CN1985290A - Gray scale display device - Google Patents

Abstract

A grayscale display device, comprising a gradient detection circuit (3) for detecting the gradient of the grayscale value of a pixel within a picture in an input image; detecting the change degree of the grayscale value of a pixel relative to time in the input image The time change detection circuit (4); according to the output of the gradient detection circuit (3) and the output of the time change detection circuit (4), the unit that detects the size of the image movement of input and the image movement direction; and according to the detected image movement size A gradation correction circuit (12) for correcting and displaying the input image signal by weighting with the moving direction of the image and the brightness of the subfield.

Description

grayscale display device

technical field

The present invention relates to a grayscale display device using subfields, in particular to a grayscale display device capable of reducing grayscale display distortion during dynamic image display, that is, a virtual contour of a dynamic image.

Background technique

In an image display device that uses sub-fields to perform grayscale display, such as a display device that generally uses a plasma display panel (PDP), noise-like noises called "moving image virtual contours" or the like may sometimes be seen in moving image parts. Image quality deterioration.

Although it is known that the dynamic image virtual contour can be improved by increasing the number of sub-fields, according to such devices as PDP, if the sub-fields are increased, it is difficult to ensure the luminous time and cannot obtain the necessary brightness. , so try to set a smaller number of sub-fields, only in the part where the virtual contour of the dynamic image is generated, and control the combination of the sub-fields relative to the gray scale you want to display, so as to take into account the quality of the dynamic image and ensure the brightness (refer to for example 2000-276100 Bulletin).

In this existing image display device, in the part where the image has movement, the number of gray scales used for display is limited, and the combination of gray scale values that are not easy to produce the virtual contour of the dynamic image is limited to display the image. In order to make up for the number of gray scales The decline of the gradation is added to the virtual gray scale through dithering to ensure a certain gray scale.

However, in conventional image display devices, motion detection does not adopt a structure that particularly considers the gradation display method using subfields. In terms of high-precision detection of image parts that tend to generate virtual contours in moving images or parts that attract attention, There is still room for improvement.

The present invention was conceived to solve such problems, and provides a gray scale display device capable of accurately detecting the occurrence of substantially virtual contours of moving images and having a simple circuit configuration.

Contents of the invention

In order to solve the above-mentioned problems, the grayscale display device of the present invention has the following characteristics, a grayscale display device that constitutes one field period by a plurality of subfields having a predetermined brightness weighting, and performs grayscale display through the plurality of subfields, Including a gradient detection unit that detects the gradient of the gray value of the pixel in the image in the input image; a time change detection unit that detects the change degree of the gray value of the pixel relative to time in the input image; according to the gradient detection unit output and the output of the time change detection unit to detect the input image movement size and image movement direction; Signal correction unit shown.

Description of drawings

FIG. 1 is a block diagram showing the structure of a grayscale display device according to an embodiment of the present invention.

Fig. 2 is a schematic diagram of range combinations and control methods of the characteristic quantities of the device.

Fig. 3 is a block diagram showing an example of a smoothness detection circuit of the device.

Fig. 4 is a block diagram showing an example of a gradient detection circuit of the device.

FIG. 5 is a schematic diagram showing an example of filter coefficients of the gradient detection circuit.

Fig. 6 is a block diagram of an example of time change detection of the device.

Fig. 7 is a characteristic diagram showing a determination circuit of the device.

Fig. 8 is a structural diagram of the overall judgment result of the device.

FIG. 9 is an explanatory diagram of a method of calculating an image shift amount from a gradient and a time change in the device.

Fig. 10 is a characteristic diagram of a gradation distortion evaluation circuit of the device.

Fig. 11 is a characteristic diagram of the gradation correction circuit of the device.

Fig. 12 is a combined diagram of luminance weighting and light emission for subfields of the device.

Fig. 13 is a diagram of the encoding method of the encoding circuit of the device.

FIG. 14 is a graph showing the relative relationship between the gradient direction of an image portion and the moving direction of an image in a grayscale display device according to another embodiment of the present invention.

Fig. 15 is an evaluation diagram of the amount of gradation distortion of the device.

Fig. 16 is a block diagram showing the configuration of a gray scale display device according to another embodiment of the present invention.

Fig. 17 is a schematic diagram of the gradient direction component VG of the motion vector V of the device.

Fig. 18 is a configuration diagram of a gradation distortion amount predicting circuit of the device.

Fig. 19 is a block diagram showing the configuration of a grayscale display device according to another embodiment of the present invention.

Fig. 20 is a block diagram showing the configuration of the gradation correction circuit of the device.

Fig. 21 is an explanatory diagram for explaining a general error diffusion coefficient.

Fig. 22 is an explanatory diagram for explaining the error diffusion coefficient control method of the device of the present invention.

Fig. 23 is a schematic diagram of the transition of the error diffusion coefficient EA of the device.

Fig. 24 is an explanatory diagram for explaining the calculation method of the error diffusion coefficient EA of the device.

Fig. 25 is an explanatory diagram for explaining the concept of interpolation of the error diffusion coefficient EA of the device.

Fig. 26 is a transition diagram of the error diffusion coefficient EB of the device.

Fig. 27 is an explanatory diagram for explaining the interpolation concept of the error diffusion coefficient EB of the device.

Fig. 28 is an explanatory diagram for explaining the interpolation concept of the error diffusion coefficient EC of the device.

Fig. 29 is an explanatory diagram for explaining the interpolation concept of the error diffusion coefficient ED of the device.

Detailed ways

Hereinafter, a grayscale display device according to an embodiment of the present invention will be described with reference to the drawings.

Embodiment 1

FIG. 1 is a block diagram showing the structure of a grayscale display device according to an embodiment of the present invention. In FIG. 1, the image signal supplied by the input terminal 1 is respectively supplied to a smoothness detection circuit 2 as a smoothness detection unit, a gradient detection circuit 3 as a gradient detection unit, and a time change detection circuit 4 as a time change detection unit, The time change detection circuit 4 detects the time direction of pixels in the input image, that is, the degree of change of the gradation value with respect to time. The smoothness detection circuit 2 detects the smoothness of the gray value of the pixel in the input image, and the gradient detection circuit 3 detects the gradient of the gray value of the pixel in the input image in the screen.

The judging circuits 5, 6, and 7 that take the respective outputs of the smoothness detection circuit 2, the gradient detection circuit 3, and the time change detection circuit 4 as inputs are judging circuits that compare input data with predetermined thresholds, and input multiple The output of the judgment circuits 5 to 7 is output from the comprehensive judgment circuit 8 as the comprehensive judgment result k.

The output S of the smoothness detection circuit 2 is input to the judgment circuit 5, and a threshold value TH1 can be set at the same time, and a judgment result k1 can be output. The output G of the gradient detection circuit 3 is input to the determination circuit 6, and two thresholds TH2, TH3 can be set simultaneously, and a determination result k2 is output. The output B of the time change detection circuit 4 is input to the determination circuit 7, and two threshold values TH4 and TH5 can be set simultaneously, and a determination result k3 can be output. These determination results k1 , k2 , and k3 are input to the comprehensive determination circuit 8 .

In addition, the output G of the gradient detection circuit 3 and the output B of the temporal change detection circuit 4 are supplied to the movement amount detection circuit 9, and the movement magnitude and direction of the image movement of the input image are detected based on the input data. Then, the output G of the gradient detection circuit 3 and the output m1 of the movement amount detection circuit 9 are supplied to the gradation distortion evaluation circuit 10, and the output m2 of the gradation distortion evaluation circuit 10 and the output k of the comprehensive determination circuit 8 are supplied to the correction amount control circuit. The circuit 11 controls the operation of the gradation correction circuit 12 serving as signal correction means based on the output of the correction amount control circuit 11 .

The image signal input from the input terminal 1 and the output m3 of the correction amount control circuit 11 are supplied to the grayscale correction circuit 12, and the output thereof is connected to a subfield grayscale display device 13. That is, the gradation correction circuit 12 corrects and displays the input image signal based on information on the magnitude and direction of image movement detected by the movement amount detection circuit 9 and information on the brightness weighting of subfields of the input image signal.

Next, the operation of each part of the grayscale display device having such a structure will be described in detail.

First, in FIG. 1 , the smoothness detection circuit 2 , the gradient detection circuit 3 , and the time change detection circuit 4 are used to detect a pixel of interest or an image feature of a designated area in an input image signal. FIG. 2 is a schematic diagram of an example of the combination and control method of the characteristic range.

That is, as shown in FIG. 2, use the smoothness detection circuit 2, the gradient detection circuit 3, the time change detection circuit 4, and the determination circuits 5 to 7 connected to these circuits to determine the characteristics and various ranges of the input image, and then use the integrated Determination circuit 8 divides the area of this attention into six categories such as [no time change], [time change is too large], [flat portion], [edge portion], [certain slope portion], [complex figure], etc., by and Which of the six categories is equivalent determines the comprehensive judgment result k. Also, in FIG. 2 , the inequality sign indicates the magnitude relationship between the feature value of each pixel and the threshold value, and the [X] symbol indicates that the magnitude relationship is an arbitrary value.

As shown in Figure 2, in the determination circuit 5, if the degree of smoothness of the area to be noticed is S, the detection becomes the range of S≥TH1 (TH1 is the threshold of the determination circuit 5), and in the determination circuit 6, if the area If the gradient of the grayscale value is G, the range of TH2≤G≤TH3 (TH2 and TH3 are the thresholds of the determination circuit 6) will be detected. In addition, in the determination circuit 7, if the change in the time direction of the grayscale value of this area When the degree is B, the detection falls within the range of TH4≦B≦TH5 (TH4 and TH5 are the threshold values of the determination circuit 7). Then, the detected range pixels are determined as an area that is likely to generate a dynamic image virtual contour or is easy to be detected, and performs gray scale correction on this portion, and performs image display.

That is, because the gradient (inclination) of the pixel gray value of the dynamic image virtual contour forming the image in the picture and the variation degree of the pixel gray value relative to time are respectively within the appropriate upper limit and lower limit range, and in the image graphics Relatively smooth parts are noticeable, so such parts are selectively detected.

Here, an example of the above-mentioned smoothness detection circuit 2, gradient detection circuit 3, and time change detection circuit 4 will be described. First, as shown in FIG. 3, the smoothness detection circuit 2 is composed of the following circuit, including a delay circuit 20 that delays each pixel signal according to the image signal input from the input terminal 1; A pixel average calculation circuit 21 for calculating the average value of the grayscale value for each pixel signal; by obtaining the difference between the output value of the pixel average calculation circuit 21 and the output value of the delay circuit 20, and converting the grayscale of each pixel signal A difference circuit 22 for calculating the difference between the value and the average value, an absolute value operation circuit 23 for obtaining the absolute value of the difference value obtained by the difference circuit 22, and an output value obtained by the absolute value operation circuit 23 The addition circuit 24 that adds the absolute values to obtain the smoothness output of each pixel gray value of the input image signal.

Next, as shown in FIG. 4 , the gradient detection circuit 3 uses a horizontal filter 30 that detects changes in pixel grayscale values in the horizontal direction and a vertical filter 31 that detects changes in pixel grayscale values in the vertical direction to obtain the above-mentioned horizontal filter An absolute value operation circuit 32 for taking the absolute value of the respective output values of the filter 30 and the vertical filter 31, and an addition circuit 33 for adding the output values of the absolute value operation circuit 32. The above-mentioned horizontal filter 30 and vertical filter 31 play the role of multiplying the surrounding pixels of the attention pixel by a predetermined coefficient and adding them together. As an example of this coefficient, as long as they respectively constitute That's it. That is, by using the horizontal filter 30 and the vertical filter 31, in the image signal input from the input terminal 1, the variation of the pixel gradation value with respect to the horizontal direction and the vertical direction is detected, and by comparing the absolute value of the detected value plus, to detect the gradient of the input image signal as the degree of inclination of the gray value of the pixel.

Next, as shown in FIG. 6 , the time change detection circuit 4 uses a field delay circuit 40 that delays one field of the input image signal to obtain the grayscale value of the pixel of the current image signal and the value of the pixel grayscale value before passing through the above-mentioned field delay circuit 40. The differential circuit 41 for the difference between the pixel grayscale values of the image signal of one field, and the absolute value calculation circuit 42 for obtaining the absolute value of the output of the differential circuit 41 are constituted. By obtaining the pixel grayscale value of the current image signal and the previous The difference between the pixel gray value of the image signal of one field is used to detect the time change of the pixel gray value of attention.

Also, in Fig. 2, the degree of correction of the input image grayscale is simply recorded as [correction=weak] and [correction=strong], but the degree of correction can be set to three levels or above In the multi-level method, smooth correction is performed by continuously switching the correction amount. 7A, 7B, and 7C respectively show the characteristics of the determination circuit 5, the determination circuit 6, and the determination circuit 7. In order to correspond to the above-mentioned continuous correction of the gray scale, they are the characteristics shown in FIGS. 7A, 7B, and 7C.

That is, if the characteristics of the determination circuit 5 as shown in FIG. 7A are described, then for the detected degree of smoothness S, a threshold value TH1 is set so that the degree of smoothness S is in a part close to the threshold value TH1, and the output of the determination circuit 5 is [0 ] and [1], when the degree of smoothness S is in the part smaller than the threshold value TH1, the output of the decision circuit 5 is a value closer to [0], and when the degree of smoothness S is in the part greater than the threshold value TH1, the output of the decision circuit 5 The output is a value closer to [1].

Regarding the decision circuit 6, as shown in FIG. 7B, the thresholds TH2 and TH3 are set so that when the input, that is, the gradient G is between the two thresholds, the output of the decision circuit 6 is a value closer to [1], and the value of the gradient G is between the two thresholds. In other cases, the output of the determination circuit 6 is a value closer to [0].

In addition, regarding the determination circuit 7 as shown in FIG. 7C, like the determination circuit 6, two thresholds TH4 and TH5 are set. The output of 7 is a value closer to [1], and the value of the degree of change B with respect to time is other than this, the output of the determination circuit 6 is a value closer to [0]. In addition, the actual outputs of the determination circuit 5, the determination circuit 6, and the determination circuit 7 may of course change in a stepwise manner.

In addition, the comprehensive determination circuit 8 outputting the comprehensive determination result k is composed of, for example, multipliers 81 and 82 shown in FIG. ~7 The features of the obtained image can smoothly obtain the comprehensive judgment result k.

On the other hand, the detection of the size of the image movement, that is, the amount of movement and the direction of image movement is based on the gradient G detected by the gradient detection circuit 3 and the degree of change B in the time direction detected by the time change detection circuit 4. 9 to proceed. If it is assumed that the shape of the displayed object does not change but the gray value of the image changes, the calculation method can be operated as follows in principle.

That is, as shown in FIG. 9, since it can be assumed that it is proportional to the change degree B of the pixel grayscale value in the time direction and inversely proportional to the change in the grayscale value in the screen, that is, the gradient G, the amount of image movement m1 can be Use m1=B/G to find out. However, when the change in the gradient G is large, the above assumption does not hold, and the amount of movement cannot be accurately obtained. In addition, since the denominator of the above-mentioned calculation formula is a small value in a portion where there is basically no gradient G, the amount of movement cannot be obtained with high accuracy even in this case. In addition, when the change in the time direction is very small, there is basically no virtual contour of the dynamic image, or conversely, when the brightness change per unit time is very large, it is difficult to perceive the virtual contour of the dynamic image. Therefore, by limiting the combination of image features shown in FIG. 2 , it is possible to detect image motion with high accuracy in a portion where dynamic image virtual contours are likely to occur. That is, by controlling the operation of correcting the virtual contour of the moving image based on the output k of the comprehensive determination circuit 8, image motion can be detected with high accuracy in the portion where the false contour of the moving image is likely to occur, and the image signal can be corrected.

In addition, the amount of movement obtained by the calculation of the above-mentioned movement amount detection circuit 9 can be obtained accurately enough as long as the characteristics of the image satisfy the above-mentioned conditions, but the detected amount of movement is the number of pixels per unit time, which is originally It is a physical quantity different from the virtual contour of the dynamic image that appears as grayscale distortion, and is not necessarily completely proportional to the value of the virtual contour of the dynamic image that is actually observed visually.

Therefore, in the present invention, a configuration is adopted in which the gradation distortion amount m2 is estimated using the gradation distortion amount evaluation circuit 10 having two-dimensional input-output characteristics as shown in FIG. 10, and the gradation distortion amount m2 is input to Correction amount control circuit 11. That is, the configuration is such that the pixel moving speed obtained by the moving amount detecting circuit 9 , that is, the image moving amount is converted into distortion of the gray scale value, and input to the correction amount control circuit 11 .

The characteristic in FIG. 10 is that when the movement amount is changed with respect to the magnitude of a constant gradient, the virtual contour of the moving image becomes the maximum value at an intermediate value of the movement amount. That is, the characteristics of the gradation distortion amount evaluation circuit 10 can be expressed as dots at a portion where the gradient is small and the movement amount is large (A in FIG. 10 ), or at a portion where the movement amount is small and the gradient is large (such as B in 10 ). A function that produces severe virtual contours in dynamic images.

Then, although not shown, the correction amount control circuit 11 can be constituted by, for example, a multiplier, multiplies the estimated gradation distortion amount m2 by the comprehensive determination coefficient k, and outputs the computed gradation correction signal m3.

In addition, in the gradation correction circuit 12 to which the gradation correction signal m3 is input, in order to suppress the virtual contour of the dynamic image which is generated by the image display using the subfield, according to the structure of the subfield, the image movement and the gradation value, corresponding Perform gray scale correction. The gradation correction circuit 12 is composed of a combination of an encoding circuit and a feedback circuit, as shown in FIG. 11 .

In FIG. 11 , the image signal input from the input terminal 1 is supplied to the encoding circuit 122 via the adder 121 , and is then output from the output terminal 125 after predetermined encoding is performed in the encoding circuit 122 . At this time, the subtractor 123 obtains the difference from the signal before encoding, and then adds the input signal to the adder 121 via the feedback circuit 124 . In addition, since the feedback circuit 124 generally includes a plurality of system delay elements and coefficient circuits, by performing gradation limitation in the encoding circuit 122 , so-called error diffusion processing is performed as the gradation correction circuit 12 .

FIG. 12 shows an example of an encoding method combining luminance weighting and light emission of subfields used by the gradation display device 13. FIG. 12 shows a case where 10 subfields (SF1 to SF10) are used. As shown in Figure 12, the brightness weighted ratios of each subfield are respectively set to [1], [2], [4], [8], [16], [24], [32], [40], [ 56], [72]. In addition, FIG. 12 shows a subfield allocation encoding method corresponding to a certain input image grayscale value, and the part of "1" in the figure indicates "light emitting".

FIG. 13 is a schematic diagram of an encoding method in the encoding circuit 122 of FIG. 11, showing an example of luminance weighting of subfields and the encoding method. That is, if the correction amount is small, grayscale display is performed using a large number of grayscales, and grayscale control is performed in this way; on the other hand, if the correction amount is large, grayscale display is performed using a small number of grayscales. Performs gray scale control while utilizing error diffusion to ensure effective gray scale and image display. In FIG. 13 , the gradation correction amount is set to eight steps from [0] to [7], and dots are placed on the gradation values to be used. That is, when the grayscale correction amount is [0], all grayscales can be used, and when the grayscale correction amount is [7], the number of usable grayscales is the minimum. In the part where the virtual contour of the dynamic image may be seriously generated, the correction amount is increased to maintain the correlation between the gray value and the luminescence distribution of the subfield, thereby suppressing the generation of the virtual contour of the dynamic image. In addition, as the amount of imaginary virtual contours in dynamic images becomes less and less, by reducing the amount of correction, the grayscale correction of the image can be continuously controlled to achieve smooth suppression of virtual contours in dynamic images and when it is difficult to generate virtual contours in dynamic images. Good grayscale correction for some parts.

In this way, according to this embodiment, there is a unit that detects the gradient of the image frame and the degree of change of the gray value relative to time, and detects the input image movement size and image movement direction according to the detected information; A signal correcting unit that corrects and displays an input image signal by weighting the magnitude and direction of image movement obtained with the luminance of the subfield can perform good gradation display with a simple structure.

However, [Multidimensional Signal Processing of TV Images] (Tekhiko Fukiba, P202-P207, Nov. date issue) and other methods. However, the gradient method described in [Multidimensional Signal Processing of TV Images] and the like is effective when the motion is small, and is not necessarily widely used in practice.

The present invention is a method discovered by observing the generation of dynamic image virtual contours in an image display device using subfields, and elucidating the relationship between the amount of generation of dynamic image virtual contours with respect to the structure of subfields, image characteristics, and image movement amounts. . That is, it was found that as long as the conditions of the portion where the gradient of the grayscale value is within the specified upper limit and the lower limit range, and the portion where the change of the grayscale value with respect to time is within the specified upper limit and the lower limit range are satisfied, it is possible to easily determine the virtual image of the moving image. The generation position or degree of the contour can basically detect the movement of the image correctly according to the gradient and time changes, and make full use of the above-mentioned found relationship, it is possible to provide a simple structure that takes into account the dynamic image characteristics and static characteristics. Methods.

In addition, it is of course possible to make various changes regarding the luminance weighting of subfields, the encoding method of subfields, the method of predicting the amount of gradation distortion from the amount of image movement, and the method of gradation correction used in the above description. out of shape.

Embodiment 2

Next, other embodiments of the present invention will be described. In this embodiment, the grayscale value is controlled and displayed based on the correction amount obtained by comprehensively determining the smoothness of the grayscale value of the input image signal, or the gradient in the screen, and the degree of change with respect to time. At the same time, pay attention to the relationship between the gradient direction of the gray value and the direction of change with respect to time, and more accurately determine the procedure of the virtual contour of the generated dynamic image and perform image correction. In this embodiment, compared with the embodiment in FIG. 1 , only the internal structure and operation of the gradation distortion amount prediction circuit 10 are different, and other structures and operations are basically the same, so only the different parts will be described.

FIG. 14 shows the relative relationship between the gradient direction of the image portion to be displayed and the image moving direction in the grayscale display device of the present embodiment. The table part of FIG. 14 is the same as that shown in FIG. 12 described in the above-mentioned embodiment. The solid line arrows and dotted line arrows shown in FIG. The difference in the amount of virtual contours in the dynamic image produced when the image is generated.

For example, in FIG. 14 , consider a case where a ramp waveform moves in the vicinity centered on the value of the gradation value [200]. As shown in Figure 14A, if the part of the image moves in the direction opposite to the direction in which the gray value increases in the screen, the probability of observing the subfield of [lighting] is less than the original one, and relatively serious Dynamic image virtual outline. On the contrary, as shown in FIG. 14B , when the image portion moves in the same direction as the direction in which the gradation value increases in the screen, although slightly more light emission than the amount of light emission that should be observed is observed, the The amount is smaller than in the case of moving in the opposite direction, and the resulting moving image has a smaller degree of false contours.

Therefore, when evaluating the amount of virtual contour generation in a moving image based on image movement, the direction of image movement and the gradient direction of grayscale values within the screen are relatively evaluated, and by changing the amount of image correction, more accurate image correction can be performed.

FIG. 15 shows the state of this control, and shows the evaluation of the amount of gradation distortion with respect to the magnitude and direction of image movement, and the gradient of gradation values. Fig. 15 is a two-variable function with image movement (horizontal axis) and gradient (vertical axis) as two parameters, and the function value (perpendicular to the direction of the paper) is the grayscale distortion, ie, the evaluation value of the virtual contour of the dynamic image.

As can be seen from FIG. 15 , even if the absolute value of the same image gradient and image shift is the same, the amount of image correction can be changed by combining the direction of image shift and the direction of the gradient of the grayscale value. In addition, in the example of FIG. 15, the absolute value of the magnitude of the image movement increases from the state of [0], and the image correction amount also increases accordingly, and reaches the maximum value at a certain point. The maximum value is set in this way, depending on the combination of image moving direction and gradient direction. For example, when the image moving direction is [+] and the gray value gradient is [+], the correction amount of the image is maximized. , or when the image moving direction is [-], and the gradient of the gray value is [-], the correction amount of the image is maximized.

As described above, according to the present embodiment, for the virtual contour of a moving image, the correction amount of the image is changed according to the combination of the moving direction of the image and the direction of the gradient, and good gradation display can be performed with a simple structure.

Embodiment 3

Next, other embodiments of the present invention will be described with reference to FIGS. 16 to 18 . This embodiment is a gradation display device that detects an image movement direction by dividing it into a horizontal direction component and a vertical direction component, and corrects a signal based on a value obtained by converting the gradient magnitude and the image movement magnitude into gradient directions. In FIG. 16 , as compared with the embodiment shown in FIG. 1 , the same reference numerals are assigned to the same basic operations as those in the embodiment shown in FIG. 1 , and description thereof will be omitted.

In FIG. 16 , the gradient detection circuit 31 outputs the horizontal component Gx and the vertical component Gy of the gradient in addition to the gradient absolute value |G| of the gradation value. The horizontal movement amount detection circuit 91 and the vertical movement amount detection circuit 92 calculate the horizontal direction of the image based on the horizontal direction component Gx of the gradient, the vertical direction component Gy of the gradient, and the change amount of the grayscale value with respect to time, that is, the change degree B. The amount of movement Vx and the amount of movement Vy in the vertical direction. Furthermore, the gradation distortion amount predicting circuit 100 calculates a value based on the absolute value of the gradient |G|, the horizontal component Gx of the gradient, the vertical component Gy of the gradient, the horizontal movement Vx of the image, and the vertical movement Vy of the image. Calculate the equivalent gray level distortion me.

FIG. 17 shows the relationship between the motion vector V represented by image motion components (Vx, Vy) and the gradient component VG of the motion vector V. FIG. This VG is calculated by the gradation distortion amount prediction circuit 100 configured as shown in FIG. 16 .

Fig. 18 is a specific structural diagram of the gradation distortion amount prediction circuit 100. In Fig. 18, the arc tangent function transformation unit 101, the arc tangent function transformation unit 102 and the subtractor 103 are used to calculate the distance between the moving vector V and the gradient direction. The corner is further transformed by the cosine function transformation unit 104, and the transformed value is multiplied by the absolute value of the image movement amount obtained by the absolute value circuit 106. In this way, the movement size component converted into the gradient of the image can be obtained VG. The table 107 is capable of predicting the generation amount of virtual contours in moving images similar to that of the gradation distortion evaluation circuit 10 in FIG. 1 .

According to the above configuration, image motion and image gradient direction can be evaluated collectively, and it is possible to accurately estimate the generation prediction of virtual contours in moving images, and perform accurate image correction and good image display.

Embodiment 4

FIG. 19 is a block diagram showing another embodiment of the present invention. In FIG. 19, the same parts as those shown in FIG. 1 are given the same reference numerals. In FIG. 19, the output G of the gradient detection circuit 3 and the time direction change detection circuit are supplied to the horizontal movement amount detection circuit 14, the vertical movement amount detection circuit 15, the 45° movement amount detection circuit 16, and the 135° movement amount detection circuit 17, respectively. Output B of 4. In addition, the output of the horizontal movement amount detection circuit 14 and the vertical movement amount detection circuit 15 is used as the input of the movement amount calculation circuit 18, and then the movement amount calculated by the movement amount calculation circuit 13 is input to the comprehensive judgment circuit 8, and the comprehensive judgment result k is output. The overall determination result k is supplied to the gradation correction circuit 19 which is a signal correction unit.

An image signal input from the input terminal 1 is input to the gradation correction circuit 19, and the gradation correction circuit 19 performs gradation correction control and error diffusion control for correcting the gradation value of the input image. The method of this gradation correction and error diffusion is passed through the comprehensive determination result k of the above-mentioned comprehensive determination circuit 8, and the horizontal movement amount detection circuit 14, the vertical movement amount detection circuit 15, the 45° movement amount detection circuit 16 and the 135° movement amount detection circuit 17 output to control. The gradation-corrected image signal obtained by the gradation correction circuit 19 is supplied to the subfield gradation display device 13 and displayed as an image.

Here it is constituted like this, it detects the size of image movement in four directions, and is used for the control of the grayscale correction circuit 19 of the subsequent stage, but the calculation of the size of the image movement itself can be based on the horizontal movement amount and the vertical movement amount. Therefore, after supplying the movement amount calculation circuit 18 to obtain the movement size, it is input to the comprehensive judgment circuit 8 to determine the value of the comprehensive judgment result k corresponding to the necessary gradation limit amount.

Next, the gradation correction circuit 19 will be described in detail. In the gradation correction circuit 19, the gradation correction of the input image is performed using a value corresponding to the obtained multi-directional image moving direction, multi-directional image moving magnitude, and image gradation limit amount, that is, the comprehensive judgment result k. , but the relationship between the size of the multi-directional image movement and the gradation limit is the same as the method described in FIG. 12 and FIG. 13 .

FIG. 20 shows a specific configuration example of the gradation correction circuit 19 . As shown in FIG. 20, the gradation correction circuit 19 has an adder 191, an encoding circuit 192, a movement amount input terminal 193, an input terminal 194, a subtractor 195, delay circuits 196-199, coefficient circuits 200-203, and coefficient control circuits. circuit 204 . Furthermore, the previously detected horizontal movement amount, vertical movement amount, 45° movement amount, and 135° movement amount are respectively input to the coefficient control circuit 204, and the respective coefficient values EA, EB, EC, and ED are obtained by the coefficient circuits 200 to 203, respectively. , the signals of the delay circuits 196-199 are calculated and processed according to the coefficient values, and then supplied to the adder 191 to form an error diffusion loop.

Also, in the structure shown in FIG. 20, the switching of the grayscale control of the grayscale value of the input image signal is carried out according to the signal input from the movement amount input terminal 193. In addition, the coding shown in FIG. The encoding circuit 192 of the correction circuit 19 is carried out.

In this way, the input image signal correspondingly limits the number of grayscales according to the size of the image movement, and supplies the display device to properly suppress the generation of virtual contours in dynamic images. At the same time, since an error diffusion ring is formed, an equivalent gray value can be ensured. Also, if the limitation of the number of gradations in the moving image portion is increased in order to increase the effect of suppressing the virtual contour of the moving image, it is felt that the image quality is degraded due to the large amount of noise caused by the error diffusion process. Therefore, the present invention controls the error diffusion coefficient according to the moving direction of the image, so as to suppress the degradation of image quality when the grayscale limit is large.

Fig. 21 is an explanatory diagram of a general error diffusion coefficient. FIG. 21 shows how the difference between the input signal and the display signal at that time is distributed to the surrounding four pixels A, B, C, and D when the pixel P is displayed with gradation limitation. FIG. 22 shows actual numerical examples of the distribution coefficients EA, EB, EC, and ED. According to Fig. 22, it can be seen that when the image movement is very small and no virtual contour of the dynamic image is actually generated, the image is regarded as a static image, and the coefficient values EA, EB, EC, and ED are respectively [7], [1], [ 5], [3] values. Also, since the value of the error diffusion coefficient is originally the coefficient of the error distribution, the sum should be [1], but for convenience, it is expressed as a value of 16 times.

Also, when the image is not a static image but moves in a specified direction, the values of the coefficient values EA, EB, EC, and ED are updated according to FIG. 22 . Parts other than [Still Image] in FIG. 22 show coefficients set for each image moving direction. In the figure, the coefficient values when there is a certain degree of image movement are shown, but in fact, according to the size of the image movement, the values are set as continuous or segmented.

FIG. 23 is an explanatory diagram of this case, and is a conceptual schematic diagram of a method of setting the coefficient EA. That is, control is performed in such a way that when it is a static image, the coefficient EA is set to [7], but the image movement becomes larger, for example, when there is image movement in the horizontal direction of the pixels of the screen, the coefficient value EA is set to the maximum according to the size of the image movement. It is set as [10]. In addition, when the moving direction of the image is the vertical direction of the pixels of the screen, the coefficient value EA gradually decreases from [7] to [0] according to the size of the moving image. Also, when the image moves in the oblique direction of the screen, it is similarly controlled so that it gradually changes from [7] to [3].

FIG. 24 is an explanatory diagram of this case, showing the relationship between the angle θ shown in FIG. 22 and the image movement. In Fig. 24, when there is image movement in the direction forming an angle θ with the horizontal, assuming that the size of the image movement is m, the image movement is represented by a vector.

The coefficient value EA corresponding to such image movement can be obtained from the value obtained by interpolation in FIG. 23 using FIG. 5 which is a graph. FIG. 25 shows values obtained by interpolating points other than the values shown in FIG. 23 with the values explicitly shown around them, and the angle θ=0 indicates the horizontal direction of the screen. In addition, the upper part of FIG. 25 (the direction perpendicular to the bottom surface) shows the coefficient value of each point. In FIG. 25, the value of point P is equivalent to that of point P in FIG. 24, and its coefficient value is represented by EA.

Since the coefficient value is set to change continuously, the coefficient value of the error diffusion can be continuously changed according to the value of the static image, the direction of the image movement, and the magnitude of the image movement, and the gradation correction can be smoothly performed according to the magnitude and direction of the image movement , performing good suppression of virtual contours in dynamic images and good error diffusion operations.

In addition, other coefficients, for example, the coefficient value EB can be represented by transitions as shown in FIG. 26 , and can be interpolated and shown in FIG. 27 . The same is true for the transition diagrams of the coefficient values EC and ED, which can be shown in FIGS. 28 and 29, respectively. In addition, although not shown, the concept of interpolation of coefficient values can be expressed using the same graph as that in FIG. 25 or FIG. 27 for the coefficient values EC and ED.

According to the present embodiment as described above, in the gradation display device using subfields, signal processing including gradation correction control and error diffusion control is performed using the magnitude and direction of image movement, and it is possible to realize dynamic image virtual contours. suppression and good grayscale display.

Also, in the above description, the error diffusion coefficient in the direction parallel to the image moving direction is relatively increased. This is because when the line of sight follows the movement of the image and follows the object on the screen, it is considered that the amount of light emitted by a plurality of pixels is "visually fused" on the observer's retina. That is, a plurality of pixels in the direction parallel to the image movement can be considered to equivalently represent an action similar to one pixel, and there should be as much error as possible between such pixels, and it is difficult to occur [visual The diffusion error of the pixels of the fusion on], that is, the pixels in the direction orthogonal to the image movement, can suppress the sense of noise that increases with the error diffusion.

In addition, in the present embodiment, an example in which interpolation of coefficient values is distributed linearly is described, but of course, curve interpolation using a higher-order function or other continuous functions may be used. In addition, an example was given in which the gradation value is controlled in several steps according to the size of the image movement, but the number of steps is not limited to the above example. Furthermore, as a special example, it is also possible to control only the error diffusion coefficient without performing the control of the number of gradations. In addition, the error diffusion coefficient described in this embodiment is not limited to the illustrated error diffusion coefficient. Of course, the same effect can be obtained by utilizing the characteristics of the visual blending effect according to the moving direction of the image.

As explained above, according to the present invention, since the gradient detection unit that detects the gradient of the gray value of the pixel in the screen in the input image; A time change detection unit; a unit that detects the input image movement size and image movement direction according to the output of the above-mentioned gradient detection unit and the output of the above-mentioned time change detection unit; and the image movement size and image movement direction according to the above detection, and the above-mentioned The brightness weighting of the sub-field is a signal correction unit that corrects and displays the input image signal, so it can detect the moving direction of the image according to the image gradient and predict the generation of the virtual contour of the dynamic image, so it can perform grayscale correction more accurately and can suppress Dynamic image virtual contour, and image display that can ensure good grayscale characteristics.

According to the present invention, it is possible to detect the image movement and the gradient of the portion where the virtual contour of the dynamic image is likely to occur with a simple structure, thereby suppressing the virtual contour of the dynamic image and realizing a good image display, and improving the grayscale display device using subfields. display quality.

Industrial Applicability

According to the present invention described above, it is possible to detect the image shift and gradient in the portion where the virtual contour of the moving image is likely to occur with a simple structure, and correct the signal and display it, thereby suppressing the false contour of the moving image and realizing a good image display. The display quality of the device is displayed using the gray scale of the subfield.

Claims (5)

1. a grey scale display unit is characterized in that,

Be to constitute a field duration and carry out gray scale gray-scale displayed display device by its a plurality of sons field by luminance weighted a plurality of sons field with regulation,

Be included in the gradient detecting unit that detects the gradient of gray-scale value in picture of pixel in the image of input; In the image of described input, detect the time change-detection unit with respect to the intensity of variation of time of the gray-scale value of pixel; The image that detects described input according to the output of the output of described gradient detecting unit and described time change-detection unit moves the unit of size and image moving direction; And move size and image moving direction, and the luminance weighted signal correction unit that the picture signal of importing is revised and shown of described son according to the image of described detection.

2. a grey scale display unit is characterized in that,

Be to constitute a field duration and carry out gray scale gray-scale displayed display device by its a plurality of sons field by luminance weighted a plurality of sons field with regulation,

Be included in the smoothness detecting unit of the smooth degree of the gray-scale value that detects pixel in the image of input; In the image of described input, detect the gradient detecting unit of the gradient of gray-scale value in picture of pixel; In the image of described input, detect the time change-detection unit with respect to the intensity of variation of time of the gray-scale value of pixel; The image that detects described input according to the output of the output of described gradient detecting unit and described time change-detection unit moves the unit of size and image moving direction; And the luminance weighted signal correction unit that the picture signal of input is revised and shown that moves size and image moving direction and described son field according to the image of described detection.

3. grey scale display unit as claimed in claim 1 or 2 is characterized in that,

Be to constitute like this, the image moving direction is divided into the horizontal direction component for it and the vertical direction component detects, and carries out signal correction according to gradient magnitude and image being moved the value that size conversion becomes gradient direction to obtain.

4. grey scale display unit as claimed in claim 1 or 2 is characterized in that,

The signal correction unit be revise input picture gray-scale value control and carry out the control of error diffusion.

5. grey scale display unit as claimed in claim 4 is characterized in that,

The signal correction unit is controlled, and to move the gray-scale value that size is revised input picture according to image, comes the signal Processing of departure diffusion simultaneously according to the image moving direction.

Images