KR19990076726A - Clarity Improvement Methods and Circuits - Google Patents

Abstract

In the sharpness improvement method, the presence of the predetermined colors R, M in the image signals y-in, u-in, v-in is determined to obtain the control signal CTRL (7), and the image signal y The sharpness of -in, u-in, v-in) is enhanced according to the control signal CTRL (1-5).

Description

Clarity Improvement Methods and Circuits

Some known circuits of this type improve the sharpness by emphasizing the current transient brightness when it is detected. This is logical because the sharpness detection of the image is mainly determined by the luminance signal.

The present invention relates to a method and a circuit for improving the sharpness, and to a multimedia device and a video display device including such a sharpness improving circuit.

1 is a basic block diagram of a first embodiment of the present invention.

2 is an explanatory diagram of a first possible sharpness detection region;

3 is an explanatory diagram of a good sharpness detection region;

4 shows a possible realization of the picking filter.

Fig. 5 shows a preferred embodiment of the video display device according to the present invention.

6 illustrates a multimedia device according to the present invention.

7 shows a preferred embodiment of a color-dependent sharpening improvement circuit according to the invention.

In particular, it is an object of the present invention to provide better sharpness improvements. To this end, a first aspect of the invention provides a method as defined in claim 1. A second aspect of the present invention provides a sharpness improvement circuit as defined in claim 9. The third and fourth aspects of the present invention provide a multimedia device and a video display device including such sharpness improvement circuits. Advantageous embodiments are defined in the dependent claims.

In the sharpness improving method according to the first aspect of the present invention, the presence of a predetermined color in the video signal is determined to obtain a control signal, and the sharpness of the video signal is enhanced in accordance with the control signal. In a preferred embodiment, the picture quality of the TV signal is improved by applying peaking on the luminance channel according to the color information to enhance the sharpness detection of the red and magenta portions of the image.

These and other aspects of the present invention will be apparent from and elucidated with reference to the embodiments described below.

Saturated colors (primarily red, magenta, blue) often appear blurry in current TV systems (PAL, NTSC, SECAM). This is partly caused by the fact that the color theory used for the concept of these TV systems is overlooked by gamma correction in the camera.

A first aspect of the invention aims to increase the perceived sharpness of saturated red, magenta and blue. The circuit increases the brightness sharpness in the saturated red and magenta portions of the screen. Noticeable results are obvious (blue is not enhanced; only noise is enhanced because there are very few items in the natural landscape). The circuit can be easily implemented in an IC.

Television signals are nonlinear because of the 1 / γ preliminary correction in the camera and the γ characteristic of the CRT. This basic characteristic is often overlooked. This is because the TV set is to be tested with test images such as color bars and gray scale. These images are generated without gamma preliminary correction.

In all standard TV systems, the camera sensor decomposes the image into three primary colors: R, G, and B. The 1 / γ preliminary correction is added immediately before matrixing the signal with three signals, Y, U, and V, which are suitable for effective transmission behind the sensor.

One of the consequences of this nonlinearity is the lack of a constant luminance principle. According to the present principle, it becomes possible to select a right angle axis for the display of the TV signal in such a manner that all the luminance information is in the Y signal. The other two axes only contain color information. Usually, R-Y (or V) and B-Y (or U) are selected for these color axes.

Because of the above-described gamma, such orthogonality does not exist. This means that there is not always crosstalk between luminance and chroma information. A simple calculation is that most of the luminance information is coded in a color signal for some colors (mainly blue, magenta, red). Other colors are less affected (yellow, cyan, green).

Since decoding the Y, U, V signals into color signals R, G, B is a reverse process of coding, there is no overall error in a system where the band is not limited. In an ideal case, it is possible to retrieve luminance information with a raw bandwidth. The present invention is based on the recognition that in real TVs, both U and V have a very reduced bandwidth (about 700 kHz compared to 4 MHz of Y) which prevents the reconstruction of the original luminance information in the TV receiver. As a result, the effective luminance bandwidth displayed on the screen will be primarily limited for saturated colors with low luminance content such as red, magenta, blue, and the like. This problem is inherent in current TV systems and is known as a lack of constant brightness.

This lack of constant luminance is important because the perceived sharpness feeling of an image is mainly determined by the luminance information. Thus, if some of the luminance information is transmitted over the chroma channels U and V, the luminance bandwidth will be less than ideally expected. This results in a saturated blue, magenta, and red portion of the blurred image.

The present invention is also based on the recognition that it is possible to improve an image by the following algorithm, i.e., an algorithm that detects when there is a red, blue or magenta color of the image and locally improves the sharpness of the color signal. Since the color difference signal is band limited to about 700 kHz, increasing the high frequency of the color difference signal is not used. Instead, luminance sharpness should be increased if sharpening enhancements are made in the YUV domain. While it is conceivable to enhance the sharpness of the color signal in the RGB domain, the enhancement of the sharpness of the luminance signal becomes a priority over cost.

In contrast to standard picking or sharpness algorithms, the extra gain is well applied at relatively low frequencies (ie 700 kHz to 2 MHz, whereas prior art picking typically operates in the frequency range between 2 and 4 MHz). This is desired because the lack of information starts at about 700 kHz. Above 2 MHz, the effect of this characteristic is not very important. Near the color subcarrier, the effect can be neglected to negate the amplification of the cross-lminance effect.

The sharpening enhancement of the present invention is preferably used in addition to the current sharpening enhancement algorithm. Current algorithms generally satisfy increasing brightness clarity, while the effect is insufficient for red, magenta, and blue. As far as these colors are concerned, current algorithms typically raise noise because they peak at too high frequencies. If this peaking is optimized for (saturated) red, magenta, and blue, the corresponding part of the image is improved, while the remaining parts (gray, green, yellow) are very blurred. Thus, the use of conventional sharpening enhancements for all colors is preferred, and the use of the sharpening enhancements proposed by the present invention for at least red and magenta is also preferred.

1 is a basic block diagram of a first embodiment of a sharpness improvement circuit SI according to the present invention. The luminance component y-in of the input image signal is input to a peaking circuit 1 which may include a second induction acquisition circuit. The picking signal output of the picking circuit 1 is multiplied by the multiplier 3 with the control signal CTRL, and then added by the adder 5 with the input luminance component y-in to output luminance component y-out. ). The control signal CTRL is derived from the input chrominance components u-in, v-in of the video signal by a color detector 7 which detects at least red, magenta and possibly blue as well. The color difference components u-in and v-in are supplied unchanged at the outputs u-out and v-out, respectively.

2 illustrates a first possible sharpness detection region in a color diagram determined by the color difference component B-Y on the horizontal axis and the color difference component R-Y on the vertical axis. Red (R), magenta (M), blue (B), cyan (C), green (G), and yellow (Y) are indicated at the corners of the color diagram. 2 shows how the U and V signals fill the color plane. If you want to detect red, blue, magenta, the detection area will be a shaded area. This area can be described as follows.

(B-Y) /2.03 + (R-Y) /1.14 ≥ 0

To avoid switching effects (one color is detected while the other similar color is not detected), the crossover between the detected color and the undetected color is preferably made gradually. It is also important that the colorless area is not detected because it sharpens the gray tone, which is not necessary with the color specific algorithm of the present invention.

The test turns out that it is better not to enhance the blue part because both are blue and there are few natural items with many items. The reason for not enhancing blue is because blue is a limited signal to the noise ratio of the camera. It is better to limit the strengthening to red and magenta. This leads to the following appropriate equation.

α * (B-Y) + β * (R-Y)-C ≥ 0

3 illustrates a good sharpness detection area that no longer contains blue.

Equations other than Equation 1 are also possible. Equation 2 below proved to be very valid.

CTRL = -C + max (min [(α * V + β * U), (γ * V + δ * U)], C)

Where V = (R-Y) /1.14

U = (B-Y) /2.03

α, β, γ, δ: the whole is not negative, preferably about 1,

C: not negative, preferably between 0 and 0.1.

4 shows that the picking filter 1 of FIG. 1 can be implemented. The input luminance component y-in is applied to the cascade connection of the two delay cells 11 and 13. The input of the delay cell 11 and the output of the delay cells 11 and 13 are respectively an adder 21 to multiply -0.5 through a multiplier 15, a multiplier 17 to multiply 1, and a multiplier to multiply -0.5 ( It is connected to each input of 19). The multiplier 17 can be connected directly. The output of the adder 21 supplies the peaking signal output. The peaking circuit of Figure 4 is a single -1/2, 1, -1/2 FIR filter. The length of the delay line determines the peaking frequency. This will be chosen to compensate for the saturation bandwidth of 700 kHz. Considering the low Q factor of the FIR filter, the optimum peaking frequency is around 1.5 MHz. A delay of about 250 ns per cell was used for the prototype. Perhaps a slightly longer delay time can be chosen for optimal results. Those skilled in the art may wish to replace the peaking filter of FIG. 4 with another peaking filter.

The amount of picking may be controlled by multiplying the picking signal and the control signal CTRL. If the control signal CTRL is zero, there is no peaking. If the control signal CTRL is 1, the peaking is maximum.

5 shows a preferred embodiment of the image display apparatus according to the present invention. Before they are processed, the input components y-in, u-in, v-in are clamped by the clamp circuits 23, 25, 27. The clamped luminance component y-in is input to the peaking filter of FIG. The clamped chrominance components u-in and v-in go through delay cells 29 and 31, respectively, to compensate for the delay and then input to multipliers 41 and 43, respectively. The outputs of multipliers 41 and 43 are summed by adder 45. The sum is clamped by clamp 47 and compared with reference voltage C = Vref by (soft) comparator 49 to obtain control signal CTRL. The compensation delay cells 29 and 31 supply the output luminance components u-out and v-out, respectively. 5 corresponds to equation (1). That is, in the embodiment corresponding to Equation 2, elements 41-49 should be replaced with another processor that calculates CTRL based on U and V.

The output components y-out, u-out and v-out are input to the matrix circuit MX which supplies red (R), green (G) and blue (B) signals. The R, G, and B color signals are processed by the red processor RP, the green processor GP, and the blue processor BP, respectively. The output signals of the color processors RP, GP, and BP are input to the display tube DP.

In FIG. 5, the actual color detector 7 of FIG. 1 is shown in more detail. The sum of U and V is compared with Vref in soft comparator 49. In practice, this comparator consists of a long-tailed pair, which yields a crossover region of about 250mV.

An additional clamp 47 between the U and V adders 45 and the soft comparator 49 is needed to correct the DC biasing. By selecting V _ref , it becomes possible to limit the detection area in such a way that the colorless area is not affected by the processing.

6 shows a multimedia device according to the present invention. This multimedia device has an input 61 for receiving an image signal to be processed by an image processor (IP) 63 comparing the sharpness enhancement circuits SI of the type shown in FIG. The CD-ROM player 65 exists to provide the video and / or data signal I1 and the audio signal S1. The speech processing unit 69 (SND1 portion of the sound card) receives a speech signal from the input 67 and supplies a data signal D1 and a speech signal S2. Another processor 73 operates on the text signal TXT received from the keyboard connector 71. The video and / or data outputs of the devices 63, 65, 69, 73 are input to a video unit (graphics card) 75 which supplies an output video signal to an output 77 to which a monitor (not shown) is connected. The audio outputs S1, S2 and the audio processing unit 69 of the CD-ROM player 65 are connected to the output portion SND2 of the sound card, and the output of this SND2 is connected to the audio output of the multimedia apparatus.

Fig. 7 shows a preferred embodiment of the color-dependent sharpness improvement circuit according to the present invention. The input luminance signal y-in = YO is input to a delay line having eight delay cells 81-88 for obtaining the delay luminance signals Y1-Y8. The signals Y0, Y4, Y8 are input to the low frequency peaking filter 1. A coring circuit 91 and a switch 93 are positioned between the fader 3 and the peaking filter 1 controlled by the color-dependent sharpness control signal CTRL. The switch 93 is controlled by a control signal CDS which indicates whether color-dependent sharpness control according to the invention is required.

The second fader 107 is located between the fader 3 and the adder 5. The signals Y1, Y2, Y4, Y6, Y7 are input to the high frequency peaking filter 94 followed by the coring circuit 95 and the fader 97. By the control signal CFS input to the high frequency peaking filter 94, a selection is made between the 2.38 MHz peaking frequency and the 3.58 MHz peaking frequency.

The total luminance signal YO-Y8 is input to the minimum amplitude and maximum amplitude detector 99, followed by an amplitude detection circuit 101 that determines AMP = MAX-MIN. The detected amplitude AMP is used to control the faders 97 and 107 through the comparators 103 and 105, respectively. The purpose of amplitude-dependent control of the peaking is to prevent blooming of the spots resulting as a result of too large a sharpening enhancement signal, thereby reducing the sharpness. The faders 97 and 107 are controlled by the detection amplitude AMP in such a way that the output signal never exceeds 100% of the maximum amplitude of the input signal.

The offset voltage sources 107-111 are switched in accordance with the control signal CDS (indicating whether color-dependent sharpness improvement control is required) and the control signal CFS (indicating the peaking frequency of the high frequency peaking filter 94). Three different offset voltages selected by 113). The offset voltage selected by the switch 113 is input to the comparator 103, while four offset voltages are input to the comparator 105 by the offset voltage source 115. These different offsets are necessary because the detection amplitude (AMP) is somewhat frequency dependent, allowing the AMP to be adjusted according to CFS. Clearly, if color-dependent sharpness improvement is switched on, the correction according to the AMP must be adjusted, which accounts for the CDS-dependent control of the switch 113.

Another variation between the embodiment of FIG. 7 and the embodiment of FIG. 5 is that the minimum amplitude detection devices 117-119, 121-123 are inserted before the multipliers 41, 43, respectively. Each minimum amplitude detection device includes a series connection of two delay cells 117, 118 and 121, 122, and the minimum amplitude of the received color difference signals (u-in, v-in) and the one supplied by the delay cell. Minimum amplitude detection circuits 119 and 123 for selecting the first and second delayed color difference signals. These minimum amplitude detection devices are used to ensure that color-dependent sharpness control is only effective for sharpening the high frequency components of the colored object while preventing unwanted overshoot from occurring at the boundaries of the colored object.

The sharpness improvement algorithm was tested on test signals with several heavily saturated red, blue and magenta regions. The first conclusion is that the circuit operates satisfactorily and reproduces saturated red and magenta with a more satisfactory effect with both PAL and NTSC signals. The sharpness of saturated parts is increased. It is unlikely that saturated blue with many items exists. Most of the blue is water or sky and the possible noise increase is more pronounced in blue than magenta and red. Therefore, the detection axis has changed to some extent, so that less blue is detected. This improves normal picture quality.

It will be clear that the sharpening enhancement is limited by louder signals such as directly recorded signals. If too much color clarity is present, extra noise may be unsatisfactory. The coring circuits 91 and 95 of FIG. 7 are used to limit sharpness enhancement.

It is noted that the foregoing embodiments rather limit the invention and that those skilled in the art can design many alternative embodiments without departing from the scope of the appended claims. For example, a picking circuit having a controllable picking frequency can be considered, and the picking frequencies for red and magenta are lower than the picking frequencies for other colors. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The present invention can be implemented by hardware and several suitably programmed computers including some unique elements.

Claims (14)

Determining the presence (7, 41-49) of the presence of a predetermined color (R, M) in the video signal (y-in, u-in, v-in) to obtain a control signal (CTRL), Enhancing the sharpness of the image signal (y-in, u-in, v-in) according to the control signal (CTRL). The method of claim 1, wherein the predetermined color comprises red (R) and magenta (M). The method of claim 1 or 2, wherein the step of enhancing the sharpness of the image signal (y-in, u-in, v-) according to the control signal (CTRL) representing the predetermined color (R, M). enhancing the sharpness of the luminance component (y-in) of in). The method of claim 1, wherein the determining step (7, 41-49) comprises the step (41-45) of making a predetermined linear combination of chrominance signals (u-in, v-in). 5. The method according to claim 4, wherein said control signal (CTRL) is obtained by thresholding said predetermined linear combination. 2. The method of claim 1, wherein enhancing (1-5, 11-21) enhancing the sharpness comprises raising a frequency between about 700 kHz and about 2 MHz. The method of claim 6, wherein the step of enhancing the sharpness (1-5, 11-21), Obtaining a difference signal (11-21) by the delay cells (11, 13) and the subtraction unit (15-21) providing a suitable delay for selecting the frequency to be raised, Multiplying the difference signal by the control signal CTRL to obtain a sharpness adjustment signal; Adding (5) the sharpness control signal to the video signal (y-in, u-in, v-in). The luminance component (y-in) of the image signal (y-in, u-in, v-in) is input to the delay cells (11, 13), the sharpness control signal is the luminance A method for improving sharpness that is added to the component (y-in) (5). Means (7, 41-49) for determining the presence of predetermined colors (R, M) in the image signals (y-in, u-in, v-in) to obtain a control signal (CTRL), Means for enhancing the sharpness of the video signal (y-in, u-in, v-in) according to the control signal (CTRL) to obtain an enhanced video signal (y-out, u-out, v-out) Clarity improvement circuit comprising a. 10. The sharpness improvement circuit of claim 9, wherein the predetermined color comprises red (R) and magenta (M). The image signal y according to claim 9 or 10, wherein the means for enhancing sharpness (1-5, 11-21) is in accordance with the control signal (CTRL) representing the predetermined colors (R, M). means for enhancing the sharpness of the luminance component (y-in) of -in, u-in, v-in). 10. The apparatus of claim 9, wherein the determining means (7) comprises a minimum amplitude detection device (117-119, 121-123) indicating the minimum amplitude of the input color difference signal (u-in, v-in) and the delayed color difference signal. And the control signal (CTRL) is only activated on a colored object. A sharpness improvement circuit (1-7) as claimed in claim 9, And means (DT) for displaying said enhanced video signal (y-out, u-out, v-out). A multimedia device for processing a video signal and processing at least one other item from a group comprising data and text and / or voice, An image processor (IP) comprising a sharpness improvement circuit (SI) as claimed in claim 9, Means (75) for supplying said enhanced video signal (y-out, u-out, v-out) and / or said at least one other item.