parameter control circuit 30 constituted by the system controller 5. Thus, in this embodiment, the amplitude value γm ′ and the initial phase Φm, n ′ that correctly reflect the flicker are calculated, and the amplitude value γm ′ and the initial phase Φm, n ′ resulting from the calculation result are supplied to the flicker generation circuit 28. input.

  The flicker generation circuit 28 outputs the flicker coefficient Γn (y) that correctly reflects the flicker by executing the calculation process of the equation (2) using the amplitude value γm ′ and the initial phase Φm, n ′. . Here, also in the calculation processing of the equation (2), under actual fluorescent lamp illumination, the flicker component can be sufficiently approximated practically by omitting high-order processing. As a result, in this embodiment, for example, the arithmetic processing of the expression (2) is executed in the range up to the second order, thereby simplifying the processing.

  Here, the equation (1) can be modified as the following equation. As a result, the arithmetic circuit 29 adds 1 to the flicker coefficient Γn (y) output from the flicker generation circuit 28 to calculate 1 + Γn (y), and then divides the image data by this calculated value. Suppresses the component.

  In the flicker component, the flicker generation circuit 28 starts in this case by being repeated in a cycle of three fields under a fluorescent lamp with a commercial power source having a frequency of 50 [Hz], for example, according to the blinking frequency of the light source. The flicker coefficient Γn (y) is calculated and output only for the three fields, and the first calculated flicker coefficient Γn (y) is retained in the memory and repeatedly output for the fields after the third field. Also good. In this way, the processing of the digital signal processing circuit 9 can be simplified.

  FIG. 1 is a functional block diagram showing a configuration of a parameter control circuit 30 configured by the system controller 5. The parameter control circuit 30 calculates a correction coefficient g for correcting the amplitude value γm by the initial phase Φm, n output from the DFT circuit 26, and corrects the amplitude value by the correction coefficient g to input the initial phase Φm. , N and the amplitude value γm ′ and the initial phase Φm, n ′ that correctly reflect the flicker are output. Therefore, in this parameter control circuit 30, the correction coefficient calculation circuit 31 calculates a correction coefficient g for correcting the amplitude value γm by the initial phase Φm, n output from the DFT circuit 26.

  That is, in the correction coefficient calculation circuit 31, the delay circuit 32 delays the initial phase Φm, n output from the DFT circuit 26 by one field period, and the subtraction circuit 33 outputs the initial phase output from the DFT circuit 26. The initial phase Φm, n output from the delay circuit 32 is subtracted from the phase Φm, n, thereby calculating the inter-field difference ΔΦm, n of the initial phase Φm, n between the corresponding lines.

  The low-pass filter (LPF) 34 smoothes and outputs the inter-field difference ΔΦm, n of the initial phase Φm, n output from the subtracting circuit 33, thereby causing excessive fluctuation of the inter-field difference ΔΦm, n due to disturbance. To reduce.

  The subtracting circuit 35 subtracts the theoretical value Δφm, n of the inter-field difference ΔΦm, n from the inter-field difference ΔΦm, n of the initial phase Φm, n output from the low-pass filter (LPF) 34, thereby the initial phase. For the inter-field difference ΔΦm, n of Φm, n, the phase ΔΦm, n−Δφm, n from the logical value Δφm, n is output. The theoretical value Δφm, n is obtained by the equation (3).

  The absolute value conversion circuit (ABS) 37 converts the phase value ΔΦm, n−Δφm, n from the logical value Δφm, n output from the subtraction circuit 35 into an absolute value and outputs it. Therefore, the output value ψm, n of the absolute value circuit 37 detected by the series of processing is that the amplitude value γm of the flicker component and the initial phase Φm, n obtained by the DFT circuit 26 correctly represent flicker. The smaller the value, the larger the amplitude value γm and the initial phase Φm, n of the flicker component are due to disturbance, movement of the subject, and the like.

  Accordingly, the gain calculation circuit 38, based on the output value ψm, n from the absolute value conversion circuit 37, corrects the correction coefficient g in which the initial phase between fields at the flicker frequency decreases as the distance from the logical value φm, n increases. Is generated. More specifically, as shown in FIG. 4, when the output value ψm, n from the absolute value conversion circuit 37 is smaller than the first threshold value thrA, the gain calculation circuit 38 outputs the correction coefficient g based on the value 1. When the output value ψm, n from the absolute value circuit 37 is larger than the second threshold value thrB, which is larger than the first threshold value thrA, the correction coefficient g with a value of 0 is output. Further, between these first threshold value thrA and second threshold value thrB, the correction coefficient g is set so that the value changes linearly corresponding to the output value ψm, n from the absolute value circuit 37. Output.

  The multiplication circuit 39 multiplies the spectrum amplitude value γm at the flicker frequency detected by the frequency analysis by the DFT circuit 26 by the correction coefficient g, thereby correcting the spectrum amplitude value γm with the correction coefficient g and outputting the result. To do.

  When the higher-order component (m ≧ 2) of the flicker frequency obtained by frequency analysis is also used for the flicker suppression processing, the parameter control circuit 5, the flicker generation circuit 28, and the arithmetic circuit 29 are provided for each subsequent component. Then, the correction coefficient g is calculated, and the pixel value of the image data D1 is finally corrected by the arithmetic circuit 29 by combining the respective correction coefficients g. In this case, after all, the correction amount by each correction coefficient g may be weighted and corrected according to the ratio of each component to the output value from the normalization circuit 25. In such processing, In each circuit block from the gain calculation circuit 38 of the parameter control circuit 5 to the calculation circuit 29 of the flicker reduction circuit 10, the correction may be performed with any one of the correction coefficient g, the amplitude value γm, and the flicker coefficient Γn (y).

  Thus, the system controller 5 calculates the correction coefficient g by executing the processing procedure shown in FIG. 5 for each line. In other words, when the frequency analysis result for one line is input from the digital signal processing circuit 10, the system controller 5 proceeds from step SP1 to step SP2, where it calculates the inter-field difference ΔΦm, n of the initial phase Φm, n. In the subsequent step SP3, the high frequency component of the inter-field difference ΔΦm, n is suppressed. In the subsequent step SP4, the difference value from the logical value φm, n of the inter-field difference ΔΦm, n is converted into an absolute value. Then, in the subsequent step SP5, the value of the correction coefficient g is calculated, and this processing procedure is performed in step SP6. finish.

(2) Operation of Example In the above-described configuration, in this imaging apparatus 1 (FIG. 2), an optical image of a desired subject is formed on the imaging surface of the imaging device 4 by the lens 2, and this optical image is formed by the imaging device 4. The image pickup signal S1 is output after the photoelectric conversion process. In the image pickup apparatus 1, the image pickup signal S 1 is subjected to signal processing by the analog signal processing circuit 7, then converted to image data D 1 by the analog / digital conversion circuit 8, and the image data D 1 is signal processed by the digital signal processing circuit 9. As a result, a video signal SV based on a result of imaging a desired subject is generated. In the imaging apparatus 1, the flicker component of the video signal SV is suppressed in the processing of the image data D1 in the digital signal processing circuit 9.

  That is, the image data D1 is sequentially integrated line by line by the integration circuit 21 in the flicker reduction circuit 10 of the digital signal processing circuit 9 (FIG. 3), and between the integration periods corresponding to adjacent fields in the difference calculation circuit 24. Thus, the integral value is subtracted, whereby a change in signal level between successive fields is detected in line units. Further, in the normalization circuit 25, the change in the signal level is divided by the integral value, whereby the change in the signal level between successive fields is normalized, and the change in the signal level between successive fields is converted into a video signal. Detected in line units with corresponding proportions.

  Thus, the change in the signal level detected in this manner is detected by including the change in the signal level due to the movement of the subject in the change in the signal level due to the flicker, but the change in the signal level due to the flicker is detected. In such a case, it periodically changes depending on the flicker frequency.

  As a result, in the image data D1, the change in signal level between successive fields in units of lines detected in this way is subjected to frequency analysis by the DFT circuit 26, and the spectrum at the flicker frequency is detected for each line. Thus, in the amplitude value of the spectrum at the flicker frequency detected in this way, the change in signal level between consecutive fields is normalized, so that Of the flicker component. Thus, it is considered that the flicker can be suppressed by correcting the signal level of the original video signal SV with this amplitude value. In addition, in the flicker component, a higher-order frequency spectrum of the flicker frequency is also generated, and by adding the amplitude value related to the higher-order frequency spectrum to the correction of the video signal within a practically sufficient range, the flicker frequency is further increased. It is considered that flicker can be suppressed with high accuracy.

  However, in such actual shooting of a subject, the spectrum may be detected by such a flicker frequency due to the movement of the subject, and thus the video signal may be erroneously corrected depending on the shooting conditions. appear.

  As a result, in the imaging apparatus 1 (FIG. 1), the amplitude value and the initial phase in each line are notified to the system controller 5 with respect to the spectrum of the flicker frequency detected in this way. Here, the delay circuit 32 and the subtraction circuit 33 are notified. Thus, a change in the initial phase is detected between consecutive fields. Here, in the change of the initial phase detected in this way, when the change in the signal level in the video signal is due to flicker, the value is based on the logical value determined by the blinking frequency of the light source, whereas the movement of the subject In the case of the above, the value differs from the logical value depending on the degree.

  As a result, in the imaging apparatus 1, the system controller 5 generates a correction coefficient whose value decreases as the change in the initial phase between the fields moves away from the logical value, and the signal level of the image data D1 is corrected by this correction coefficient to flicker. Is reduced.

  As a result, the imaging apparatus 1 can correctly reduce flicker by simple processing even when the shooting conditions change in various ways.

  As a result, in the imaging apparatus 1, after the initial phase change between the fields is smoothed by the low-pass filter 34, the subtraction circuit 35 calculates a difference value from the logical value, and the difference value is converted into the absolute value conversion circuit 37. Is converted to an absolute value. Further, the gain calculation circuit 38 generates a correction coefficient whose value decreases as the change in the initial phase between fields becomes farther from the logical value, based on this absolute value. In the imaging apparatus 1, the amplitude value of the corresponding spectrum is corrected by the correction coefficient and is output to the flicker reduction circuit 10 together with the corresponding initial phase. In the flicker reduction circuit 10, the corrected amplitude value and the corresponding initial phase are output. Thus, a flicker coefficient that correctly reflects the flicker component is generated, and the signal level of the image data D1 is corrected by the flicker coefficient to reduce the flicker.

  In the processing relating to these correction coefficients, in the imaging apparatus 1, the multiplication circuit 39 corrects the amplitude value of the spectrum at the flicker frequency detected by the frequency analysis means with the correction coefficient, and then the correction is performed in the flicker generation circuit 28. The flicker coefficient indicating the ratio of the flicker component to the video signal is calculated based on the amplitude value, and the signal level of the video signal is corrected by the flicker coefficient. Correction can reduce flicker.

  In addition, the correction coefficient is generated by smoothing the change of the initial phase at the flicker frequency, so that erroneous generation of the correction coefficient due to disturbance can be effectively avoided.

  On the other hand, in the normalization process, the integration results in units of lines are averaged between consecutive fields, and the normalization process is executed using the average value obtained as a result. False correction is effectively avoided.

(3) Advantages of the embodiment According to the above configuration, the change in the signal level is subjected to frequency analysis to detect the spectrum of the flicker frequency, and the change in the initial phase of this spectrum in the continuous field or frame is determined to determine the video signal. By suppressing the flicker component, flicker can be reliably reduced by a simpler process than in the prior art.

  In this series of processing, after correcting the amplitude value of the spectrum at the flicker frequency with the correction coefficient, the flicker coefficient indicating the ratio of the flicker component to the video signal is calculated based on the corrected amplitude value. By correcting the signal level, it is possible to reduce the flicker by correcting the signal level of the video signal correctly using the frequency analysis result effectively.

  Further, by generating the correction coefficient by smoothing the change of the initial phase at the flicker frequency, it is possible to effectively avoid the erroneous generation of the correction coefficient due to disturbance.

  Further, by averaging the integration results in line units between consecutive fields and performing normalization processing using the average value obtained as a result, it is possible to effectively avoid flicker error correction due to motion.

  In the above-described embodiment, a case has been described in which a difference value is calculated between fields from the integration result in units of lines, and the difference value is normalized to perform frequency analysis. However, the present invention is not limited to this, and is practically sufficient. If a sufficient characteristic can be ensured, the line-by-line integration result may be normalized to perform frequency analysis.

  In the above embodiment, the case where the video signal is integrated and processed in units of one line has been described. However, the present invention is not limited to this, and an integration period is set longer than one horizontal scanning period. The video signal may be integrated and processed in units.

  In the above-described embodiment, the case where various difference values are calculated between fields has been described. However, the present invention is not limited to this, and these difference values may be calculated between frames.

  In the above-described embodiment, the case where the imaging result is acquired at the field frequency of 60 [Hz] has been described. However, the present invention is not limited to this, and the imaging result is acquired at the frame frequency of 50 [Hz]. Can be widely applied.

  In the above-described embodiment, the case where the imaging result is acquired and processed by the CMOS image sensor has been described. However, the present invention is not limited to this, and when an XY address scanning type image sensor other than the CMOS image sensor is used. Can also be widely applied.

  In the above-described embodiments, the case where the present invention is applied to the imaging apparatus and the imaging result is processed in real time has been described. However, the present invention is not limited thereto, and such an imaging result is processed by a computer, an editing apparatus, or the like. It can be widely applied to processing. In these cases, the configuration of the flicker reduction circuit of the digital signal processing circuit according to the above-described embodiment and the parameter control circuit of the system controller may be integrally configured by the processing of the arithmetic processing means.

  The present invention relates to a video signal processing device, a video signal processing method, a video signal processing program, and a recording medium on which a video signal processing program is recorded, and is applied to, for example, a video camera using an XY address scanning type imaging device. Can do.

It is a functional block diagram which shows the structure of the parameter control part in the system controller which concerns on Example 1 of this invention. 1 is a block diagram illustrating an imaging apparatus according to Embodiment 1 of the present invention. FIG. 3 is a functional block diagram illustrating a configuration of a flicker reduction circuit in a digital signal processing circuit in the imaging apparatus of FIG. 2. It is a characteristic curve figure with which it uses for description of operation | movement in the parameter control part of FIG. It is a flowchart which shows the process sequence of the system controller which concerns on the structure of FIG. It is a top view with which it uses for description of a flicker.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Imaging device, 4 ... Imaging device, 5 ... System controller, 9 ... Digital signal processing circuit, 10 ... Flicker reduction circuit, 21 ... Integration circuit, 22 ... Memory, 23 ... Average value calculation Circuit, 24... Difference calculation circuit, 25... Normalization circuit, 26... DFT circuit, 28... Flicker generation circuit, 29. 35... Subtracting circuit, 34... Low pass filter, 37... Absolute value circuit, 38... Gain calculation circuit, 39.

Claims (8)

Video signal acquisition means for acquiring a video signal;
Integration means for integrating the video signal and outputting an integration result in units of integration periods of one or more lines of the video signal;
A difference value of the integration result with the corresponding integration period in an adjacent field or frame, or normalizing means for normalizing the integration result;
Frequency analysis means for analyzing the processing result of the normalization means, and detecting the amplitude value and initial phase of the spectrum at the flicker frequency for each integration period;
Phase change detecting means for detecting a change in the initial phase between the corresponding integration periods of the adjacent fields or frames;
Correction coefficient calculation for generating a correction coefficient whose value decreases as the detection result moves away from the logical value, based on a difference value between the detection result by the phase change detection means and the logical value of the initial phase change by flicker And a video signal correcting means for correcting the signal level of the video signal based on the correction coefficient. The video signal correcting means includes
Amplitude correction means for correcting the amplitude value of the spectrum at the flicker frequency detected by the frequency analysis means by the correction coefficient;
Flicker generation means for calculating a flicker coefficient indicating a ratio of a flicker component to the video signal based on the amplitude value corrected by the amplitude correction means and the initial phase;
The video signal processing apparatus according to claim 1, further comprising an arithmetic unit that corrects a signal level of the video signal based on a flicker coefficient calculated by the flicker generation unit. The normalizing means includes
Averaging means for averaging the integration results between the successive fields or frames;
A division value for outputting a normalization processing result by dividing the difference value of the integration result from the corresponding integration period in an adjacent field or frame, or by dividing the integration result by the average value calculated by the averaging means The video signal processing apparatus according to claim 1, further comprising: means. The correction coefficient calculation means includes
The video signal processing apparatus according to claim 1, wherein the correction coefficient is generated by smoothing a change in an initial phase at the flicker frequency detected by the phase change detection unit. The video signal acquisition means includes
The video signal processing apparatus according to claim 1, wherein the video signal processing apparatus is an imaging unit that uses an XY address scanning type imaging device that images a desired subject and outputs the video signal based on the imaging result. A video signal acquisition step of acquiring a video signal;
An integration step of integrating the video signal and outputting an integration result in units of integration periods of one or more lines of the video signal;
A difference value of the integration result with the corresponding integration period in an adjacent field or frame, or a normalizing step of normalizing the integration result;
Frequency analysis of the processing result of the normalization step, and for each integration period, a frequency analysis step of detecting an amplitude value and an initial phase of a spectrum at a flicker frequency;
A phase change detection step of detecting a change in the initial phase between the corresponding integration periods of the adjacent fields or frames;
A correction coefficient that generates a correction coefficient whose value decreases as the detection result moves away from the logical value, based on a difference value between the detection result of the phase change detection step and the logical value of the initial phase change due to flicker Calculation steps,
And a video signal correction step of correcting a signal level of the video signal based on the correction coefficient. In a video signal processing program for reducing flicker of a video signal by causing an arithmetic processing means to execute a predetermined processing procedure,
The processing procedure is as follows:
A video signal acquisition step of acquiring the video signal;
An integration step of integrating the video signal and outputting an integration result in units of integration periods of one or more lines of the video signal;
A difference value of the integration result with the corresponding integration period in an adjacent field or frame, or a normalizing step of normalizing the integration result;
Frequency analysis of the processing result of the normalization step, and for each integration period, a frequency analysis step of detecting an amplitude value and an initial phase of a spectrum at a flicker frequency;
A phase change detection step of detecting a change in the initial phase between the corresponding integration periods of the adjacent fields or frames;
A correction coefficient that generates a correction coefficient whose value decreases as the detection result moves away from the logical value, based on a difference value between the detection result of the phase change detection step and the logical value of the initial phase change due to flicker Calculation steps,
And a video signal correction step of correcting a signal level of the video signal based on the correction coefficient. In a recording medium recorded with a video signal processing program for reducing flicker of a video signal by causing an arithmetic processing means to execute a predetermined processing procedure,
The processing procedure is as follows:
A video signal acquisition step of acquiring the video signal;
An integration step of integrating the video signal and outputting an integration result in units of integration periods of one or more lines of the video signal;
A difference value of the integration result with the corresponding integration period in an adjacent field or frame, or a normalizing step of normalizing the integration result;
Frequency analysis of the processing result of the normalization step, and for each integration period, a frequency analysis step of detecting an amplitude value and an initial phase of a spectrum at a flicker frequency;
A phase change detection step of detecting a change in the initial phase between the corresponding integration periods of the adjacent fields or frames;
A correction coefficient that generates a correction coefficient whose value decreases as the detection result moves away from the logical value, based on a difference value between the detection result of the phase change detection step and the logical value of the initial phase change due to flicker Calculation steps,
And a video signal correction step of correcting a signal level of the video signal based on the correction coefficient. A recording medium recording a video signal processing program.

Images