JPH08265795A - Luminance signal/color signal separation filter - Google Patents

Abstract

PURPOSE: To provide a luminance signal/color signal separation filter capable of accurately separating luminance signals and color signals without picture quality degradation caused by making the scale of a memory to be used small while lowering a cost thereby. CONSTITUTION: Composite video signals sampled by a frequency synchronized with a carrier color signal reference frequency or a horizontal scanning frequency are divided into blocks for picture elements for which the phase of carrier color signals is the same. A maximum/minimum value detection circuit 3 obtains a maximum value and a minimum value for the respective blocks and a gradation threshold preparation circuit 7 prepares plural gradation range threshold values on the basis of the maximum values and the minimum values. Then, sample values are compared with the plural gradation range thresholds and signals for indicating gradation ranges to which the respective sample values belong are obtained, stored in the memory, then read and used for color signal extraction.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a luminance signal / chrominance signal separation filter for separating a luminance signal and a chrominance signal from a composite television signal of the NTSC system, for example.

[0002]

2. Description of the Related Art FIG. 7 is a block circuit diagram of a conventional NTSC system luminance signal / color signal separation filter. In the figure,
Reference numeral 101 is an input terminal for inputting an NTSC composite color television signal, 102 is a frame memory for storing the composite color television signal, and 103 is subtraction for calculating the composite color television signal and the output signal of the frame memory. , 104 is a multiplier, 105 is a subtractor, and 106 and 107 are output terminals.

A conventional luminance signal / chrominance signal separation filter is provided with a frame memory 102 for directly storing image information of one frame, and sample values of an image for one frame are subtracted by using subtractors 103, 105 and a multiplier 104. By the calculation, it was possible to realize the separation without any disturbing component in the still image.

[0004]

A conventional luminance signal / chrominance signal separation filter has a frame memory and calculates a sample value of an image for one frame to perform separation without any interfering components in a still image. While you can
There is a problem that the cost becomes high because a large capacity memory is used.

The present invention has been made in order to solve the above problems, and reduces the cost by reducing the size of the memory to be used, while at the same time, it is possible to obtain an accurate luminance signal without deterioration of image quality. It is an object of the present invention to obtain a luminance signal / color signal separation filter capable of separating signals.

[0006]

According to a first aspect of the present invention, there is provided a luminance signal chrominance signal separation filter having a grayscale level represented by a first bit number at a frequency synchronized with a horizontal scanning frequency or a carrier color signal reference frequency. The converted composite video signal,
A luminance signal / color signal separation filter that performs processing by dividing it into a plurality of blocks according to the position of the pixel corresponding to each sample value on the screen, and obtains the maximum value and the minimum value of the sample value for each block. Creating a threshold value of a gradation range corresponding to a gradation level of a second bit number smaller than the first bit number based on the minimum value detection means (3) and the maximum value and the minimum value. Means (4-7), and encoding means (8) for comparing each sampled value with the thresholds of the plurality of gradation ranges and replacing it with a code (φij) representing the gradation range to which each sampled value belongs. , A memory means (9) for storing a code representing the gradation range, and a code stored in the memory means (9) is read out, and the read code and the maximum value and the minimum value are associated with them. Based on the data (LA, LD) showing the value Decoding means (11) for forming a gradation level signal having a third bit number which is larger than the second bit number for this value, and a color signal using the gradation level signal having the third bit number. A luminance signal color signal separation filter comprising: a color signal extraction filter (12) for performing extraction; and a luminance signal extraction means (13) for extracting a luminance signal using the color signal.

According to a second aspect of the present invention, there is provided a luminance signal / color signal separation filter according to the first aspect, wherein the means (4 to 7) for creating the threshold value of the gradation range uses the maximum value and the minimum value. Reference value creating means (4) for obtaining two reference values (P1, P2) and the average level of the block from the above two reference values
Means (5) for calculating (LA), means (6) for calculating the gradation width index (LD) from the above two reference values, and the average level
(LA) and means (7) for creating threshold values of a plurality of gradation ranges from the gradation range index (LD), and as the value associated with the maximum value and the minimum value, the average level ( LA) and gradation range index (LD) are used.

A luminance signal / chrominance signal separation filter according to a third aspect is the filter according to the first or second aspect, wherein the composite video signal is an NTSC composite video signal or a PAL composite video signal, and each block is on each line. It is characterized in that the block division is performed so that the pixels are formed by every four pixels having the same phase of the carrier color signal.

According to a fourth aspect of the present invention, there is provided the luminance signal / chrominance signal separation filter according to the third aspect, wherein the first of the blocks
Of the pixels on each line have the same phase of the carrier color signal, and the second block constitutes the first block of the pixels on the same line. It is characterized by being composed of the pixel next to the pixel.

According to a fifth aspect of the present invention, in the luminance signal / color signal separation filter according to the third or fourth aspect, block division is performed so that each block is composed of pixels having the same carrier color signal phase even between different lines. It is characterized by being

According to a sixth aspect of the present invention, there is provided a luminance signal / color signal separation filter according to any one of the first to fifth aspects, wherein each of the blocks is composed of 16 sample values of 4 rows and 4 columns. To do.

A luminance signal / color signal separation filter according to a seventh aspect is the filter according to any one of the first to fifth aspects, wherein each of the blocks is composed of 16 sample values of 2 rows and 8 columns. And

A luminance signal / chrominance signal separation filter according to an eighth aspect is the filter according to any one of the first to fifth aspects, wherein each of the blocks is composed of eight sample values of 2 rows and 4 columns. And

[0014]

In the luminance signal chrominance signal separation filter according to claim 1, the composite video signal sampled at a frequency synchronized with the horizontal scanning frequency or the carrier chrominance signal reference frequency is divided into a plurality of blocks, and each block has a maximum value. Values and minimum values are obtained, thresholds for a plurality of gradation ranges are created based on the maximum and minimum values, and the sampled values are compared with the thresholds for the plurality of gradation ranges to perform encoding. Done. Therefore, it is possible to reduce the size of the memory used, reduce the cost, and perform accurate separation of the luminance signal and the color signal without deterioration of the image quality.

In the present invention, the two reference values are obtained by using the maximum value and the minimum value, the average level of the block is calculated from the two reference values, and the gradation width index is calculated from the two reference values. Is calculated, and threshold values for a plurality of gradation ranges are created from the average level and the gradation width index.
Therefore, with a relatively simple circuit configuration, it is possible to realize the encoding in which the image quality of the block shape is less deteriorated.

In claim 3, the composite video signal is NT.
It is an SC composite video signal or a PAL composite video signal, and each block realizes accurate encoding / decoding using pixel data having a strong correlation of the carrier color signal among the pixels on each line. Therefore, it is possible to perform accurate encoding / combining using pixel data having a strong correlation.

According to another aspect of the present invention, the first block of the blocks is composed of four pixels having the same carrier color signal phase among the pixels on each line, and the second block is the same. Of the pixels on the line, the pixel is formed by the pixel next to the pixel forming the first block.
Therefore, although each block is composed of pixels every four pixels, block coding can be performed for all pixels.

According to another aspect of the present invention, block division is performed so that each block is composed of pixels having the same carrier color signal phase even between different lines. Therefore, accurate encoding / decoding can be performed using the pixel data having a strong correlation across a plurality of lines.

In the sixth aspect, each of the blocks is composed of 16 sample values of 4 rows and 4 columns. Therefore, it is possible to reduce the size of the memory to be used, reduce the cost, and separate the luminance signal and the color signal with less image quality deterioration.

In the present invention, each of the blocks is composed of 16 sample values in 2 rows and 8 columns. Therefore, it is possible to further reduce the size of the line memory that stores the pixel data input to the maximum / minimum value detection circuit.

In the eighth aspect, each of the blocks is composed of eight sample values of 2 rows and 4 columns. Therefore, as compared with claims 6 and 7, the compression rate is slightly reduced, but the encoding / decoding with less error is realized.

[0022]

【Example】

Example 1. FIG. 1 is a schematic block diagram showing an embodiment of the present invention. Reference numeral 1 is an input terminal to which an NTSC composite color television signal sampled at a frequency synchronized with a horizontal scanning frequency is given. Reference numeral 2 denotes a composite color television signal input from the input terminal 1 in 1 of 4 rows and 4 columns.
The line memory 3 stores the data so that it is divided into a plurality of blocks each consisting of 6 sample values, and 3 is a maximum / minimum value detection circuit.
The maximum value and the minimum value of the 6 pixels are detected. Reference numeral 4 is a reference value creating means, which obtains representative values of the maximum value and the minimum value, determines two sample value groups from the two representative values, and calculates the average value of the two sample value groups as a reference value. To do. Reference numeral 5 is an average level calculation circuit for calculating the average level of the block from the two reference values, 6 is a gradation width index calculation circuit for calculating the gradation width index of the block from the two reference values, and 7 is the average level and the gradation width index. A gradation range threshold creating circuit for creating thresholds for a plurality of gradation ranges, 8 compares 16 sample values with the above-mentioned gradation range threshold, and each of the 16 sample values indicates which floor A comparison circuit that expresses whether it is included in the key range with a bit number smaller than the bit number of the input signal, 9 is a memory circuit, 10a and 10b are buffer memories, 11 is a decoding circuit, 12 is a color signal extraction filter,
13 is a subtracter, and 14 and 15 are output terminals.

The operation of the embodiment shown in FIG. 1 will be described below. The line memory 2 stores the composite color television signal input from the input terminal 1 so as to be divided into a plurality of blocks each consisting of 16 pixel data of 4 rows and 4 columns. The color subcarriers of the composite color television signal have the same phase every four pixels. Therefore, the method of extracting the 16 pixel data is to select the pixel data of the same phase of the color signals among the pixel data of 4 rows and 16 columns as shown in FIG.

That is, each block is constituted by pixels of every four pixels on each line. Then, the next block (second block) is formed by the pixel (on the right side on the line) next to the pixel forming one block (first block). Further, a pixel next to the pixel forming the second block forms a third block. further,
A pixel next to the third block forms a fourth block. The fifth block is formed next (on the right side) to the area occupied by the first to fourth blocks. The blocks are constructed in the same manner below. By doing this,
Accurate encoding / combining can be performed using pixels having a strong correlation, and all pixels can be assigned to any block and block encoding can be performed for all pixels.

Further, it is desirable to form a block by pixels having the same color subcarrier phase even between different lines.
By doing this, it is possible to perform accurate encoding / compositing by using pixels having a strong correlation that extend over a plurality of lines.

The pixel data of each block is encoded as follows. The maximum / minimum value detection circuit 3 has a maximum value (Lmax) and a minimum value (Lmi) of 16 pixel data.
n) is detected. The reference value creating means 4 uses Lmax and Lm.
The minimum representative value P1 and the maximum representative value P2 are obtained from in as follows. P1 = (Lmax + 3Lmin) / 4 P2 = (3Lmax + Lmin) / 4

A sample value group having a value of P1 or less is determined, the average value of the sample value group is calculated as a reference value Q1, and a sample value group having a value larger than P2 is determined.
The average value of this sample value group is calculated as the reference value Q8. The average level calculation circuit 5 calculates the average level LA of the block from Q1 and Q8 as follows. LA = (Q1 + Q8) / 2

The gradation width index calculating circuit 6 calculates the gradation width index LD of the block from Q1 and Q8 as follows. LD = Q8-Q1

The gradation range threshold generation circuit 7 prepares a plurality of gradation range thresholds from LA and LD. The gradation level corresponding to the gradation range is represented by 3 bits, and is smaller than the number of bits (for example, 8 bits) representing the gradation level of the original sample value. The threshold value of the gradation range is determined as shown in FIG. 3, for example. Here, L1 to L6 have the following values. L1 = LA- (3/8) LD L2 = LA- (1/4) LD L3 = LA- (1/8) LD L4 = LA + (1/8) LD L5 = LA + (1/4) LD L6 = LA + (3/8) LD

The comparison circuit 8 compares each of the 16 sampled values with the gradation range threshold values L1 to L6, LA,
Which gradation range each of the sample values is included in, ie, which gradation level it corresponds to, is represented by φij of 3 bits as shown in FIG.

The component φij encoded as described above is temporarily stored in the memory circuit 9 and then supplied to the decoding circuit 11. On the other hand, the gradation width index LD and the average level LA are applied to the decoding circuit 11 after being delay-adjusted by the buffer memories 10a and 10b, respectively.

The decoding circuit 11 outputs a signal decoded from the input φij and LD, LA as shown in Table 1.

[0033]

[Table 1]

By this decoding, the original 256 gradation level signal is obtained. The color signal extraction filter 12 inputs the output a of the decoding circuit 11 and the output b of the line memory 2 as shown in FIG. 4, and calculates them by the subtracter 21 and the multiplier 22 to extract the color signal c. it can.

At the time of decoding, a signal represented by 8 bits is formed like the original signal, but the number of bits representing this gradation level is different from the number of bits representing the original gradation level. It may be.

Example 2. In the first embodiment, as shown in FIG. 2, 16 pixel data in which the color signals have the same phase are extracted from the pixel data in 4 rows and 16 columns. Instead of this, a block of 2 rows and 8 columns in which the color signals have the same phase may be selected from the pixel data of 2 rows and 32 columns as shown in FIG. In this way, a smaller line memory 2 can be used.

Example 3. Decoding circuit 11 according to the first embodiment
The output value increase is 2LD / 16 or 3LD / 16
Is a multiple of LD / 16, which is advantageous in limiting the hardware scale. Alternatively, the increment of the output value may be constant. An example is shown in Table 2. In this example, the increment of the output value is 2LD / 14
And is constant. By doing so, a more natural image can be obtained.

[0038]

[Table 2]

Embodiment 4 FIG. The line memory 2 in the fourth embodiment stores the composite color television signal input from the input terminal 1 so as to be divided into a plurality of blocks each of which is composed of 8 sample values of 2 rows and 4 columns. Eight pieces of pixel data are taken out so as to select pixel data of 2 rows and 16 columns and having the same color signal phase, as shown in FIG. As a result, although the compression rate is lower than in the case of block division in the first and second embodiments, encoding / decoding with less error can be realized.

In the above embodiment, the NTSC signal in which the phase of the color signal under one line is inverted has been described. However, the pixel data in which the phase of the color signal under one line changes by 90 degrees By selecting the ones having the same phase, it is possible to support a PAL system composite color television signal. Further, in the above embodiment, the pixel data of one line below is taken, but one line is skipped and 2
It goes without saying that pixel data having the same phase under the line may be extracted.

[0041]

According to the luminance signal / chrominance signal separation filter of the first aspect, it is possible to reduce the cost by reducing the size of the memory used and to accurately separate the luminance signal / color signal without deterioration of image quality. .

According to the second aspect, it is possible to realize the encoding in which the image quality of the block shape is less deteriorated with a relatively simple circuit configuration.

According to the third aspect, it is possible to perform accurate encoding / compositing using pixel data having a strong correlation.

According to the fourth aspect, although each block is composed of every four pixels, block coding can be performed for all pixels.

According to the fifth aspect, it is possible to perform accurate encoding / decoding by using the pixel data having a strong correlation across a plurality of lines.

According to the sixth aspect, it is possible to reduce the cost by reducing the size of the memory to be used, and to separate the luminance signal and the chrominance signal with less image quality deterioration.

According to the seventh aspect, it is possible to further reduce the size of the line memory for storing the pixel data input to the maximum / minimum value detection circuit.

According to the eighth aspect, encoding / decoding with a relatively small compression rate but a smaller error is realized.

[Brief description of drawings]

FIG. 1 is a schematic block diagram of an embodiment of an NTSC system luminance signal / color signal separation filter according to a first embodiment.

FIG. 2 is an explanatory diagram showing an array for extracting 16 pixel data from a composite color television signal in the first embodiment.

FIG. 3 is an explanatory diagram showing level values in a gradation range threshold generation circuit in FIG.

FIG. 4 is a block diagram showing an embodiment of a color signal extraction filter in FIG.

FIG. 5 is an explanatory diagram showing an array for extracting 16 pixel data from a composite color television signal according to the second embodiment.

FIG. 6 is an explanatory diagram showing an array for extracting eight pixel data from a composite color television signal according to a fourth embodiment.

FIG. 7 is a block circuit diagram of a conventional luminance signal / color signal separation filter.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1: Input terminal, 2: Line memory, 3: Maximum / minimum value detection circuit, 4: Reference value creation means, 5: Average level calculation circuit, 6: Gradation width index calculation circuit, 7: Gradation range threshold generation Circuit, 8: comparison circuit, 9: memory circuit, 10a, 10b: buffer memory, 11: decoding circuit, 12: color signal extraction filter, 13: subtractor,
14, 15: output terminals.

Claims (8)

[Claims] 1. A composite video signal sampled at a gradation level represented by a first bit number at a frequency synchronized with a horizontal scanning frequency or a carrier color signal reference frequency is displayed on a screen of pixels corresponding to each sampled value. Is a luminance signal / color signal separation filter that performs processing by dividing it into a plurality of blocks according to the position of, and maximum / minimum value detection means (3) for obtaining the maximum value and the minimum value of sample values for each block. Means (4 to 7) for creating a gradation range threshold value corresponding to a gradation level represented by a second bit number smaller than the first bit number based on the maximum value and the minimum value.
And an encoding means (8) for comparing each sampled value with a threshold value of the plurality of gradation ranges and replacing it with a code (φij) representing the gradation range to which each sampled value belongs, and representing the gradation range. Memory means for storing codes (9)
And reading the code stored in the memory means (9), and based on the read code and the data (LA, LD) indicating the maximum value and the minimum value or the values associated therewith, Decoding means (11) for forming a gradation level signal represented by a third bit number larger than the second bit number with respect to the sampled value, and the gradation level signal of the third bit number are used. A brightness signal color signal separation filter comprising a color signal extraction filter (12) for extracting a color signal and a brightness signal extraction means (13) for extracting a brightness signal using the color signal. 2. A means (4-7) for creating a threshold value of the gradation range is a reference value creating means (2) for obtaining two reference values (P1, P2) using the maximum value and the minimum value. 4), a means (5) for calculating the average level (LA) of the block from the above two reference values, a means (6) for calculating a gradation range index (LD) from the above two reference values, and the above average level (LA) and means (7) for creating threshold values for a plurality of gradation ranges from the gradation width index (LD), and as the value associated with the maximum value and the minimum value, the average level ( The luminance signal / chrominance signal separation filter according to claim 1, wherein LA) and gradation range index (LD) are used. 3. The composite video signal is an NTSC composite video signal or a PAL composite video signal, and each block is composed of pixels for every four pixels having the same carrier color signal phase among the pixels on each line. 3. The block division is performed as follows.
The luminance signal / color signal separation filter described in. 4. The first block of the blocks is composed of four pixels having the same carrier color signal phase among the pixels on each line, and the second block is
The luminance signal / color signal separation filter according to claim 3, wherein the pixel is formed by a pixel next to a pixel forming the first block among pixels on the same line. 5. The block division is carried out so that each block is composed of pixels having the same phase of the carrier color signal even between different lines.
The luminance signal / color signal separation filter described in. 6. The luminance signal chrominance signal separation filter according to claim 1, wherein each of the blocks is composed of 16 sample values of 4 rows and 4 columns. 7. The method according to claim 1, wherein each of the blocks is composed of 16 sample values in 2 rows and 8 columns.
6. The luminance signal / color signal separation filter according to any one of items 1 to 5. 8. The luminance signal chrominance signal separation filter according to claim 1, wherein each of the blocks is composed of eight sample values of 2 rows and 4 columns.