JPH09214957A - Video compression recorder using correlation - Google Patents

Abstract

(57) [Abstract] [Purpose] To provide a compression method of a stereoscopic image capable of obtaining high horizontal resolution. [Structure] One field is pre-scanned and a compression rate by quantization and coding is set. One frame is compressed at the compression rate set by this prescan.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a system for compressing and recording a video signal.

[0002]

2. Description of the Related Art Recently, a method of compressing a digitized video signal using a high-efficiency coding technique and recording the data on a magnetic tape, a disk, a semiconductor memory or the like has become popular. As a typical compression method, there are MPEG, JPEG, SDVTR standards, etc., but each method uses a method in which the amount of data after compression is constant regardless of the pattern. this is,
This is because the aim is to secure a certain recording time by embezzling the recording medium, and there is a limit to the capacity (recording rate) that can be recorded in a unit time depending on the recording medium. JPEG and SDVTR without time axis compression
As a standard, a method that has been widely used recently as a compression algorithm for performing intra-field or intra-frame compression is a method of performing DCT (discrete cosine transform), further quantizing, and entropy coding. DCT is a so-called orthogonal transform that transforms a video signal converted into digital data from a spatial coordinate axis to a frequency coordinate axis so that it can be easily encoded. Quantization omits a component with a small frequency energy (high frequency component). The technique, entropy coding, is a technique in which short codes are assigned to high-frequency data and long codes are assigned to low-frequency data.

When such DCT, quantization, and entropy coding are performed with a constant quantization coefficient, the amount of data after compression changes depending on the pattern. That is, the amount of data after compression is large for a fine pattern such as a forest or a hand-knitted sweater, and the amount of data after compression is small for a pattern with few high-frequency components such as a smooth white wall. When moving images are compressed continuously, there is no problem when the amount of data is small, but when the amount of data is large, there is a limit to the capacity (recording rate) that can be recorded per unit time depending on the recording medium, as described above. Therefore, the compressed data that continuously appears cannot be recorded within a predetermined time, that is, overflow occurs.

There are various methods for avoiding this, but as a general method, a process of estimating the amount of data after the compression is performed before the actual compression is performed, and then the quantization is performed so that the target data size is reached. A method is used in which actual compression is performed by changing the coefficient and encoding settings.

FIG. 3 shows a block diagram of a conventional example, and FIG. 4 shows a timing chart of the conventional example. In this example, since the code amount calculation is performed before compression, the A / D unit 100, D / A
The data transfer speed of the unit 110 is doubled by inputting / outputting the data transfer speed to / from the compression / expansion unit via the field memories 125 and 126 to perform double speed transfer. The read / write section is omitted here,
Only write speed and read speed are transfer speed (fre
q).

The signal flow will be described with reference to the block diagram of FIG. First, the A / D unit 100 converts the luminance signal Y and the color signals Cr and Cb from analog signals into digital signals. The video signal of one frame is stored in the odd and even field memories 125 and 126, and writing and reading are performed by a read / write control unit (not shown). Here, the read / write section is omitted, and only the writing speed and the reading speed are described as the transfer speed. The data from the field memories 125 and 126 is compressed and recorded in the recording unit.

The compression unit and the decompression unit are pre-processed to reduce the redundancy, but here the DCT (DESCRET C
OSINE TRANSFER) is used. The video signal whose redundancy is reduced by the DCT is quantized and encoded to compress the data amount and recorded in the recording unit 700. A buffer memory 6 is provided between the compression / expansion unit 200 and the recording unit 700.
00, the data is temporarily stored in the buffer memory and then recorded in the recording unit 700.

Quantization / inverse quantization section 220 and encoding
The control of the inverse encoding unit 230 is performed by feedback, and the output data of encoding is once fed back to the quantization control unit 240 or the encoding control unit 250 to set the compression rate before the processing. There is.

Therefore, the prescan is performed before compressing the data for one field, and then the data for one field is adaptively compressed (see FIG. 4). Note that FIG. 4A shows a signal for switching between an odd field and an even field, and FIGS. 4B and 4C show an odd field (O).
The read / write timing of the field memory (DD) is shown. 4D and 4E show even fields (E
The read / write timing of the (VEN) field memory is shown. Further, FIG. 4F shows output data from the compression unit.

According to this structure, one system of compression system performs pre-scanning and code amount calculation before actual compression, and based on the result, the quantization coefficient and the encoding setting are redone and then the actual compression process is performed. It can be performed.

However, the sampling clock frequencies of the A / D and D / A sections can be increased only up to 1/2 of the data transfer rate of the compression / expansion section, and the frequency band in the image when reproduced (expanded). Has a drawback that it is narrow and the so-called horizontal resolution is low.

[0012]

That is, since the compression rate is set for each field by the prescan, the ratio of the writing speed and the reading speed can be as high as 1/2 and the horizontal resolution is deteriorated. there were.

The present invention has been made in view of the above problems, and an object of the present invention is to provide a moving image recording apparatus capable of achieving high horizontal resolution.

[0014]

In order to achieve the above-mentioned object, in the invention according to claim 1 of the present invention, the image information of any one field is read out twice from the field memory, and the quantization and the coding are performed. It is characterized in that a compression rate is set by compression, image information of all fields constituting one frame is compressed, and the compressed image information is recorded on a recording medium.

Specifically, it can be easily realized by using the video data of one of the fields constituting one frame for the prescan.

[0016]

Embodiments will be described with reference to the drawings. FIG. 1 is a block diagram showing a stereoscopic image compression system according to an embodiment of the present invention. The same components as those in FIG. 3 are designated by the same reference numerals and will be described.

The applicant of the present invention performs prescanning using one of two fields by utilizing the correlation without performing prescanning for each field in order to change the ratio of the writing and reading speeds and increase the horizontal resolution. The processing time was shortened. Further, since the image information of two fields can be selectively transferred to the compression / expansion unit and processed, a single processing system is sufficient.

The signal flow will be described below with reference to FIG. In the figure, odd numbers are marked as ODD and even numbers are marked as EVEN. Further, a number is added after ODD or EVEN so that the processing of the field can be understood. 2A shows a signal for switching between the odd field and the even field, and FIGS. 2B and 2C show the read / write timing of the field memory of the odd field (ODD). 2D and 2E show the read / write timing of the even-numbered field (EVEN) field memory. Further, FIG. 2F shows output data from the compression unit.

When the switching signal shown in FIG. 2A is high, writing to the field memory is performed as shown in FIG. 2B. Here, the ratio of the writing speed and the reading speed is changed. Writing is 4 / 3f and reading is 2f. Therefore, the prescan and the processing of the odd field ODD1 and the even field EVEN1 are completed within one frame period.

As can be seen from FIG. 2 (c), the prescan is only once in the odd field (ODD1),
Since the prescan is not performed in the even field, the processing time can be shortened.

[0021]

As described above, according to the present invention, the sampling frequency of the A / D section is reduced by utilizing the correlation between fields to reduce the number of times the compression ratio is set and the processing time. It is a high resolution and high horizontal resolution.

[Brief description of the drawings]

FIG. 1 is a diagram when a compression device showing an embodiment according to the present invention is applied to a recording device.

FIG. 2 is a timing chart for explaining the operation of FIG.

FIG. 3 is a diagram showing a conventional compression device.

FIG. 4 is a timing chart for explaining the operation of FIG.

[Explanation of symbols]

 100 ... A / D section 110 ... D / A section 200 ... Compression / expansion section 700 ... Recording section

Claims (1)

[Claims] Claim: What is claimed is: 1. Pre-processing for reducing redundancy of image information, and compression means for compressing the pre-processed image information by quantizing and encoding the image information of a plurality of fields. A plurality of field memories for storing information, a writing means for writing the image information to these field memories, a reading means for reading the image information from these field memories, and an image information from the plurality of field memories Output means for selectively outputting to the compression means, giving a predetermined ratio between the writing speed by the writing means and the reading speed by the reading means, and the image information of any one field is stored in the field memory. From the above-mentioned reading means by the above-mentioned reading means, and by the first reading, the quantization and coding Sets the compression ratio, moving picture compression recording apparatus characterized by compressing the remaining fields constituting the second reading and the frame at the compression ratio.