JPH11272443A - Data processor and storage medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve processing efficiency without the need of a data operation in a bit unit by executing a conversion processing on data with a first length and data with a second length to which non-used data is added, based on a conversion table. SOLUTION: When compiling (compressing) data where six subsequent bits are effective in one byte, three types of compilation tables are available. 3-8 bits in eight left bits and 3-4 bits in eight right bits, that is, eight bits in total, on a first byte and a second byte (objective area α1 ) from its head are extracted by using a compilation table A111, and they are stored in an output area 1. In the compilation table A111, integer values constituted of effective bits are stored in respective array elements being one-dimensional arrays, which correspond to the integer values of two bytes and have the sizes of 2<16> . When a value stored in the address of the integer value represented by the objective area α1 is taken in, it is equivalent to the extraction of the effective bit.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION As shown in FIG. 1, according to the present invention, a compression process for packing only valid bit data into bit data in a data area or a data operation for each group of valid bit data is performed. The present invention relates to bit data editing processing, such as expansion processing for expanding data into units (for example, byte units) that are easy to perform.

2. Description of the Related Art In many cases, data such as earth observation images obtained by satellites are not always in a data format of a unit (byte unit) that can be easily processed by a computer due to the characteristics (resolution) of a sensor. (For example, one pixel has 6 bits or 10 bits.) The present invention realizes efficient processing in such a case.

[0003]

2. Description of the Related Art In conventional bit data editing processing, data has been extracted for each bit and transferred (shifted).

[0004]

In the prior art, since data is processed in units of bits, processing efficiency is low (processing time is long). The present invention has been made in view of such circumstances, and has as its object to realize efficient processing.

[0005]

The object of the present invention is to provide a conversion table storing a relationship between a data position to be converted and a data position to be converted, and a data having a first length based on the conversion table. Conversion means for performing a conversion process between data having a longer length than the first length and data of a second length obtained by adding unused data to data of the first length. Is done.

[0006]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below in detail.
FIG. 2 shows a system configuration in a case where the present invention is applied to bit data editing processing. The system includes a data 10 in which valid bits and invalid bits coexist, a data 20 composed of only valid bits, and an editing program 100 that inputs these data and performs compression processing or decompression processing of bit data, respectively. And an editing table for compression 110 used in the compression process and an editing table for expansion 120 used in the expansion process.

First, the concept of the compression process using the compression editing table will be described by taking as an example a case where the last 6 bits in a 1-byte area are packed with valid data and edited (compressed). As described in detail below, there are three types of editing tables in this case. In FIG. 3, 10 is the first 2 in 1 byte.
The data where the bit is invalid and the last 6 bits are valid, 111,
Reference numerals 112 and 113 denote editing tables for compression, and reference numeral 20 denotes data compressed by removing invalid bits.

First, for the first and second bytes (target area α ₁ ) from the beginning, 3 to 8 of the left (upper) 8 bits
A total of 8 bits, 3 to 4 bits out of the 8 bits and the right (lower) 8 bits, are extracted using the editing table A, and the output area 1
To be stored. For this purpose, the first editing table A is used. The editing table A is, as shown in FIG.
It is a one-dimensional array having a size of 2 ¹⁶ corresponding to an integer value of a byte (0 to 2 ¹⁶ -1). Each array element has valid bits (3 to 8 bits of the upper 8 bits and lower 8 bits) Integer value (a ₃ × 2 ⁷ + a ₄ × 2 ⁶ + a ₅ × 2 ⁵ + a ₆ × 2 ⁴ + a ₇ × 2 ³ + a ₈ × 2 ² +)
b ₃ × 2 ¹ + b ₄ × 2 ⁰ ) are stored. Therefore, taking in the value stored at the address of the integer value represented by the target area α ₁ corresponds to extracting the valid bit.

Next, for the second and third bytes (target area β ₁ ) from the top, 5 to 8 of the left (upper) 8 bits
A total of 8 bits of 3 to 6 bits out of the 8 bits and the right (lower) 8 bits are extracted using the editing table B, and the output area 2
To be stored. For this purpose, the second editing table B is used. The editing table B is, as shown in FIG.
It is a one-dimensional array having a size of 2 ¹⁶ corresponding to a byte integer value (0 to 2 ¹⁶ -1). Each array element has valid bits (4 to 8 bits of the upper 8 bits and lower 8 bits) integer comprised of eight bits) of 3-6 bits of ^{_{(b 5 × 2 7 + b}} 6 × 2 6 + b 7 × 2 5 + b 8 × 2 4 + c 3 × 2 3 + c 4 × 2 2 +
c ₅ × 2 ¹ + c ₆ × 2 ⁰ ) is stored. If can capture this reason stored in the address of the integer value representing the target region beta ₁ value is to correspond to the extracted valid bit.

Next, for the third and fourth bytes (target area γ ₁ ) from the beginning, 7 to 8 of the left (upper) 8 bits
A total of 8 bits of 3 to 8 bits out of the 8 bits and the right (lower) 8 bits are extracted using the editing table C, and the output area 3
To be stored. For this purpose, the third editing table C is used. The edit table C is, as shown in FIG.
It is a one-dimensional array having a size of 2 ¹⁶ corresponding to an integer value of a byte (0 to 2 ¹⁶ -1). Each array element has valid bits (7 to 8 bits of the upper 8 bits and lower 8 bits). integer comprised of eight bits) of 3-8 bits of ^{_{(c 7 × 2 7 + c}} 8 × 2 6 + d 3 × 2 5 + d 4 × 2 4 + d 5 × 2 3 + d 6 × 2 2 +
d ₇ × 2 ¹ + d ₈ × 2 ⁰ ) is stored. If can capture this reason stored in the address of the integer value representing the target region gamma ₁ value is to correspond to the extracted valid bit.

The processing from the next fifth byte onward is a repetition of the processing using the tables A, B and C described above.
FIG. 10 shows the flow of this processing. Next, the concept of the expansion process using the expansion edit table will be described with an example in which, when the unit of the effective bits is 6 bits, the effective bits are edited (developed) in the lower 6 bits of the 8-bit area. As described in detail below, there are two types of editing tables in this case. In FIG. 4, reference numeral 20 denotes data in which effective bits in units of 6 bits are tightly packed, and 121 and 122.
Denotes an editing table for development, and 10 denotes data in which the effective bit is edited in the lower 6 bits of the 8-bit area.

First, for the first and second bytes (target area α ₁ ) from the top, 1 to 8 of the left (upper) 8 bits
The edit table D is composed of 6-bit 6-bit data and 6-bit data obtained by adding 1 to 4 bits of the right (lower) 8 bits to 7 to 8 bits of the upper 8 bits. And stored in output areas 1 and 2, respectively.

For this purpose, a first editing table D is used. The edit table D is, as shown in FIG.
It is a two-dimensional array having a size of 2 ¹⁶ × 2 corresponding to a byte integer value (0 to 2 ¹⁶ -1), and each array element has valid bits (1 to 6 bits of the upper 8 bits and upper bits). Integer value (a ₁ × 2 ⁵ + a ₂ × 2 ⁴ + a ₃ ) composed of 7 bits of 8 bits and 2 bits of 6 bits obtained by adding 1 to 4 bits of lower 8 bits to 8 bits of lower 8 bits × 2 ³ + a ₄ × 2 ² + a ₅ × 2 ¹ +
a ₆ × 2 ⁰ and _{^{a 7 × 2 5 + a 8}} × 2 4 + b 1 × b 3 + b 2 × 2 2 + b 3 × 2 1 + b 4
× 2 ⁰⁾ is stored. Therefore, taking in the value stored at the address of the integer value represented by the target area α ₁ is equivalent to storing the effective bit in the lower 6 bits.

Next, for the second byte and the third byte (target area β ₁ ) from the top, 5 to 5 of the left (upper) 8 bits
6-bit data obtained by adding 1 to 2 bits of the right (lower) 8 bits to 8 bits, and 3 of the lower 8 bits
Two 6-bit data of 6-bit data of up to 8 bits are extracted using the editing table E and stored in the output areas 3 and 4, respectively.

For this purpose, the first editing table E is used. Specifically, as shown in FIG.
It is a two-dimensional array having a size of 2 ¹⁶ × 2 corresponding to the integer value of a byte (0 to 2 ¹⁶ -1). Each array element has valid bits (5 to 8 bits out of the upper 8 bits and lower bits to 8 to 8 bits). integer composed of 8 two 6-bit data of 3-8 bits of 6 bits and lower 8 bits plus 1-2 bits of the ^{_{bit) (b 5 × 2 5 +}} b 6 × 2 4 + b 7 _{^{× 2 3 + b 8 × 2}} 2 + c 1 × 2 1 +
c ₂ × 2 ⁰ and _{^{c 3 × c 5 + c 4}} × 2 4 + c 5 × b 3 + c 6 × 2 2 + c 7 × 2 1 + c 8
× 2 ⁰⁾ is stored. If you can capture this reason stored in the address of the integer value representing the target region beta ₁ value is to correspond to storing the valid bit in the lower 6 bits.

The processing from the fourth byte onward is a repetition of the processing using the tables D and E described above. Although the present invention has been described according to the embodiment in which the number of valid bits is 6 bits, the present invention is not limited to this.
Even if the effective bit length is arbitrary other than 6 bits, it is possible to edit various bit data by preparing an editing table corresponding to the effective bit length.

[0017]

As described above, according to the present invention, in data compression (stuffing) and data expansion (expanding effective data in byte units), data manipulation in bit units is performed by using a dedicated edit table. Need not be performed, and the processing efficiency can be improved.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram of a bit data editing process.

FIG. 2 is a system configuration diagram when the present invention is applied.

FIG. 3 is an explanatory diagram of a concept of compression processing using an editing table.

FIG. 4 is an explanatory diagram of a concept of a developing process using an editing table.

FIG. 5 is an explanatory diagram of a compression editing table.

FIG. 6 is an explanatory diagram of an editing table for compression.

FIG. 7 is an explanatory diagram of a compression editing table.

FIG. 8 is an explanatory diagram of a development editing table.

FIG. 9 is an explanatory diagram of a development editing table.

FIG. 10 is an explanatory diagram of a processing flow in the case of compression processing.

[Explanation of symbols]

10 Expanded bit data including invalid (unused) data 20 Compressed bit data composed only of valid data 100 Program for editing bit data 110 Editing table for compression 120 Editing table for expansion

Claims (2)

[Claims] An association between data of a first length and data of a second length which is longer than the first length and is obtained by adding unused data to the data of the first length. In the data processing device to be managed, a conversion table storing a relationship between a data position to be converted and a data position of a conversion destination; a first length data and a second length data based on the conversion table. And a conversion means for performing conversion processing between the data processing apparatus and the data processing apparatus. 2. An association between data of a first length and data of a length longer than the first length, wherein unused data is added to the data of the first length. A computer-readable storage medium storing a program for management, comprising: a conversion table storing a relationship between a data position to be converted and a data position to be converted; A computer-readable storage medium storing a program for causing a conversion unit to perform a conversion process between data of a second length and data of a second length.