JP2004118047A - Image processing apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce a flicker phenomenon in moving picture display. <P>SOLUTION: An image processing apparatus is equipped with an image storage part which stores an input image signal (a) inputted at specified time t<SB>1</SB>, a 1st multiplier which multiplies an input image signal (b) inputted one frame after the specified time t<SB>1</SB>by a specified coefficient α (0≤α<1), a 2nd multiplier which multiplies the stored input image signal by a coefficient (1-α), and an adder which sums up the α-multiplied signal (b) and (1-α)-multiplied signal (a), and outputs the sum image signal obtained by the adder. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an image processing apparatus, and more particularly, to an image processing apparatus for reducing a flicker phenomenon when displaying a moving image.
[0002]
[Prior art]
2. Description of the Related Art In a liquid crystal display device (hereinafter, referred to as an LCD), flicker occurs when a moving image is displayed due to various causes such as a type of data to be displayed, a response of the LCD, luminance information of application software, and a drawing speed.
One method of reducing the occurrence of flicker is to reduce the luminance difference between frames of data that display one screen. However, this method cannot faithfully reproduce color space information that should be originally expressed. Will be.
[0003]
Further, when the frequency of the image signal input to the LCD is different from the display frequency of the LCD, the temporal order of the video is reversed at an irregular cycle, so that a lack of smoothness in moving image display is required. was there.
[0004]
In a conventional image processing apparatus, an apparatus for eliminating the unnaturalness of the moving image display has been proposed (for example, refer to Patent Document 1).
FIG. 10 shows a configuration diagram of a frequency conversion device of this patent document proposed to obtain a smooth moving image. In FIG. 10, reference numeral 201 denotes an input terminal of an image signal, 202 and 206 denote frame buffers (image memories), 203 and 205 denote coefficients (1−α), multipliers with α (0 ≦ α <1), and 204, respectively. It is an adder.
[0005]
At a certain time t, the image signal u (t) input to the input terminal 201 is stored in the frame buffer 202, read out sequentially thereafter, and multiplied by the coefficient (1−α) in the multiplier 203. The multiplied image signal (1-α) u (t) is input to the adder 204.
[0006]
On the other hand, the input image signal at time t−1 corresponding to one frame before is stored in the frame buffer 206. The adder 204 multiplies the multiplied image signal (1−α) u (t) and the image signal F (t−1) at time t−1 one frame before read out from the frame buffer 206 by a multiplier. In 205, the signal (αF (t−1)) multiplied by the coefficient α is added. The added signal is stored in the frame buffer 206.
Here, the signal multiplied by the coefficient α by the multiplier 205 is a residual image component, and a visually smoother moving image is realized by adding the residual image component by the adder 204.
[0007]
Further, a motion level determination circuit is used to detect the amount of motion of a moving image. However, if a spike noise component is present in the input image signal, it is determined that the image is a moving image despite being a still image. There was a malfunction. There has been proposed a technique for detecting a pixel at an isolated point into which such spike-like noise is mixed to prevent a malfunction in the motion level determination processing (for example, see Patent Document 2).
[0008]
[Patent Document 1]
JP-A-9-101765 [Patent Document 2]
JP-A-1-290389
[Problems to be solved by the invention]
However, since the image signal stored in the frame buffer is used as an output image signal as it is in the above-described Patent Document 1, an image memory that stores a complete image signal for at least one frame is used. It had to be provided, which increased the cost.
[0010]
Further, the technique described in Patent Document 2 is an effective technique for removing an isolated point where spike noise in moving image display has occurred, but cannot reduce a flicker phenomenon caused by an image signal between frames. .
[0011]
Therefore, the present invention has been made in view of the above circumstances, and even when the frequency of the input image signal is different from the display frequency of the image display device, without impairing the color reproducibility of the still image display. It is another object of the present invention to provide an image processing apparatus capable of reducing flicker development in displaying a moving image.
Another object of the present invention is to provide an image processing apparatus which can obtain an afterimage effect of an image of one frame before without storing all image signals of one frame.
It is still another object of the present invention to provide an image processing apparatus capable of obtaining a smoother afterimage effect on a moving image by correcting an image signal in consideration of an image signal of a peripheral pixel of a changed pixel.
[0012]
[Means for Solving the Problems]
The present invention relates to an image storage unit and the input image signal b a predetermined coefficient α multiplied inputted after one frame of the predetermined time t ₁ (0 for storing the input image signal a is input at a predetermined time t ₁ ≤α <1), a second multiplier for multiplying the stored input image signal a by a factor (1−α), and a signal b multiplied by α and (1−α) multiplied. A first adder for adding the signal a and outputting an image signal added by the first adder.
According to this, since the input image signals of two temporally adjacent frames are added after being multiplied by a predetermined coefficient, the color reproducibility at the time of displaying a still image is not impaired. Flicker can be reduced.
[0013]
The input image signal stored in the image storage unit may be a predetermined number of upper bits of each pixel constituting one frame. According to this, the storage capacity of the image storage unit that stores the image signal can be saved.
For example, in the case of FIG. 6 of a second embodiment described later, only 2 bits of an input image signal composed of 8 bits need be stored, so that the storage capacity can be reduced to about 1/4. .
[0014]
Further, the present invention compares the input image signal b with an input image signal a stored in an image storage unit, detects a changed pixel, and calculates a coefficient β corresponding to a changed state. A comparison unit, a filter processing unit that calculates an average value of the changed pixel and a predetermined peripheral pixel of each of the changed pixels detected by the image comparison unit, and a coefficient ( A third multiplier for multiplying the image signal c output from the first adder by a factor β, an average value multiplied by (1-β), and A second adder for adding the image signal c to the image signal c, and outputting the pixel signal d added by the second adder.
According to this, since a smoother afterimage effect can be obtained, flicker of moving image display can be further reduced.
Here, eight pixels adjacent to the changed pixel may be used as the peripheral pixels of the changed pixel.
Further, the coefficient β for the changed pixel may be set to 0 <β <1, and the coefficient β for the unchanged pixel may be set to β = 1.
[0015]
In the following embodiments, the image storage unit is called an image memory. As the image storage unit, a rewritable storage element such as a RAM, a field memory, and a FIFO is used. Further, in order to obtain a high processing speed, each component is preferably configured by hardware logic including a logical operation element such as a buffer, a counter, and a flip-flop. In the present invention, the image signal refers to a luminance signal representing a luminance level given to each pixel of the display device.
[0016]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, the present invention will be described in detail based on an embodiment shown in the drawings. Note that the present invention is not limited by this.
<Example 1>
FIG. 1 is an overall configuration diagram of an image processing apparatus according to a first embodiment of the present invention.
In FIG. 1, an image processing device 106 includes an image input unit 102, an image processing unit 103, and an image correction unit 104.
The image output device 101 is a device that generates an image to be displayed, and for example, a personal computer or the like can be used.
The image display device 105 is a display device such as a liquid crystal display device (LCD) or a CRT, and receives and displays an image signal converted for display by the image processing device 106.
[0017]
The present invention has a feature in the configuration of the image processing apparatus 106 that performs a correction process or the like for reducing flicker of a moving image, and particularly in the configuration of the image correction unit 104.
An image input unit 102 of the image processing device 106 is a unit that receives an image signal provided from the image output device 101 and converts the image signal into a signal that can be processed by the image processing unit 103 at the next stage. And a process for converting a differential digital signal such as DVI into a digital signal.
The image processing unit 103 is a part that converts the image signal converted by the image input unit 102 into a signal having the same frequency as the display frequency indicating the basic display timing in the image processing device 105. When the input frequency and the output frequency are different from each other in the frequency conversion in the image processing unit 103, the temporal order of the video is inverted at an irregular cycle which causes a lack of smoothness when displaying a moving image.
As will be described in detail below, the image correction unit 104 is a unit that performs correction for reducing the flicker phenomenon of a moving image by generating an afterimage effect on the image signal whose frequency has been converted by the image processing unit 103.
[0018]
FIG. 2 is a configuration diagram of an embodiment of the image correction unit 104 of the image processing device 106 according to the present invention.
As shown in FIG. 2, the image correction unit 104 includes an image memory 112, multipliers 113 and 114, and an adder 115.
Here, the image memory 112 corresponds to an image storage unit, the multiplier 113 corresponds to a first multiplier, the multiplier 114 corresponds to a second multiplier, and the adder 115 corresponds to a first adder.
[0019]
The multiplier 113 multiplies the input image signal 111 by a coefficient α, and the multiplier 114 multiplies the signal read from the image memory 112 by a coefficient (1−α). Here, the coefficient α is set to 0 ≦ α <1.
The input image signal 111 converted to the same frequency as the display frequency of the image display device 105 by the image processing unit 103 is input to the image memory 112 of the image correction unit 104 and the multiplier 113. The image memory 112 stores the image signal u (t) at a certain time t.
[0020]
On the other hand, the image signal 111 input to the multiplier 113 is multiplied by a coefficient α, and given to the adder 115. That is, the signal αu (t) is provided.
Further, the input image signal 111 (u (t-1)) input and stored one frame before (time t-1) is read out from the image memory 112, and given to the multiplier 114, where the coefficient ( 1-α) is multiplied. That is, (1−α) u (t−1) is output from the multiplier 114 and provided to the adder 115. In the adder 115, the signal αu (t) from the multiplier 113 and the signal (1−α) u (t−1) from the multiplier 114 are input at the same timing and added.
[0021]
Therefore, αu (t) + (1−α) u (t−1) is output from the adder 115 as the output image signal 116.
Here, the output image signal 116 includes an input signal component one frame before, so that a residual effect can be obtained and flicker at the time of displaying a moving image can be reduced.
[0022]
FIG. 3 is an explanatory diagram of an image signal correction process performed by each component of the image correction unit 104 in FIG.
Here, a case is considered where an image signal composed of 5 × 5 pixels is corrected. In FIG. 3, white pixels are pixels having a luminance level of zero, and black pixels are pixels having a certain positive value whose luminance level is not zero. For example, in FIG. 3, the input image signal 111 (u (t)) at a certain time t indicates that the lower left ten pixels have a certain luminance level value.
[0023]
The image signal 117 (u (t−1)) output from the image memory 112 indicates the luminance level of the pixel one frame before, and the 16 pixel portions other than the pixels located at the top and right are , A pixel having a certain luminance level.
Since the coefficient α is 0 ≦ α <1, the image signal 118 (αu (t)) after passing through the multiplier 113 is obtained by multiplying the luminance level of each pixel by α as shown in the figure. .
Similarly, the image signal 119 ((1−α) u (t−1)) after passing through the multiplier 114 has a luminance level of each pixel multiplied by (1−α).
[0024]
In the adder 115, the two types of image signals 118 and 119 are added for each pixel to form an output image signal 116 as shown.
In the output image signal 116, pixels having no change in luminance level between the time t and the time t−1 can obtain the same image signal as before the coefficient is applied, but a portion where the luminance level has changed, that is, The six pixels in the portion where the luminance level has changed to zero at time t are output as image signals multiplied by the coefficient (1−α). From this, it can be seen that in a still image that does not change with time, the same image signal as before the coefficient is applied is obtained because the luminance level does not change even with time, and it can be displayed without impairing color reproducibility.
[0025]
When such correction is not performed, the display changes from the image signal 117 at the time t-1 to the input image signal 111 at the next time t, and this change is expressed as flicker of the moving image. By performing the above-described correction of the present invention, the display changes from the image signal 117 at the time t-1 to the output image signal 116 including the pixel of the intermediate luminance level at the next time t. It can be seen that a smooth display with less flicker can be displayed.
[0026]
<Example 2>
FIG. 4 is a configuration diagram of the image correction unit 104 according to the second embodiment of the present invention.
Here, a smoother moving image display is performed by performing a filter process in addition to the correction process of the first embodiment.
4, the image correction unit 104 according to the second embodiment further includes line memories 406 and 407, a 3 × 3 filter processing unit 408, an image comparison unit 409, multipliers 410 and 411, and an adder 412.
[0027]
The multiplier 410 multiplies the coefficient β, and the multiplier 411 multiplies the coefficient (1−β). Here, the coefficient β is 0 <β ≦ 1.
The multiplier 410 corresponds to the fourth multiplier, the multiplier 411 corresponds to a third multiplier, and the adder 412 corresponds to a second adder.
Further, in FIG. 4, the filter processing unit is constituted by the 3 × 3 filter processing unit 408 and the two line memories 406 and 407.
[0028]
To configure a 3 × 3 filter, two line memories are required. The line memory 1 (406) stores one line of the input image signal 111, and the line memory 2 (407) further includes This stores data read from the line memory 1 (406).
The 3 × 3 filter processing unit 408 captures the input image signal 111 and image signals of eight pixels around the input image signal 111, calculates an average value Uav (t) of the image signals of these nine pixels, and outputs the average value Uav (t). FIG. 5 is a schematic explanatory diagram of the 3 × 3 filter processing unit 408 according to the second embodiment.
[0029]
Three image signals are input to the 3 × 3 filter processing unit 408. For example, _assuming that an image signal at a certain time t with a central pixel of interest (m rows, n columns) _{Am, n} is input as the input image signal 111, the input image signal 501 from the line memory 1 (406) , Data one line before the input image signal 111 is input, and data two lines before the input image signal 111 are input as the input image signal 502 from the line memory 2 (407).
The end 3 × 3 filter processing unit 408, pixel _{A m,} and an image signal of _n, the pixels _{A m,} an image signal of the eight peripheral pixels of _{n (from A m-1, n-1} to _{A m + 1, n + 1} ) Is input. Then, an average value of a total of nine image signals is calculated and output.
[0030]
The image comparison unit 409 is a part that determines the coefficient β, and the determined coefficient is provided to the multipliers 410 and 411. The coefficient β compares the input image signal 111 (u (t)) with the image signal u (t−1) of one frame before read from the image memory 112, and determines whether the comparison result is the same. Is determined.
For example, if the signal u (t) and the signal u (t-1) are the same, β = 1, and if different, β is determined in the range from 0 to 1.
[0031]
Next, a correction process according to the second embodiment will be described.
As in the first embodiment, the image signal converted into a signal that can be displayed by the image input unit 102 and the image processing unit 103 is input as the input image signal 111 to the image correction unit 104 in FIG.
The input image signal 111 (u (t)) is subjected to coefficient multiplication (113, 114) and addition (115) as in the first embodiment, and the output image signal 116 is provided to the multiplier 410. Multiplier 410 multiplies output signal 116 by coefficient β and outputs the result to adder 412.
[0032]
Further, the input image signal 111 (u (t)) and the image signal 421 (u (t−1)) one frame before (time t−1) read from the image memory 112 are sent to the image comparison unit 409. Entered and compared.
After this comparison, the coefficient β is determined by the above-described processing, and is provided to the multipliers 410 and 411.
On the other hand, the average value signal Uav (t) calculated by the 3 × 3 filter processing 408 is provided to the multiplier 411, and is multiplied by the coefficient (1−β).
Further, in the adder 412, the output image signal 116 multiplied by β and the average value signal Uav (t) multiplied by (1−β) are added and output to the image display device 105 as an output image signal 422. .
In the second embodiment, by further multiplying the corrected output image signal 116 of the first embodiment by a predetermined coefficient (β), a smoother afterimage effect can be obtained, and flicker in moving image display can be reduced. .
[0033]
Here, when the two image signals (111 and 421) input to the image comparison unit 409 are the same, β = 1, and the same image signal 116 as in the first embodiment is output.
If the two image signals (111 and 421) are different, β becomes a value of 0 <β <1, so that the β-multiplied image signal 116 is added to the signal (1-β) Uav (t) from the multiplier 411. Is the output image signal 422.
[0034]
FIG. 7 is a diagram illustrating an image signal correction process performed by each component of the image correction unit 104 according to the second embodiment.
Here, an image signal state of 5 × 5 pixels is shown as an example. In FIG. 7, the pixel states denoted by reference numerals 111, 117, 118, 119 and 116 are the same as those shown in FIG. In FIG. 7, the 3 × 3 filter processing unit 408 performs a process of calculating the average value of the pixel signals of the nine pixels as described above for each pixel.
[0035]
The filter processing unit 408 shown in FIG. 7 shows an average value calculated only for representative six pixels. For example, in the case of averaging nine pixels centering on the pixel at (3 rows, C columns), an average value Uav (t) of three black pixels and six white pixels is calculated. The average value Uav (t) is the image signal value of the pixel at (3 rows, C columns).
As described above, the average value Uav (t) calculated for each pixel is input to the adder 411 and multiplied by the coefficient (1−β).
[0036]
Here, the coefficient β is determined by the image comparison unit 409 shown in FIG. 4. The pixel having a change has a value of 0 <β <1, but the pixel having no change has β = 1.
That is, since the coefficient 1-β in the multiplier 411 becomes zero in a pixel that has not changed, the output from the multiplier 411 also becomes zero. Therefore, a non-zero pixel signal is output from the multiplier 411 only to a pixel that has changed.
This is intended to emphasize the afterimage effect only in a pixel that has changed in consideration of pixel signals of surrounding pixels.
[0037]
On the other hand, the image signal 116 obtained in the same manner as in the first embodiment is provided to the multiplier 410 and multiplied by the coefficient β.
Here, as described above, since β = 1 in a pixel that has not changed, the image signal 116 is output as it is.
In addition, since 0 <β <1, the image signal multiplied by the coefficient β is output from the multiplier 410 in the pixel that has changed.
[0038]
This is to obtain an afterimage effect by further multiplying the changed pixel by the coefficient β. Then, the image signals of the two multipliers 410 and 411 are simply added by the adder 412 for each pixel, and become an output image signal 422 as shown in FIG.
Comparing the output image signal 116 obtained in the first embodiment with the output image signal 422 obtained in the second embodiment, the output image signal 422 shows the change of the image signal for the pixel at the boundary portion where the luminance change is larger. Can be said to be smoother, so that the afterimage effect is more emphasized. That is, the flicker phenomenon of the moving image display can be further reduced.
[0039]
Although image signals for all pixels for one frame are stored in the image memory 112 shown in FIG. 4, if the image comparison unit 409 only detects a motion, some of the input image signals 111 (For example, two bits Dt7 and Dt6 in FIG. 6).
In this case, since only the upper bits (Dt7, Dt6) of each image signal need be stored in the image memory 112, the storage capacity of the image memory 112 can be saved.
[0040]
FIG. 6 shows a configuration in which the upper two bits (Dt7, Dt6) of the input image signal 111 are stored in the image memory 112 and only the upper two bits are provided to the image comparison unit 409 in the second embodiment.
Here, the image comparison unit 409 stores the bits Dt7 and Dt6 at time t and the bits Dt7 (t-1) and Dt6 (t-1) at time t-1 one frame before and read from the image memory 112. Is input.
<Example 3>
[0041]
FIG. 8 is a configuration diagram of the image correction unit 104 according to the third embodiment of the present invention.
Here, the basic elements constituting the image correction unit 104 are the same as those in the second embodiment, but the image signals input to the 3 × 3 filter processing unit 408 and the two line memories 406 and 407 are stored in the image memory 112. The second embodiment is different from the second embodiment in that the signal is 421 (u (t-1)) which is the signal of one frame before and read out from the frame. Also in the third embodiment, 0 ≦ α <1, 0 <β ≦ 1.
[0042]
In FIG. 8, the image signal 421 (u (t-1)) one frame before is input to the 3 × 3 filter processing unit 408 and the two line memories 406 and 407, and the 3 × 3 filter processing unit 408 The average value 423 (Uav (t-1)) of the nine pixels as described above is calculated.
This average value 423 (Uav (t−1)) is provided to the multiplier 411, multiplied by a factor of 1−β, and provided to the adder 412.
Therefore, the only difference is whether the image signal for which the average value is to be obtained is at time t or at time t−1 one frame before, but an afterimage effect can be obtained as in the second embodiment. Therefore, flicker at the time of displaying a moving image can be reduced.
[0043]
<Example 4>
FIG. 9 is a configuration diagram of the image correction unit 104 according to the fourth embodiment of the present invention.
Here, the configurations and processing contents of the two line memories 406 and 407, the 3 × 3 filter processing unit 408, the image comparison unit 409, the multipliers 410 and 411, and the adder 412 are the same as those in the second embodiment.
However, the configurations of the image memory 112 and its peripheral circuits 802, 803, and 805 are different from those of the first and second embodiments.
[0044]
That is, the input image signal 111 is first input to the multiplier 802, multiplied by a factor (1−α), and input to the adder 803.
The image signal (u (t-1)) of the previous frame is stored in the image memory 112, but the image signal (u (t-1)) of the previous frame read from the image memory 112 is , Is input to the multiplier 805, multiplied by the coefficient α, and fed back to the adder 803.
[0045]
Then, the adder 803 adds the output signal (1-α) u (t) of the multiplier 802 and the output signal αu (t−1) of the multiplier 805, and stores the added signal in the image memory 112. Is done.
Here, also in the fourth embodiment, 0 ≦ α <1 and 0 <β ≦ 1.
In the fourth embodiment, in particular, since the image signal obtained by feeding back the image signal one frame before and the current input image signal are multiplied by the respective coefficients and added, it can be said that time response can be further provided. Therefore, the effect that the afterimage effect can be added can be obtained.
[0046]
【The invention's effect】
According to the present invention, since the correction is performed such that the input image signal and the image signal one frame before the input image signal are each multiplied by a coefficient, the color reproducibility at the time of displaying a still image is corrected. Can be reduced without impairing the image quality.
In addition, since the average value including the image signals of the peripheral pixels is obtained for the changed pixel between temporally adjacent frames and the image signal of the changed pixel is corrected, a smooth afterimage effect is obtained, Flicker at the time of display can be further reduced.
[Brief description of the drawings]
FIG. 1 is an overall configuration diagram of an image processing apparatus according to a first embodiment of the present invention.
FIG. 2 is a configuration diagram of an image correction unit of the image processing apparatus according to the first embodiment of the present invention.
FIG. 3 is an explanatory diagram of pixel state conversion in Embodiment 1 of the present invention.
FIG. 4 is a configuration diagram of an image correction unit of an image processing apparatus according to a second embodiment of the present invention.
FIG. 5 is an explanatory diagram of a 3 × 3 filter processing unit according to a second embodiment of the present invention.
FIG. 6 is an explanatory diagram of signals provided to an image memory and an image comparison unit according to the second embodiment of the present invention.
FIG. 7 is an explanatory diagram of pixel state conversion in Embodiment 2 of the present invention.
FIG. 8 is a configuration diagram of an image correction unit of an image processing device according to a third embodiment of the present invention.
FIG. 9 is a configuration diagram of an image correction unit of an image processing apparatus according to a fourth embodiment of the present invention.
FIG. 10 is a configuration diagram of a frequency conversion device in a conventional image processing device.
[Explanation of symbols]
101 image output device 102 image input unit 103 image processing unit 104 image correction unit 105 image display device 106 image processing device 111 input image signal 112 image memory 113 multiplier 114 multiplier 115 adder 116 output image signal

Claims (5)

An image storage unit for storing the input image signal a is input at a predetermined time t _1, the input image signal b by a predetermined coefficient to be input one frame after the predetermined time t ₁ alpha times (0 ≦ α <1 ), A second multiplier for multiplying the stored input image signal a by a coefficient (1−α), and a signal b multiplied by α and a signal a multiplied by (1−α). An image processing apparatus, comprising: a first adder for adding, and outputting the image signal added by the first adder. 2. The image processing apparatus according to claim 1, wherein the input image signal stored in the image storage unit is a predetermined number of upper bits of each pixel forming one frame. An image comparison unit that compares the input image signal b with an input image signal a stored in an image storage unit, detects a changed pixel, and calculates a coefficient β corresponding to a change state; For each of the changed pixels detected in step (a), a filter processing unit for calculating an average value of the changed pixel and its predetermined peripheral pixels, and a filter (1-β) times the average calculated by the filter processing unit. A third multiplier, a fourth multiplier that multiplies the image signal c output from the first adder by a coefficient β, and adds the average value multiplied by (1−β) and the image signal c multiplied by β. The image processing apparatus according to claim 1, further comprising a second adder, and outputting the pixel signal added by the second adder. 4. The image processing apparatus according to claim 3, wherein peripheral pixels of the changed pixel are eight pixels adjacent to the changed pixel. The image apparatus according to claim 3, wherein the coefficient β for the changed pixel is 0 <β <1, and the coefficient β for the unchanged pixel is β = 1.

Images