JP3009898B2 - Data band compression encoding method and band compression encoding apparatus for image signal - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Summary] The present invention relates to a data band compression encoding method and a band compression encoding device for an image signal using a change of encoded data having a low significance in DCT encoding or the like. Bandwidth compression coding in which the output code amount increases in proportion to the number of data other than the specific value data in the time-series data of a predetermined length including a plurality of data, with the aim of contributing to the Detecting the maximum absolute value data in the input predetermined-length time-series data,
For each of the data, when the ratio of the data to the detected absolute value maximum data is smaller than a predetermined value, the specific value data is forcibly output instead of the data, and a plurality of data In a band compression encoding apparatus for an image signal in which the output code amount increases in proportion to the number of data other than the specific value data included in the predetermined-length time-series data composed of a maximum value detection circuit, a delay circuit,
And a judgment circuit.

[Industrial applications]

The present invention relates to a data band compression coding method and a band compression coding device for an image signal using a change of coded data of low significance in DCT coding or the like.

In the transmission of image signals, the construction of a transmission system has been promoted so as to transmit the minimum necessary information while eliminating redundancy included in the image signals as much as possible. In order to achieve the object, a band compression coding method of an image signal is used.

[Conventional technology]

As one of the band compression coding methods of the image signal, DC
There is T (Discrete Cosine Transform) coding. In this method, a screen is divided into blocks of a certain size (for example, 8 pixels × 8 lines), and DCT operation is performed for each block to convert pixel area data (brightness and color density) into frequency domain data. Things. Then, quantization and variable length coding are performed and transmitted. In DCT coding, scanning is performed to transmit two-dimensional data through a one-dimensional transmission path. An example of the scan is shown in FIG. The 64 data starting from the upper left (1,1) to the lower right (8,8) are scanned in order. In encoding the data converted into one dimension in this way, since a large number of 0s are generally included, instead of converting the data one by one into a code, as shown in FIG. Channel coding is performed in such a manner that one code is assigned by combining the length of 0 (zero run) and the next non-zero coefficient. That is, the number of generated codes is equal to the number of non-zero data.

[Problems to be solved by the invention]

In the coding of the conventional band compression coding method described above, consecutive 0s in the time-series data scanned and output and non-zero data following it are assigned to one code. Is merely determined whether it is 0 or not, and all data that is non-zero is transmitted. Therefore, even if the non-zero data is data that is not important for data transmission, the data is transmitted. Therefore, the encoded information of such data is still included in the transmission information as redundant information. Will be.

The present invention has been created in view of such problems,
It is an object of the present invention to provide a data band compression encoding method and a band compression encoding device for image signals, which contribute to enhancement of the degree of compression of encoded data and reduce communication costs.

[Means for solving the problem]

FIG. 1 is a block diagram showing the principle of the present invention. As shown in this figure, the data band compression encoding method of the present invention employs a data band encoding method in which the output code amount increases in proportion to the number of data other than the specific value data in the time-series data of a predetermined length including a plurality of data. In the band compression encoding, data having a maximum absolute value in the input time-series data having a predetermined length is detected, and for each of the data, a ratio of the data to the data having the maximum detected absolute value is determined in advance. When the value is smaller than a given value, the specific value data is forcibly output instead of the data. A band compression encoding apparatus for an image signal according to the present invention includes a band compression encoding apparatus for an image signal whose output code amount increases in proportion to the number of data other than zero data included in predetermined-length time-series data including a plurality of data. The gasifier is composed of the following components. Its components are:
A maximum value detecting circuit 2 for detecting data having a maximum absolute value in the input predetermined-length time-series data, and a delay circuit 4 for delaying the data by a time required for detecting the maximum value
And a determination circuit 6 for outputting zero instead of the data when the ratio of the detected data to the data having the maximum absolute value is smaller than a predetermined value.

(Operation)

For example, in DCT coding of an image signal, for example, in coding with continuous zero and non-zero data immediately following this, data having a maximum absolute value in time-series data (block to be scanned) of a predetermined length of time series data is obtained. , For example, in the maximum value detection circuit 2.

Next, the input data is compared with the data having the maximum absolute value from the maximum value detection circuit 2. When the ratio of the input data to the data having the maximum absolute value is smaller than a predetermined value, the determination circuit 6 outputs a predetermined value such as zero instead of the input data.

Since coefficients with low significance in the conventional DCT coding and the like are deleted, it is useful for enhancing the compression of coded data to be transmitted.

〔Example〕

FIG. 2 shows a first embodiment of the present invention. In this figure, 1 is a DCT circuit, 2 is a maximum value detection circuit, 4 is a delay circuit, and 6 is a judgment circuit.

Each time the DCT circuit 1 receives the data in the block in the frequency domain, the maximum value detection circuit 2 detects the data having the maximum absolute value in the block to which the data belongs up to the data. It has the configuration as shown. In FIG. 3, reference numeral 21 denotes an absolute value output circuit, which is constituted by a ROM or the like. 22 is a comparison circuit, which is a flip-flop circuit (hereinafter, referred to as a flip-flop circuit).
It is called FF circuit. 2.) Compare the maximum value before the input data in the block held by 23 with the absolute value of the input data. Reference numeral 24 denotes an inverter that receives a block head instruction signal (BLK). An AND gate 25 receives the outputs of the inverter 24 and the comparison circuit 22. Reference numeral 26 denotes a selector for outputting either the output value of the absolute value detection circuit 21 or the output value of the FF circuit 23 to the data inputs of the FF circuits 23 and 27 according to the output of the AND gate 25. The block input signal is supplied to the set input of the FF circuit 27.

The determination circuit 6 determines a predetermined ratio between the data having the maximum absolute value from the maximum value detection circuit 2 and the output data of the DCT circuit 1 delayed by the delay circuit 4 by the time required for the detection operation in the maximum value detection circuit 2. Depending on whether the value is equal to or less than the value, either the input data is output as is or 0, and the configuration is shown in FIG. In FIG. 4, reference numeral 61 denotes a conversion circuit for outputting data obtained by multiplying the data having the maximum absolute value by a predetermined ratio. The predetermined ratio is given to the coefficient information input of the conversion circuit 61. 62 is an absolute value output circuit for outputting the absolute value of the data from the delay circuit 4. A comparison circuit 63 compares the output value of the conversion circuit 61 with the output value of the absolute value output circuit 62 and outputs a selection signal. A selector 64 outputs input data (output data of the delay circuit 4) or 0 for each data according to the selection signal.

 The operation of the embodiment of the present invention having the above configuration will be described below.

When the number of data in one block is n, 1 in the block
When the data (d ₁ ) (see (2) in FIG. 5) is input, the BLK signal is 1 as shown in (1) in FIG. 5, and is 0 otherwise. The operation of the selector 26 is to output the lower input signal when the selection signal is 0 and to output the upper input signal when the selection signal is 1.

The input first data is first converted into an absolute value by an absolute value output circuit 21 and supplied to a comparison circuit 22 and a selector 26. As described above, at the time of the first data input, since the BLK signal is 1 and the output of the inverter 24 is 0,
The selection signal 0 is supplied to the selector 26 regardless of the output of the comparison circuit 22.
Is supplied. The absolute value of the first data appears at the output of the selector 26 and is latched by the FF circuit 23. This is indicated by P in FIG. 5 (3). P represents the one having the largest absolute value among the data up to that point.

Next, when the second data (see (2) in FIG. 5) is input, P, ie, the first
The absolute value of the data is given, and the absolute value of the second data is given to the lower input. The comparison circuit 22 outputs 0 when the absolute value of the lower side, that is, the input data is larger than the upper absolute value, that is, the maximum absolute value up to that point, and outputs 1 otherwise. Since the BLK signal is 0 and the signal given from the inverter 24 to the AND gate 25 is 1, the output of the comparison circuit 22 is used as it is as the selection signal. As a result, the selector 26 outputs the larger value. This is latched by the FF circuit 23. (3) in FIG.
D _i is the larger of the absolute values of D _i-1 and d _i which is shown as P in. In this way, when the last data d _n of the block is input, the output of the selector 26 appears the largest absolute value of that block (see (4) of FIG. 5). By latching this in the FF circuit 27 at the rising edge of the BLK signal of the next block, the desired maximum absolute value in the block can be output to the MAX signal line 28. This value is held for one block period in the configuration of FIG.

The output of the maximum value detection circuit 2, that is, the maximum value i in the block, is converted into an output O by the conversion circuit 61. The output O is, for example, O = i / 100. The conversion ratio 1/100 is input as coefficient information in FIG. The absolute value of the data from the delay circuit 4 is output from the absolute value output circuit 62. The absolute value is compared with the value of the output O by the comparison circuit 63. If the input data is larger, 0 is output, and if the input data is smaller, 1 is output. This output is input to the selector 64 as a selection signal. When the selection signal is 0, the lower input, that is, input data is selected, and when the selection signal is 1, the upper input (fixed to 0 in this case) is selected. As a result, input data smaller than a threshold value (O) obtained by multiplying the maximum absolute value (i) in the block by a certain value (for example, 0.01) is replaced with 0.

In this way, it is possible to delete information that is not important in the reception and reproduction of the above-mentioned conventional method, which is useful for enhancing the compression capability of encoding. It also brings benefits to communication costs.

FIG. 6 shows a coefficient information generating circuit for constituting another embodiment of the present invention. The coefficient information generating circuit 7
As shown in the figure, it comprises a code conversion circuit 71, an information amount counter 73, and a conversion circuit 75. The output of the conversion circuit 75 is provided as coefficient information to the coefficient information input of the conversion circuit 61 in FIG. That is, in another embodiment of the present invention, the coefficient information is variably set according to the code amount.

The coefficient information is generated from the coefficient information generation circuit 7 as follows.

The output OUT of the determination circuit 6 is supplied to a code conversion circuit 71, which converts the output OUT into a code length when transmitted or stored. By an information amount counter 73 that accumulates this for a certain period,
Amount of code (number of bits) from accumulation start to each data
Is weighed. This is input to the conversion circuit 75. Conversion circuit
In step 75, a value obtained from input / output characteristics as shown in FIG. 7, for example, is given to the determination circuit as coefficient information in accordance with the input code amount.

The variable setting of the coefficient information increases the value to be multiplied by the maximum absolute value as the code amount increases, and increases the number of data to be set to 0 in the determination circuit. As a result, the reduction in the number of transmission codes is further enhanced. This contributes to the control of the generated code amount.

In the above-described embodiment, an example of encoding focusing on the continuation of 0 values has been described. However, the application of the present invention to encoding in which attention is focused on the continuation of another specific value is lost at all. is not.

〔The invention's effect〕

As is apparent from the above description, according to the present invention, insignificant information is deleted in reception and reproduction, so that the encoding compression capability is improved. Helps reduce communication costs. If the coefficient information is variably set, its effectiveness is enhanced.

[Brief description of the drawings]

 1 is a block diagram showing the principle of the present invention, FIG. 2 is a diagram showing an embodiment of the present invention, FIG. 3 is a detailed diagram of a maximum value detecting circuit, FIG. Is an operation timing chart of the maximum value detection circuit, FIG. 6 is a detailed diagram of the coefficient information generation circuit, FIG. 7 is a code amount-coefficient information conversion curve diagram, FIG. It is a figure showing data grouping. 1 to 4 and 6, 1 is a DCT circuit, 2 is a maximum value detection circuit, 4 is a delay circuit, 6 is a judgment circuit, and 7 is a coefficient information generation circuit.

──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-62-133881 (JP, A) JP-A-63-123426 (JP, A) JP-A-63-132530 (JP, A) (58) Survey Field (Int.Cl. ⁷ , DB name) H03M 7/30

Claims (2)

(57) [Claims] In a data band compression coding in which an output code amount increases in proportion to the number of data other than specific value data in a predetermined length of time-series data including a plurality of data, said input predetermined length Detecting data having a maximum absolute value in the time-series data, and specifying the data instead of the data when a ratio of the data to the data having the detected absolute maximum is smaller than a predetermined value. A data band compression encoding method characterized by forcibly outputting value data. 2. A band compression encoding apparatus for an image signal in which an output code amount increases in proportion to the number of data other than zero data included in a predetermined-length time-series data composed of a plurality of data. A maximum value detection circuit (2) for detecting data having a maximum absolute value in the predetermined-length time-series data; a delay circuit (4) for delaying the data by a time required for the detection of the maximum value; A decision circuit (6) for outputting zero instead of the data when the ratio of the data to the detected absolute value maximum data is smaller than a predetermined value; Compression encoding device.