JPH08107360A - Data storage device and data reader - Google Patents

Abstract

PURPOSE: To store compressed data by a compression algorithm in which a time for expansion is small and the data length is reduced simultaneously. CONSTITUTION: After data are compressed at a 1st compression means 11a and a 2nd compression means 11b with different algorithm, expanded by a 1st expansion means 11c and a 2nd expansion means 11d, a time for expansion is measured by an expansion time measurement means 13, a data length after each compression is counted by a data length count means 12 and a data selection means decides data compressed by either one of the sets of algorithm to be stored in a data storage means 14 based on both the expansion time and the data length.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data storage device for storing a plurality of data and a data reading device.

[0002]

2. Description of the Related Art In recent years, with the advent of so-called multimedia equipment, it has become frequent to store a large amount of data such as image data and audio data in a data storage device having a large-capacity storage medium such as an optical disk. Are becoming available.

Conventionally, when storing such a large amount of data in a storage medium, so-called encoding of individual data in a completely recoverable form of the original data such as run-length encoding or entropy encoding is called. Data compression technology by lossless encoding is used, or JPEG (Joint Ph
otographic coding Experts Group) is used to reduce the data length by intentionally deleting information that is difficult to be recognized due to the human five senses to shorten the data length. By doing so, effective use of the storage medium is achieved.

Further, as a technique that takes this a step further, which one of a plurality of compression algorithms is used to obtain the shortest data length depends on the content of each data. Rather than applying a single compression algorithm to data, a technique has been proposed in which multiple compression algorithms are tried for each data and the one with the shortest data length is selected and stored (for example, , JP-A-2-228769).

[0005]

However, the data storage device is required not only to effectively use the storage medium but also to be able to read data at high speed. In particular, when compressing and storing data, the reading time must be considered including the time required for decompressing the data.

However, the request to shorten the data length after compression and the request to decompress the data at a high speed are not always compatible with each other, but rather often conflict with each other.

For example, MH (Modified Huffman) compression and MR used for international standard group 3 facsimile
Taking (Modified Relative addressing) compression as an example and comparing both with respect to a large number of data, MH compression, which has a simple algorithm, can compress and decompress data at a relatively high speed. The data length does not become very short. On the other hand, MR compression, which has a relatively complex algorithm, takes longer time to compress and decompress data than MH compression, but the data length of generated data is short. Tends to become.

As described above, since there is a tendency for the demand for shortening the length of data after compression and the demand for performing data decompression at a high speed to conflict with each other, as in the prior art, the data after compression must be compressed. If the emphasis is placed only on shortening the length, there is a problem in that the time required for decompressing the data increases and the data read speed including this time decreases.

This problem becomes particularly serious in an environment in which it is required to read a plurality of data continuously at fixed time intervals.

For example, an image filing device that stores image data and continuously displays the stored images in a page-turning manner at regular time intervals (see, for example, Japanese Patent Laid-Open No. 57-1).
No. 08961), the reading speed of data including the decompression time is significantly reduced for some data, and the time interval from the display of one image to the display of the next image is inconsistent. When stable, it becomes difficult for the user to view the display screen with peace of mind. Conventionally, as a technique for keeping the image display time interval constant even if such a situation occurs, if a large amount of time is spent for decompressing data and the entire image for one page cannot be displayed within a predetermined time, A technique has been proposed in which the display of the image is stopped halfway and the display of the next image is started (see, for example, Japanese Patent Laid-Open No. 4-37957). However, in this technique, in exchange for making the display time interval of the image constant,
For each image, the warranty that the entire image is displayed is disclaimed, so for example, the characteristic illustrations or charts that are the key when the user searches for images are near the bottom of the page. If it exists in a portion to be displayed later, that portion may end without being displayed, which causes a new problem that it becomes difficult to search for a desired image.

A similar problem occurs when storing and reproducing acoustic data such as music and voice. Read the acoustic data at a predetermined sampling cycle, compress and store it, read them one after another at the time of playback, decompress them,
Let's assume an acoustic memory playback device that sounds a speaker or the like.
In this case, letting down the decrease in the data reading speed including the decompression time, the time required for data reading and decompression will be longer than the sampling period used when storing the acoustic data. In that case, it becomes impossible to correctly reproduce the stored acoustic data with the speaker.

The present invention has been made in view of the above problems, and guarantees that the expansion of data is completed within a preset allowable time, and at the same time compresses the data with a compression algorithm that shortens the data length. It is an object of the present invention to provide a data storage device that can be stored as a device and a device that reads data.

[0013]

In order to achieve the above-mentioned object, the present invention has the following constitution.

That is, the data storage device comprises compression means for compressing data by a plurality of different compression algorithms,
Decompression means for decompressing the data compressed by each compression algorithm, decompression time measuring means for measuring the time required for decompressing the compressed data, data length counting means for counting the data length of the compressed data, and decompression A data selecting means for selecting one of the compressed data based on the measurement result of the time measuring means and the count result of the data length counting means, and a data storage means for storing the data selected by the data selecting means. It was configured.

Further, the data selecting means has a comparing means for comparing the time measured by the expanding time measuring means with an upper limit value allowable as the expanding time, and the comparison result of the comparing means and the counting result of the data length counting means. The data is selected based on the above.

Further, the data selecting means selects the data having the shortest data length counted by the data length counting means from the data whose time measured by the decompression time measuring means is smaller than the upper limit value allowable as the decompression time. It was configured.

Further, the data selecting means is configured to cause the decompressing means to decompress only the compressed data having a shorter data length than the data before compression.

Further, the data selecting means is configured to select the data which is first detected that the time measured by the expansion time measuring means is smaller than the upper limit value allowable as the expansion time.

Further, the data selecting means is constituted so that the compressed data is expanded by the expanding means in order from the one having the shortest data length.

Further, when the time measured by the expansion time measuring means exceeds the upper limit value allowable as the expansion time,
The configuration is such that the expansion of the data by the expansion means is terminated.

Further, the data selecting means is configured to select data which is not compressed when any of the compressed data is larger than the upper limit value of the expansion time which is measured by the expansion time measuring means.

Further, the data selection means, when any of the compressed data has a data length counted by the data length counting means greater than the data length of the data before compression,
The configuration is such that data that is not compressed is selected.

Further, the image input means for optically reading the image and photoelectrically converting the generated image data to the compression means is provided.

Further, the sound input means for compressing the sound signal generated by converting the sound into the electric signal by the compression means is provided.

In addition, the data reading device individually confirms that each data can be expanded within a predetermined time among a plurality of different compression algorithms and compresses the data compressed by the selected algorithm. And a reading means for reading the compressed data and the corresponding identifier from the data storing means, and a decompressing means for decompressing the data read by the reading means according to the read identifier. The conversion unit converts the data decompressed by the decompression unit into information that can be recognized by the human senses at predetermined time intervals.

The data stored in the data storage means includes image data, and the conversion means has a display means for displaying an image.

Further, the display means is configured to display the image as a moving image. Further, the data stored in the data storage means includes acoustic data, and the conversion means has a sound generation means for generating sound.

[0028]

The above-described structure provides the following functions.

That is, the data storage device compresses the data by the compression means by a plurality of different compression algorithms, decompresses the data compressed by each compression algorithm by the decompression means, and increases the time required for decompressing the compressed data. The data length of the compressed data measured by the expansion time measuring means is counted by the data length counting means, and the data selecting means determines which compression algorithm is used to store the compressed data in the data storage means. Since the determination is made based on not only the counting result of the counting means but also the measurement result of the decompression time measuring means, the data length is not stored in the storage means, but the data length is not simply stored in the storage means by the compression algorithm that minimizes the data length. And the time required to decompress the data are both selected and stored in the data storage means. It becomes possible.

Further, the comparison means compares the time measured by the extension time measuring means with an upper limit value which is allowable as the extension time, and the data selecting means sets the data based on the comparison result and the counting result of the data length counting means. By selecting, it becomes possible to guarantee that the data stored in the data storage means can be expanded within the allowable time.

Further, the data selecting means selects the data having the shortest data length counted by the data length counting means, out of the data whose time measured by the expansion time measuring means is smaller than the upper limit value allowable as the expansion time. So
Of the compression algorithms that can be expanded within the allowed time,
It becomes possible to store the data compressed by the algorithm with the shortest data length in the data storage means and effectively use the storage medium.

Further, since the data selecting means causes the decompressing means to decompress only the compressed data having a shorter data length than the data before compression, decompression is performed on all the compressed data. It is possible to determine which algorithm compresses the data to be stored in the data storage unit faster than the case of performing.

Further, the data selecting means selects the data which is first detected that the time measured by the expansion time measuring means is smaller than the upper limit value allowable as the expansion time. It becomes possible to determine at a higher speed whether or not to store in the storage means.

Further, since the data selecting means causes the decompressing means to decompress the compressed data in order from the one having the shortest data length, the time measured by the decompressing time measuring means may be smaller than the upper limit value allowable as the decompressing time. It is possible to guarantee that the data detected first has the shortest data length after compression.

Further, when the time measured by the extension time measuring means exceeds the upper limit value which is allowable as the extension time,
Since the decompression of the data by the decompression means is ended and the decompression of the new compressed data is started, it is faster than the completion of the decompression of the previous data and the decompression of the new compressed data. It becomes possible to decide whether to store the data compressed by the algorithm in the data storage means.

Further, the data selecting means stores the uncompressed data which does not require decompression when the time measured by the decompression time measuring means is larger than the upper limit value which is acceptable as the decompression time, for all the compressed data. Since it is selected as the data to be stored in, it is possible to guarantee that the expansion is completed within the allowable time for any data.

Further, the data selection means, when any of the compressed data has a data length counted by the data length counting means greater than the data length of the data before compression,
Since the data that is not compressed is selected, it is possible to prevent the data length from becoming rather large due to the compression.

Further, since the image is optically read by the image input means and the image data generated by photoelectric conversion is compressed by the compression means, the optically read image can be stored in the data storage means.

Further, since the sound signal obtained by converting the sound into the electric signal by the sound input means is compressed by the compression means, the sound can be stored in the data storage device.

In addition, the data reading device individually confirms that each of the data can be expanded within a predetermined time among a plurality of different compression algorithms and compresses the data compressed by the selected algorithm and the expansion method of the data. Is read from the data storage means by the reading means, the read data is expanded by the expansion means according to the identifier, the expanded data is converted into information that can be recognized by the human five senses, and at predetermined time intervals. It can be presented to humans.

Further, the data stored in the data storage means includes image data, and by displaying the image by the display means in the conversion means, the data stored in the data storage means is presented to a human as a continuous image. Can be presented.

Further, it becomes possible to display a continuous image as a moving image by the display means.

Further, the data stored in the data storage means includes acoustic data, and by the sound generation means generating a sound, continuous information can be presented to a human as a sound.

[0044]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings.

(Embodiment 1) FIG. 1 is a block diagram showing the configuration of a first embodiment of a data storage device of the present invention.

In FIG. 1, 10 is a temporary storage means realized by a hard disk. Further, 11 is a compression / expansion means constituted by using a data compression / expansion IC (integrated circuit) for facsimile communication, and has a first compression means 11a, a second compression means 11b and a first expansion means 11 therein.
c and the second expansion means 11d. Further, 12 is a data length counting means, and 13 is an extension time measuring means realized by a crystal oscillator and a frequency divider. Also, 14 is a CPU
(Central processing unit), memory and data selecting means realized by software. Further, reference numeral 15 is a data storage means having a large capacity optical disc built therein as a storage medium. Further, 19 is an image inputting means, which has a scanner 19a with an automatic feeder 19b therein. These respective means are connected by signal lines 510 to 519.

The operation of each means shown in FIG. 1 will be described below. The operation of the data storage device is started by setting a paper on which an image is drawn on the auto feeder 19b of the scanner 19a provided in the image input means 19, and thereafter, the data storage operation is automatically performed. .

The image input means 19 is an automatic feeder 19
When the paper is set on b, the image of one screen drawn on the paper is optically read by the scanner 19a, photoelectrically converted, and then the signal line 510 is formed as the image data consisting of the binary of the white pixel and the black pixel. Output to.

The temporary storage means 10 reads the image data supplied via the signal line 510, and the uncompressed data 10 is read.
Store as c.

The first compression means 11 in the compression / expansion means 11
a is the uncompressed data 10 stored in the temporary storage means 10.
c is read through the signal line 511, compressed by the MH compression algorithm, and the compressed result is written as the first compressed data 10a in the temporary storage means 10 through the signal line 512.

After that, the second compression means 11b converts the non-compressed data 10c stored in the temporary storage means 10 into the signal line 5.
The data is read via 11, compressed by the MR compression algorithm, and the compressed result is written as the second compressed data 10b in the temporary storage means 10 via the signal line 512.

The data length counting means 12 includes the first compressed data 10a, the second compressed data 10b and the non-compressed data 10.
For each c, the signal line 51 from the temporary storage means 10
3, the number of bytes in the data length is calculated, and the calculation result is the data length of the first compressed data 10a, the data length of the second compressed data 10b, and the data length of the non-compressed data 10c. In that order, to the signal line 514.

After that, the first decompression means 11c in the compression / decompression means 11 reads the first compressed data 10a stored in the temporary storage means 10 through the signal line 511, and MH-expands the read data. Decompress using an algorithm. The data generated as a result of decompression is not necessary here and is therefore discarded. A first decompression start signal is output to the signal line 515 when decompression of the first compressed data 10a is started, and a first decompression end signal is output to the signal line 515 when decompression is completed.

The second expanding means 11d is a temporary storage means 1
The second compressed data 10b stored in the signal line 51
1 is read and the read data is expanded using the MR expansion algorithm. The data generated as a result of decompression is not necessary here and is therefore discarded. A second decompression start signal is output to the signal line 515 when decompression of the second compressed data 10b is started, and a second decompression end signal is output to the signal line 515 when decompression is completed.

The expansion time measuring means 13 measures the time elapsed from the supply of the first expansion start signal through the signal line 515 to the supply of the first expansion end signal in milliseconds (thousandths). The measurement result is output to the signal line 516 as the first extension time. For example, when the time elapsed from the supply of the first expansion start signal to the supply of the first expansion end signal is 170 milliseconds,
The value output to the signal line 516 is 170. Similarly, the time elapsed from the supply of the second decompression start signal through the signal line 515 to the supply of the second decompression end signal is measured in milliseconds, and the measurement result is Is output to the signal line 516 as the extension time.

The data selecting means 14 holds a predetermined calculation formula as an evaluation function. This evaluation function is, for example, the product of the decompression time value measured in milliseconds and the data length for the compressed data 10a and 10b, and 100 for the uncompressed data 10c. Set as double. This evaluation function can be freely set and changed by the user by connecting a keyboard to the data selection means 14 as appropriate.

The data selecting means 14 has the data length of the first compressed data 10a, the data length of the second compressed data 10b and the non-compressed data 10c supplied via the signal line 514.
Of the first compressed data 10a, the second compressed data 10b, and the non-compressed data 10c by using the first data length and the second data length supplied via the signal line 516. Then, the value of the evaluation function is calculated, and the calculated values of the three evaluation functions are compared to select the one having the smallest value. If the selected value is calculated for the first compressed data 10a, the temporary storage means 1
The first compressed data 10a stored in 0 is read via the signal line 517, output to the signal line 518 as it is, and the value 1 is output to the signal line 519 as an identifier indicating the first decompression means 11c. On the other hand, when the selected value is calculated for the second compressed data 10b, the second compressed data 10b stored in the temporary storage means 10 is read via the signal line 517. Then, it outputs to the signal line 518 as it is, and outputs the value 2 to the signal line 519 as an identifier indicating the second decompression means 11d. If the selected value is calculated for the non-compressed data 10c, the non-compressed data 10c stored in the temporary storage means 10 is read via the signal line 517 and output to the signal line 518 as it is. Then, the value 3 is output to the signal line 519 as an identifier indicating that the data need not be expanded.

The data storage means 15 stores the data supplied via the signal line 518 and the identifier supplied via the signal line 519 in association with each other.

After that, the temporary storage means 10 erases the first compressed data 10a, the second compressed data 10b and the non-compressed data 10c.

Here, attention is focused on one image data, and the process from the reading of the image data by the image input means 19 to the storage of the data in the data storage means 15 has been described step by step. A plurality of images are sequentially read by the image input means 19, and the data storage means 15 is read.
Can be stored.

According to the present embodiment, which algorithm is used to compress the data supplied via the signal line 510 to be stored in the data storage means 15 is determined based on only the compressed data length. Since it can be decided by considering both the data length after compression and the time required to decompress compressed data, the data is compressed using an algorithm that requires a huge amount of time to decompress the data in order to shorten the data length. It is possible to eliminate the risk of doing this in advance.

Further, since the user can arbitrarily set the evaluation function of the data selection means 14, for example, when the data storage means 15 has sufficient storage capacity and high-speed data reading and decompression are required, The compression algorithm is determined with an emphasis on shortening the time required for decompression, and when the storage capacity of the data storage unit 15 does not have a sufficient margin and a high speed is not required for reading and decompressing data, the data after compression is compressed. Decide the compression algorithm with an emphasis on the time required for decompression, such as deciding the compression algorithm with an emphasis on shortening the data length, or determine the compression algorithm with an emphasis on reducing the data length after compression. It is also possible for the user to select as appropriate.

The present data storage device may be incorporated as a part of a so-called image filing device, or it may be realized as an independent device in which the storage medium of the data storage means 15 is detachable and the data may be read out. Alternatively, a read-only device may be separately provided, and the device may read data from the storage medium.

(Embodiment 2) FIG. 2 is a combination of the data storage device and the data reading device of the present invention.
It is a block diagram which shows the structure of the Example of an image filing apparatus.

In FIG. 2, reference numeral 22 is a data selecting means realized by a CPU (central processing unit), a memory and software. Reference numeral 22a is a data length memory provided in the memory included in the data selecting means 22. Also,
Reference numeral 22b is a counter provided in the data selection means 22. Further, 22c is a comparison means included in the data selection means 22. Reference numeral 22d is a register included in the comparison data selection means 22. Further, 23 is a read instruction means,
Reference numeral 24 is a timer, 25 is a data reading means, 26 to 27 are image memories, and 28 is a display means. Further, reference numeral 29 is an image input means, which internally has a scanner 29a and a display means 29b.
And an image processing unit 29c including an image processing processor for a general-purpose computer. Reference numeral 30 is a keyboard. Further, 31 is a compression / expansion means constituted by using a data compression / expansion IC for facsimile communication, and internally has a first compression means 11a, a second compression means 11b, a first expansion means 11c and a second compression means. It has a stretching means 11d. Reference numerals 520 to 530 are signal lines. In addition, the same components as those in FIG. 1 are denoted by the same numbers as those in FIG.

Further, (Table 1) A to D show examples of values stored in the data length memory 22a. In Table 1, A1,
A2 and A3 are addresses, the address A1 is the data length of the first compressed data 10a, and the address A2 is the second data length.
The data length of the compressed data 10b and the data length of the uncompressed data 10c are stored at the address A3.

[0067]

[Table 1]

Further, (Table 2) A to D are the first expansion means 11c and the second expansion means 11c measured by the expansion time measuring means 13.
An example of the data decompression time of the decompression means 11d is shown.

[0069]

[Table 2]

Table 3 shows an example of values stored in the register 22d.

[0071]

[Table 3]

Further, (Table 4) A to D are comparison means 22c.
An example of the comparison result output by is shown.

[0073]

[Table 4]

The operation of the image filing apparatus shown in FIG. 2 will be described below separately for storing image data and reading image data.

First, the operation of storing image data will be described. To store the image data, the user sets a sheet of paper on which an image is drawn on the scanner 29a and presses the keyboard 30.
It is started by operating to instruct to read the image.

The keyboard 30 outputs a character code string representing a user's instruction to the signal line 530.

The image input means 29 discriminates the character code string supplied through the signal line 530, and when instructed to read the image, the image drawn on the paper is optically read by the scanner 29a and photoelectrically read. After being converted, it is displayed on the display unit 29b as image data composed of binary values of white pixels and black pixels. The display means 29b may be realized by a CRT (cathode ray tube) or may be realized by using another display element such as a liquid crystal or a plasma display.

When the user looks at the image displayed on the display means 29b and wants to process the image such as cutting out or rotating the image before storing the image,
An instruction for performing the processing is input from the keyboard 30.

Based on this instruction, the image input means 29 processes the instructed image using the image processing means 29c, and displays the processed image again on the display means 29b.

The user repeats the same operation until an acceptable image is obtained and processes the image.

After a satisfactory image is obtained, when the user gives an instruction to store the image using the keyboard 30, the image data storage operation is performed as follows.

First, the image input means 29 is the display means 29.
The image data displayed in b is output to the signal line 510.

The temporary storage means 10 reads the image data supplied via the signal line 510, and the uncompressed data 10 is read.
Store as c.

The first compression means 11 in the compression / expansion means 31.
a and the second compression means 11b compress the uncompressed data 10c using the MH compression algorithm and the MR compression algorithm in the same manner as the first compression means 11a and the second compression means 11b in FIG. The compression result is the first compressed data 10a and the second compressed data 1
It is written in the temporary storage means 10 as 0b.

The data length counting means 12 is similar to the data length counting means 12 of FIG.
The data length of a, the data length of the second compressed data 10b, and the data length of the non-compressed data 10c are calculated and output to the signal line 514. In the following, the data length of the first compressed data 10a is 10,000, the data length of the second compressed data 10b is 7,000, and the data length of the non-compressed data 10c is 20,000.
The case will be described as an example.

The value of the counter 22b in the data selecting means 22 is initialized to 0. When the data length of the first compressed data 10a is supplied via the signal line 514, the data selection means 22 stores the supplied value in the data length memory 22.
a to the address A1 of the counter a and the counter 22b
Increase the value of by one. When the data length of the second compressed data 10b is supplied via the signal line 514,
The supplied value is written to the address A2 of the data length memory 22a and the value of the counter 22b is incremented by 1. In addition, the uncompressed data 10c is transmitted via the signal line 514.
When the data length is supplied, the supplied value is written in the address A3 of the data length memory 22a. For example, in the example defined above, the data length of the first compressed data 10a is 1
0000, the data length of the second compressed data 10b is 700
0, the data length of the non-compressed data 10c is 20000. Therefore, as shown in (Table 1) A, the data length memory 22a
10000 for address A1 and 7 for address A2
000, and 20000 is written in the address A3. At this time, the value of the counter 22b is set to the value 2 which is equal to the number of the compression means built in the compression / expansion means 31.

After that, the first expanding means 11c and the second expanding means 11d in the compressing and expanding means 31 are operated in the same manner as the first expanding means 11c and the second expanding means 11d in FIG. The first compressed data 10a and the second compressed data 10b are expanded.

The decompression time measuring means 13 measures the time required for the data decompression by the first decompressing means 11c and the second decompressing means 11d in the same manner as the decompressing time measuring means 13 of FIG. The signal is output to the signal line 516 as the first extension time and the second extension time. In the following, the first extension time is 170 milliseconds and the second extension time is 250 milliseconds, as an example. As shown in (Table 2) A, the first output time is output to the signal line 516. The description will be made assuming that the value of the extension time of 1 is 170 and the value of the second extension time is 250.

The register 22d is used for the extension time measuring means 13
Stores the upper limit value that is allowable as the data decompression time measured by, and the stored value is always output to the signal line 520. In the following, the case where the upper limit value of this time is 200 milliseconds is taken as an example, and the value stored in the register 22d is 200 as shown in (Table 3).

When the first expansion time is supplied via the signal line 516, the comparison means 22c reads the first expansion time and the value supplied via the signal line 520, and compares the two values. As a result of the comparison, if the first expansion time is smaller than the value supplied via the signal line 520, the first
As a comparison result of, the value "true value" representing one of the alternatives is output. On the other hand, if the first decompression time is greater than the value supplied via signal line 520 or they are equal,
As a first comparison result, a value "false value" that is one of the alternatives and represents a different one from the "true value" is output. Also,
When the second extension time is supplied via the signal line 516, the second extension time and the value supplied via the signal line 520 are read, and the two are compared. The result of the comparison, the second
If the expansion time of is smaller than the value supplied via the signal line 520, the “true value” is output as the second comparison result. On the other hand, if the second expansion time is larger than the value supplied via the signal line 520 or they are equal to each other, a “false value” is output as the second comparison result. For example, the first
And the second extension time is the value of A (Table 2),
When the value stored in the register 22d and output to the signal line 520 is the value of (Table 3), as shown in (Table 4) A, the “true value” is the first comparison result. A "false value" is output as the second comparison result.

When the comparison means 22c outputs the first comparison result, the data selection means 22 reads the value. When the read first comparison result is a "false value", the value of the address A1 of the data length memory 22a is set to the special value "INVALI" indicating that the data length is invalid.
D "and the value of the counter 22b is decreased by 1. On the other hand, when the first comparison result is the “true value”, the value of the data length memory 22a is not changed at all,
The value of the counter 22b is decreased by 1. In addition, when the second comparison result is output from the comparison means 22c, the value is read. When the read second comparison result is a "false value", the value of the address A2 of the data length memory 22a is rewritten to "INVALID" and the value of the counter 22b is decreased by 1. On the other hand, when the second comparison result is the "true value", the value of the data length memory 22a is not changed at all and the value of the counter 22b is decreased by 1. For example, when the first comparison result and the second comparison result of (Table 4) A are supplied while the data length memory 22a stores the value of (Table 1) A, (Table 1) ) B
As shown in, the value of the address A2 of the data length memory 22a is rewritten to "INVALID" to change the address A2.
The values of 1 and A3 cannot be rewritten. Further, at this time, the value of the counter 22b is decreased by 2 in total and becomes 0 as a result.

When the value of the counter 22b becomes 0 due to a decrease, the data selecting means 22 has a data length memory 22a.
Of the addresses A1 to A3 of the
Select the address with the smallest stored value, except for those with "NVALID". The selected address is A1
If it is, the first compressed data 10a stored in the temporary storage means 10 is read via the signal line 517 and output to the signal line 518 as it is.
The value 1 is output to the signal line 519 as an identifier indicating 1c. On the other hand, when the selected address is A2, the second compressed data 1 stored in the temporary storage means 10 is stored.
0b is read through the signal line 517, output to the signal line 518 as it is, and the value 2 is output to the signal line 519 as an identifier indicating the second expansion unit 11d. When the selected address is A3, it is not necessary to read the uncompressed data 10c stored in the temporary storage means 10 via the signal line 517 and output it to the signal line 518 as it is to expand the data. Value 3 as an identifier indicating
Output to 9.

For example, in the data length memory 22a (Table 1)
When the value of B is stored, the address A1 is selected, the first compressed data 10a is output to the signal line 518, and the identifier value 1 is output to the signal line 519.

As another example, when the first extension time and the second extension time output to the signal line 516 by the extension time measuring means 13 are 220 and 190 as shown in (Table 2) B, Assuming that, in this case, the first comparison result and the second comparison result output by the comparing means 22c are, as shown in (Table 4) B, “false value” and “true value”, respectively. Therefore, at the beginning (Table 1) A
In the data length memory 22a, in which the value of “1” was stored, the value of the address A1 was rewritten to “INVALID” (Table 1), resulting in C as shown in FIG.
The compressed data 10b is output, and the identifier value 2 is output to the signal line 519.

As still another example, the extension time measuring means 1
The first decompression time and the second decompression time output to the signal line 516 by 3 are 170 as shown in (Table 2) C.
, And 190, in this case, the first comparison result and the second comparison result output by the comparison means 22c are both "true values" as shown in Table 4C. Therefore, the data length memory 22a in which the value of A is initially stored (Table 1) is not rewritten, and as a result, the second compressed data 10 is stored in the signal line 518.
b is output, and the identifier value 2 is output to the signal line 519.

As still another example, the extension time measuring means 1
The first expansion time and the second expansion time output to the signal line 516 by 3 are 220 as shown in (Table 2) D.
, And 250, the first comparison result and the second comparison result output by the comparison means 22c are both "false values" as shown in Table 4D. Therefore, in the data length memory 22a in which the value of A is initially stored (Table 1), the values of the addresses A1 and A2 are both rewritten as "INVALID" to become (Table 1) D, and as a result, The uncompressed data 10c is output to the signal line 518, and the identifier value 3 is output to the signal line 519.

As described above, the selecting means 22 first specifies the decompression time by the register 22d from the first compressed data 10a and the second compressed data 10b generated based on the non-compressed data 10c. Those that exceed the predetermined time and those that are equal to the predetermined time are excluded, and the shortest data length is selected from the remaining compressed data and non-compressed data 10c and output to the signal line 518. An identifier indicating a decompression means for decompressing the data is output to.

The data storage means 15 stores the data supplied via the signal line 518 and the identifier supplied via the signal line 519 in association with each other.

After that, the temporary storage means 10 erases the first compressed data 10a, the second compressed data 10b and the non-compressed data 10c.

Here, one image data is focused on, and the process from the reading of the image data by the image input means 29 to the storage of the data in the data storage means 15 has been described step by step. A plurality of images are sequentially read by the image input means 29, and the data storage means 15
Can be stored.

As described above, the data storage means 15 is compressed by the algorithm which guarantees that the plurality of image data read by the image input means 29 can be expanded in a time shorter than the time indicated by the value of the register 22d. Is
Alternatively, it is stored without being compressed.

Next, the operation at the time of reading image data will be described. The operation of reading the image data is started by the user instructing the reading of the data by the keyboard 30, and thereafter, the user operates the keyboard 30.
It continues until a predetermined key is pressed to instruct to stop reading.

The keyboard 30 outputs a character code string representing a user's instruction to the signal line 530.

The read instructing means 23 discriminates the character code string supplied through the signal line 530 and outputs a read start instruction to the signal line 522 when the image read is instructed. On the other hand, if the character code string supplied via the signal line 530 is a read stop instruction, the signal line 522
A read stop command is output to.

The timer 24 repeats the data read command to the signal line 523 periodically at predetermined time intervals during the period from when the read start command is supplied through the signal line 522 until the read stop command is supplied. Output. The case where the predetermined time interval is 250 milliseconds will be described below as an example.

Each time a data read command is supplied via the signal line 523, the data read means 25 first sets the set of the data stored in the data storage means 15 and the identifier stored corresponding to the set. Only one set from the stored one is sequentially read through the signal line 524, and the set of the data and the identifier is directly output to the signal line 525. That is, the data storage means 1 is first connected to the signal line 525.
A pair of identifiers corresponding to the data stored in 5 is output, and 250 milliseconds after that, 2 is stored in the data storage means 15.
The stored pair of data and identifier will be output. In the following, a case will be described as an example in which it is possible to guarantee that the reading of the set of data and the identifier from the data storage unit 15 by the data reading unit 25 is completed within 45 milliseconds.

Image memory 26 and image memory 27
Stores white pixels, which are white when data is displayed on the display means 28, in the entire storage area as an initial value.

The compression / expansion means 31 reads the set of data and the identifier supplied via the signal line 525 and performs the processing according to the value of the identifier. That is, when the value of the identifier is 1, the data is decompressed by the first decompression means 11c and the result is signal line 526 or signal line 527.
Image memory 26 or image memory 2
Write in any of 7. The method of deciding which to write will be described later. On the other hand, when the value of the identifier is 2, the data is decompressed by the second decompression means 11d and the result is either the image memory 26 or the image memory 27 via either the signal line 526 or the signal line 527. Write in the crab. A method of determining which signal line to output will be described later. When the value of the identifier is 3, the read data is directly used as the signal line 526 or the signal line 52.
Writing to either the image memory 26 or the image memory 27 via any one of 7 Whether to write the data into the image memory 26 or the image memory 27 does not depend on the contents of the read data and identifier set, but is based on only the number of times the data and identifier set is read from the signal line 525. If it is determined and an odd-numbered data / identifier pair is read, it is written in the image memory 26, and if an even-numbered data / identifier pair is read, the image memory 2
Write to 7. Therefore, the data read first by the data reading means 25 is written in the image memory 26, and the data read second is written in the image memory 27.

Here, it is assumed that the decompression result of the decompression means 31 is written into the image memories 26 and 27 sufficiently faster than the data decompression time, and the time required for this is 5 milliseconds at the maximum. The following explanation will be given.

Further, as already described in the description of the data storage operation, the data stored in the data storage means 15 is compressed and expanded even if it is compressed.
Guarantees that it can be extended in less than 200 milliseconds.

At this time, the data reading time by the data reading means 25 from the data storage means 15 is 45 at maximum.
The maximum value of the data expansion time by the compression / expansion means 31 is 200 ms, and the image memory 26, 2
Since the writing time to 7 is up to 5 milliseconds, the data read out from the data storage means 15 will be stored in the image memory after 250 milliseconds at the maximum after the timer 24 outputs the data read command to the signal line 523. 26 or image memory 2
7 will be stored.

On the other hand, since the period in which the timer 24 outputs the data read command to the signal line 523 is 250 milliseconds, the data read command is once supplied to the signal line 523 until the next data read command is supplied. During the period, the image data read from the data storage means 15 is decompressed by the compression / decompression means 31, and then the writing to the image memory 26 or the image memory 27 is completed.

When a data read command is supplied via the signal line 523, the display means 28 outputs a signal from the image memory 27 via the signal line 529 if the supplied data read command is an odd number, and a signal if the data read command is an even number. The written data is read as image data from the image memory 26 via the line 528 and displayed. Therefore, when the first data read command is supplied through the signal line 523, the initial value stored in the image memory 27, that is, the whole screen white pixel is displayed, and thereafter, the n-th (n ≧ 2) data is displayed. When the read command is supplied, the image data read out from the data storage means 15 at the (n-1) th position is displayed.

From the timer 24, a data read command is issued 25
Since the signal is periodically supplied to the signal line 523 at 0 millisecond intervals, the image filing apparatus as a whole has a page-turning mode in which the image data stored in the data storage unit 15 is sequentially stored from the first stored one. , Will be periodically displayed on the display means 28 at a speed of 4 sheets per second.

Here, the reading of the set of data and the identifier from the data storage means 15 is completed within 45 milliseconds,
The case where the writing of the data to the image memories 26 and 27 is completed within 5 milliseconds has been described as an example. However, if these times are different, it is possible to change these two from the time when one image is displayed in the page turning mode. The value obtained by subtracting the total of the two times is set as the upper limit of the data expansion time in the register 22d.
You can store it in.

According to the present embodiment, when compressing and storing the data in the data storage means 15, a compression algorithm that guarantees that the data can be expanded within a predetermined time is used. Even if the data is continuously read at a predetermined time interval, when the reading of new data is started, the previously read data is expanded and the image memory 2 is read.
Writing to pages 6 and 27 has been completed, and it becomes possible to continuously display image data in a page-turning manner at a stable speed.

Further, since it is possible to guarantee that the entire image is displayed for each image without being partially lost, the user can display the image in comparison with the technique disclosed in Japanese Patent Laid-Open No. 4-37957. Even if the characteristic illustration or chart that is the key for searching exists in the most recently displayed part such as near the bottom of the page, that part will not end without being displayed and the desired It is possible to solve the problem that it becomes difficult to search for images.

In the data storage means 15, of the data compressed by the algorithm satisfying the condition that the data can be expanded within a predetermined time and the data before compression,
Since the one having the shortest data length is selected and stored, it is possible to reduce the usage amount of the storage medium.

(Embodiment 3) FIG. 3 is a combination of the data storage device and the data reading device of the present invention.
It is a block diagram which shows the structure of the Example of an audio memory reproduction | regeneration apparatus.

In FIG. 3, reference numeral 71 is a compression / expansion means, which has a first compression means 11a and a second compression means 11 inside.
It has b, the 1st expansion means 11c, and the 2nd expansion means 11d. Reference numeral 72 is a data selecting means realized by a CPU (central processing unit), a memory and software. Further, 72a is a data length memory provided in the memory included in the data selecting means 72. Further, 72c is a comparison means included in the data selection means 72. Further, 72d is a register included in the data selecting means 72. Also,
Reference numeral 73 is an extension time measuring means realized by a crystal oscillator and a frequency divider. Reference numerals 76 to 77 are acoustic memories. Further, reference numeral 78 is a sound generating means, and the D / A converter 7 is provided inside.
8a, an amplifier 78b and a speaker 78c. Reference numeral 79 is a sound input means, which has a microphone 79a, a D / A converter 79b, and a memory 79c inside. Further, 570 to 574 and 576 to 579 are signal lines.
In addition, the same components as those in FIG. 2 are denoted by the same numbers as those in FIG.

Further, (Table 5) shows an example of the comparison result output by the comparing means 72d.

[0122]

[Table 5]

The operation of this acoustic memory reproducing apparatus will be described below.
The description will be made separately for the storage of acoustic data and the reproduction of acoustic data.

First, the operation of storing acoustic data will be described. The operation of storing the acoustic data is started by the user operating the keyboard 30 to instruct the reading of the acoustic signal and inputting a voice into the microphone 79a.

The keyboard 30 outputs a character code string representing a user's instruction to the signal line 530.

The sound input means 79 discriminates the character code string supplied through the signal line 530, and when the reading of the sound is instructed, the sound is converted into a sound signal which is an electric signal by the microphone 79a, and A / D converter 7 for acoustic signals
After digitized by 9b, the digitized acoustic data is written in the memory 79c.

The user operates the keyboard 30 again,
When the end of reading the audio signal is instructed, the audio input means 7
The microphone 9 finishes reading the acoustic data from the microphone 79a to the memory 79c.

After that, the sound inputting means 79 is operated by the memory 79.
The acoustic data stored in c is divided every 250 milliseconds of acoustic signals, and the divided acoustic data is output to the signal line 510 one by one from the front.

The temporary storage means 10 reads the acoustic data supplied via the signal line 510, and the uncompressed data 10 is read.
Store as c.

The first compression means 11 in the compression / expansion means 71.
a and the second compression means 11b compress the uncompressed data 10c using the MH compression algorithm and the MR compression algorithm in the same manner as the first compression means 11a and the second compression means 11b in FIG. The compression result is the first compressed data 10a and the second compressed data 1
It is written in the temporary storage means 10 as 0b.

The data length counting means 12 is similar to the data length counting means 12 of FIG.
The data length of a, the data length of the second compressed data 10b, and the data length of the non-compressed data 10c are calculated and output to the signal line 514. In the following, the data length of the first compressed data 10a is 10,000, the data length of the second compressed data 10b is 7,000, and the data length of the non-compressed data 10c is 20,000.
The case will be described as an example.

When the data length of the first compressed data 10a is supplied via the signal line 514, the data selecting means 72 sets the supplied value to the address A of the data length memory 72a.
Write to 1. When the data length of the second compressed data 10b is supplied via the signal line 514, the supplied value is written in the address A2 of the data length memory 72a.
Further, when the data length of the uncompressed data 10c is supplied via the signal line 514, the supplied value is written in the address A3 of the data length memory 72a. For example, in the example defined above, the data length of the first compressed data 10a is 1000
0, the data length of the second compressed data 10b is 7000, and the data length of the non-compressed data 10c is 20000.
As shown in (Table 1) A, 10000 is assigned to the address A1 and 7000 is assigned to the address A2 of the data length memory 72a.
20,000 will be written in the address A3.

Next, the data selecting means 72 searches for the smallest value among the values stored in the addresses A1 to A3 of the data length memory 72a. Then, when the value found by searching is the value stored in the address A3, that is, the data length of the uncompressed data 10c, the uncompressed data 10c stored in the temporary storage means 10 is transferred to the signal line 517. , And outputs to the signal line 518 as it is, outputs the value 3 to the signal line 519 as an identifier indicating that the data need not be expanded, and ends the data selection operation. On the other hand, the value found as a result of the search is the address A1.
If the value stored in the first compressed data 10a is the data length of the first compressed data 10a, the first compressed data 10a stored in the temporary storage means 10 is read via the signal line 517 and the signal is directly input. The value 1 is output to the line 570, and the value 1 is output to the signal line 571 as an identifier indicating the first decompression unit 11c. Nothing is output to the signal lines 518 to 519.
On the other hand, when the value found as a result of the search is the value stored in the address A2, that is, the data length of the second compressed data 10b, the second compressed data 10b stored in the temporary storage means 10 is stored. Is output via the signal line 517 and is output to the signal line 570 as it is.
The value 2 is output to the signal line 571 as an identifier indicating 1d. Nothing is output to the signal lines 518 to 519. For example, when the value of (Table 1) A is stored in the data length memory 72a, the second compressed data 10b is output to the signal line 570, and the value 2 of the identifier is output to the signal line 571. Therefore, nothing is output to the signal lines 518 to 519.

The compression / expansion means 71 reads the data supplied via the signal line 570 and the identifier supplied via the signal line 571, and performs the processing according to the value of the identifier. That is, when the value of the identifier is 1, the data is expanded by the first expansion means 11c using the MH expansion algorithm. The data generated as a result of decompression is not necessary here and is therefore discarded. In addition, when starting the expansion of data, the expansion start signal is sent to the signal line 573.
To the signal line 573 when the expansion is completed. On the other hand, when the value of the identifier is 2, the data is expanded by the second expansion means 11d using the MR expansion algorithm. The data generated as a result of decompression is not necessary here and is therefore discarded.
Further, it outputs a decompression start signal to the signal line 573 when decompressing the data, and outputs a decompression end signal to the signal line 573 when the decompression is completed.

When the expansion start signal is supplied via the signal line 573, the expansion time measuring means 73 sets the initial value to 0 and starts measuring the time elapsed thereafter. Time is measured in milliseconds. Then, until the expansion end signal is supplied via the signal line 573, the measurement result is continuously output to the signal line 574 as the expansion time. For example, the value output to the signal line 574 becomes 10 when 10 milliseconds have passed after the decompression start signal was supplied, and the value output to the signal line 574 was 25 when 15 milliseconds passed thereafter. Become. Further, when the expansion end signal is supplied through the signal line 573, the expansion time measuring unit 73 stops measuring the expansion time and stops outputting the measured value to the signal line 574.

The register 72d has a decompression time measuring means 73.
Stores the upper limit value that can be allowed as the data decompression time measured by, and the stored value is always output to the signal line 520. In the following description, the upper limit of this time is 200
Taking the case of milliseconds as an example, description will be given assuming that the value stored in the register 72d is 200 as shown in (Table 3).

The comparing means 72d constantly reads the expansion time and the value supplied via the signal line 520 while the expansion time is being supplied via the signal line 574, and compares them. If, as a result of the comparison, the extension time is not equal to the value supplied via the signal line 520, nothing is output. On the other hand, when the expansion time becomes equal to the value supplied via the signal line 520, a “false value” is output as the comparison result. Further, when the expansion end signal is supplied via the signal line 573, the expansion time supplied to the signal line 574 immediately before that is compared with the value supplied via the signal line 520. As a result of the comparison, when the expansion time is smaller than the value supplied via the signal line 520, the “true value” is output as the comparison result. On the other hand, if the expansion time is larger than the value supplied via the signal line 520, or if both are equal, it is considered that the comparison result has already been output as a "false value", and nothing is newly output. do not do. For example, it takes 250 milliseconds for the compression / expansion means 71 to expand the compressed data 10b to the end (Table 2) A, and the value stored in the register 72d is 200 as shown in Table 3). Assuming a certain case, 200 milliseconds after the compression / expansion unit 71 starts the expansion of the compressed data 10b, that is, before the expansion is completed, as shown in (Table 5), as a comparison result, a "false value" is obtained. Will be output.

The data selecting means 72 reads the value when the comparison result is output from the comparing means 72c. If the read comparison result is the “true value”, the data output to the signal line 570 and the identifier output to the signal line 571 are output to the signal line 518 and the signal line 519 as they are, and the data selection operation ends. To do. On the other hand, if the comparison result is a "false value", the value stored in the data length memory 72a is searched for and found earlier, and the value is "IN".
VALID ”and outputs a decompression stop command to the signal line 572. For example, the data length memory 72
When “a false value” is output as the comparison result from the comparing means 72c as shown in (Table 5) when a stores the value of (Table 1) A, (Table 1) B So that the memory 62
The value of the address A2 of a is rewritten to "INVALID", and the values of the addresses A1 and A3 are not rewritten.

When the decompression stop command is supplied through the signal line 572, the compression / decompression means 71 stops the decompression operation using the first decompression means 11c or the second decompression means 11d that was being performed at that time. , And outputs a decompression stop signal to the signal line 573. For example, assuming that it takes 250 milliseconds to complete decompression of the second compressed data 10b as shown in (Table 2) A, the decompression stop command is issued via the signal line 572 200 milliseconds after decompression is started. 5 is supplied to the signal line 573 compared to the case where the expansion operation is not stopped midway.
A decompression end signal will be output 0 ms earlier.

After that, the data selecting means 72 searches again for the smallest value among the values stored in the addresses A1 to A3 of the data length memory 72a, except for the value "INVALID". When the value found as a result of the search is the value stored in the address A3, that is, the data length of the uncompressed data 10c, the uncompressed data 10c stored in the temporary storage means 10 is transferred to the signal line 5.
The data is read through 17 and output to the signal line 518 as it is, and the value 3 is output to the signal line 519 as an identifier indicating that the data need not be expanded, and the data selection operation is completed. On the other hand, when the value found as a result of the search is the value stored in the address A1, that is, the data length of the first compressed data 10a, the first compressed data 10a stored in the temporary storage means 10 is stored. Is read out via the signal line 517 and is output to the signal line 570 as it is, and the value 1 is output to the signal line 571 as an identifier indicating the first decompression unit 11c. Nothing is output to the signal lines 518 to 519. On the other hand, when the value found as a result of the search is the value stored in the address A2, that is, the data length of the second compressed data 10b, the second compressed data 10b stored in the temporary storage means 10 is stored. Is read through the signal line 517 and is output to the signal line 570 as it is.
The value 2 is output to the signal line 571 as an identifier indicating d.
Nothing is output to the signal lines 518 to 519. For example, when the value of B in (Table 1) is stored in the data length memory 72a, the value found by the search is 10000, and the value is stored in the address A1. The first compressed data 10a is output, and the value 1 of the identifier is output to the signal line 571.

Thereafter, with respect to the first compressed data 10a, the expansion operation by the compression / expansion means 71, the measurement by the expansion time measuring means 73, and the comparison by the comparison means 72d are performed again, and the comparison result is output. . Further, the data selection means 72 determines the data to be output to each signal line based on the value output from the comparison means 72c.

The data selection means 72, which has been described so far,
Compression / expansion means 71, expansion time measuring means 73, comparison means 7
The overall operation of 2d is as follows. First, the data selection unit 72 stores the data lengths of the first compressed data 10a, the second compressed data 10b, and the non-compressed data 10c supplied from the data length counting unit 12 via the signal line 514 in the data length memory. 72a. Then, the first compressed data 10a, the second compressed data 10b, and the non-compressed data 10c are searched for the one having the shortest data length. When the retrieved data is the uncompressed data 10c, the identifier corresponding to the uncompressed data 10c is output to the signal line 518 and the signal line 519, and the operation is ended. On the other hand, when the retrieved data is the first compressed data 10a or the second compressed data 10b, the compression / expansion means 71 is caused to expand it, and the expansion time measuring means 73 is made to measure the time required for the expansion. . Further, the comparing means 72d compares the decompression time with the upper limit value allowable as the decompression time stored in the register 72d. Then, as a result of the comparison, when the expansion time is smaller than the upper limit value which is allowable as the expansion time, the same data as expanded by the compression / expansion means 71 is output to the signal line 518, and the corresponding identifier is output to the signal line. It outputs to 519 and ends the operation. On the other hand, when the decompression time reaches the upper limit that can be allowed as the decompression time, the compression / decompression means 71 is caused to stop the decompression operation at that stage. The above operation is performed in order from the data having the shortest data length, and finally, the compressed data which can be expanded within a predetermined time or the uncompressed data 10c having the shortest data length is the signal. Line 5
To the signal line 519, the identifier corresponding to the data is output to the signal line 519, and the operation ends.

The data storage means 15 stores the data supplied via the signal line 518 and the identifier supplied via the signal line 519 in association with each other.

After that, the temporary storage means 10 erases the first compressed data 10a, the second compressed data 10b and the non-compressed data 10c.

Here, attention is focused on one piece of acoustic data, and the process from the output of the acoustic data to the signal line 510 by the acoustic input means 79 to the storage of the acoustic data in the data storage means 15 has been described step by step. The storage / playback device sequentially outputs the acoustic data stored in the memory 79c to the signal line 510, and performs the above storage operation until all the acoustic signals input from the microphone 10a are stored as acoustic data in the data storage unit 15. Execute repeatedly.

Next, the operation of reading the acoustic data will be described. The operation of reading the acoustic data is started by the user instructing the reading of data by the keyboard 30, and thereafter, the user operates the keyboard 30.
It continues until a predetermined key is pressed to instruct to stop reading.

The keyboard 30 outputs a character code string representing a user's instruction to the signal line 530.

The read instructing means 23 discriminates the character code string supplied through the signal line 530, and outputs a read start command to the signal line 522 when instructed to read the acoustic data. On the other hand, if the character code string supplied via the signal line 530 is a read stop instruction, the signal line 5
A read stop command is output to 22.

The operations of the timer 24 and the data reading means 25 are exactly the same as the operations of the timer 24 and the data reading means 25 of FIG.

The acoustic memories 76 and 77 store a value representing a silent state in the initial state at all addresses.

Compression / expansion means 71 for reading acoustic data
The operation of is compared with the operation of the compression / expansion means 11 of FIG.
Instead of writing data to the image memory 26 via the signal line 526, data is written to the acoustic memory 76 via the signal line 576, and instead of writing data to the image memory 27 via the signal line 527, via the signal line 577 Except for writing the data in the acoustic memory 77, it is exactly the same, so a detailed description will be omitted, but as in the case of the second embodiment of FIG. 2, the acoustic memories 76 and 77 store data at 250 millisecond intervals. The acoustic data stored in the storage means 15 is
It is decompressed if necessary and then written.

When the data read command is supplied through the signal line 523, the sound generating means 78 outputs from the sound memory 77 through the signal line 579 if the supplied data read command is an odd number, and if the data read command is an even number. After the written data is read from the acoustic memory 76 via the signal line 578 and converted into an analog signal by the D / A converter 78a,
The amplified signal is amplified by the amplifier 78b, and the amplified signal is output to the speaker 78c.
To generate sound. Therefore, when the first data read command is supplied through the signal line 573, the initial value stored in the acoustic memory 77, that is, the data representing the silent state is read and no sound is generated.
When the nth (n ≧ 2) data read command is supplied, the n−1th acoustic data read from the data storage means 15 is output as a sound from the speaker 78c.

Here, the data stored in the data storage means 15 is input from the microphone 79a, and the A / D converter 7
The acoustic data digitized by 9b is stored by dividing it by 250 milliseconds, and one data read from the data storage means 15 is sounded by the speaker 78c as an acoustic sound in 250 milliseconds. It takes time.
On the other hand, since the data read command is periodically supplied from the timer 24 to the signal line 523 at intervals of 250 milliseconds, the entire sound storage / playback apparatus of the present invention stores data in the data storage means 15.
The acoustic data that is divided and stored every 50 milliseconds is continuously speakered without interruption or partial skipping as if all the acoustic data is continuously stored without being divided. You can ring at 78c.

According to this embodiment, the sound data input from the sound input means 79 is compressed by the algorithm which can guarantee the expansion within a predetermined time among the plurality of compression methods, and the data storage means 15 is then used. Since it is stored in, it is possible to reduce the usage amount of the storage medium of the data storage means 15 as compared with the case of storing without compression. In addition, after dividing the acoustic data at a predetermined time interval, it is possible to guarantee that each data can be expanded within a predetermined time, and the compression algorithm with the shortest data length is selected. Even if compared with the case of compressing the whole with one algorithm without doing it, the part in the silent state is an algorithm suitable for the silent state, and the part with a lot of high frequency components is an algorithm suitable for it.
It is possible to reduce the usage amount of the storage medium of the data storage unit 15 by compressing with an algorithm suitable for each partial portion. Further, since acoustic data that cannot be expanded within a predetermined time by any compression algorithm is stored in the data storage means 15 in an uncompressed format, any data is reproduced when the acoustic data is reproduced. Also, it is possible to guarantee that the data stored in the data storage means 15 can be read out within the predetermined time as data in the uncompressed form in the acoustic memories 76 and 77, and thus divided into a plurality of pieces at predetermined time intervals. While the certain acoustic data stored by being reproduced is reproduced by the speaker 78c, the next acoustic data is read from the data storage means 15 and expanded,
The acoustic data can be stored in the acoustic memories 76 to 77, and the acoustic data divided and stored in the data storage means 15 is
As if all the acoustic data is stored continuously without being divided, the speaker 78c can be continuously played without interruption or skipping.

Further, the data selection means 7 in this embodiment.
No. 2 is stored in the data storage unit 15 after the measurement of the expansion time to be performed by the compression / expansion unit 71 and the expansion time measurement unit 13 is completed for both the first compressed data 10a and the second compressed data 10b. Since the compressed data first found to have a data decompression time shorter than the predetermined time is selected instead of selecting the data to be stored, as compared with the data selection means 22 in the image filing apparatus of the second embodiment, Data can be selected at high speed.
Further, since the measurement of the decompression time is performed in order from the compressed data having the shorter data length, the compression algorithm that was the first to confirm that the decompression time is shorter than the predetermined time is the algorithm that minimizes the data length. Can be guaranteed. Further, when the decompression time measuring means 13 and the comparison means 72c detect that the decompression time has reached a predetermined time, the compression / decompression means 71 stops the subsequent decompression of the data, and the decompression time of another compressed data is set. Since the measurement is started, the data to be stored in the data storage means 15 can be selected at a higher speed.

The present invention is not limited to the above-described embodiments, but can be implemented with various modifications without departing from the scope of the invention.

For example, in each embodiment, the data storage means 15 may be realized by using a storage medium other than the optical disk such as a hard disk or flash memory.

Similarly, the temporary storage means 10 may be realized by using a storage medium other than a hard disk such as a semiconductor memory or a floppy disk, or the data storage means 15 may be used.
You may implement | achieve using a part of storage medium in it. In particular,
When the temporary storage means 10 is realized by using a part of the data storage means 15, not only the time required for the expansion time measuring means to expand the compressed data by the compression / expansion means 11, 31, 71, but also prior to that. The time required to read the compressed data from the temporary storage means 10 may also be included in the measurement. In this way, the total time required for reading and decompressing the data is within a predetermined time, taking into consideration the change in the data read time from the data storage unit 15 due to the change in the data length due to the compression. -It is possible to select a decompression algorithm. Further, when the temporary storage means 10 is realized by utilizing a part of the data storage means 15, the data selected by the data selection means 14, 22, 72 is stored in the data storage means 15 again and is temporarily stored. Of course, instead of erasing the data stored in 10, it is possible to leave only the data selected by the data selecting means 14, 22, 72 and erase the unselected data.

The extension time measuring means 13 may be realized by using the AC frequency of a commercial power source instead of the crystal oscillator.

In the first to third embodiments, the first
The compression algorithm used by the compression means 11a and the second compression means 11b is a compression algorithm based on the LZ method,
JBIG (Joint Bi-level Image coding experts Grou
p) compression algorithm, MMR (Modified Modified Relative) used for international standard group 4 facsimile
addressing) compression algorithm, compression algorithm for color image JPEG, compression algorithm for sub-band coding method suitable for audio signals, MH, M
You may change to a compression algorithm other than R. In this case, it goes without saying that the decompression algorithm used by the first decompression means 11c and the second decompression means 11d is also changed to one corresponding to the compression algorithm. Further, the compression / expansion means 11 and the compression / expansion means 71 having these inside are C
It may be realized by PU and software, or three or more compression / expansion algorithms to be applied may be provided.

The sound input means 79 in the sound storage / playback apparatus of the third embodiment is provided with a sound recorder such as a tape recorder, a record player, or a CD player in place of the microphone 79a, and the sound reproduced by the sound player is reproduced. Music may be given to the D / A conversion means 79b.

Further, the image input means 19 in the data storage device of the first embodiment and the image input means 29 in the image filing device of the second embodiment may read a multi-tone monochrome image or a color image. You may do it. Further, instead of reading the image from the paper with the scanner, the image may be read from other media such as movie film. In this case, it goes without saying that the compression / expansion algorithm used by the compression / expansion means 10 is selected according to the property of the image.
MPEG should be included in one of the compression algorithms, especially if there is a strong correlation between adjacent images, such as motion picture film.

In the image filing apparatus of the second embodiment, the display speed in the page turning mode may be slower or faster than 4 sheets per second, or a dial as used in a microfilm reading device may be added. Alternatively, the speed may be set. In particular, when implementing an image filing device using high-speed hardware, a display speed of about 30 images per second can be set, and one frame of a moving image such as a movie or TV broadcast is sequentially stored, The moving image may be displayed by displaying the page turning mode. In this case, the sound storage / playback device of the third embodiment is built-in to store voices and music such as movies and TV broadcasts so that a moving image and voice / music can be played back simultaneously, so-called VTR player or laser disc player. It may be used instead of the conventional AV equipment such as.

In addition, the image filing apparatus of the second embodiment uses the same technique as the conventional apparatus to set an attribute to the image data when storing the image data and to set an attribute to the data when reading the image.
It is also possible to extract only image data having a desired attribute and read only the extracted image data.

[0165]

As described above by using various examples,
According to the present invention, which compression algorithm is used to store data compressed in the data storage means is determined based on not only the counting result of the data length counting means but also the measurement result of the decompression time measuring means. , Rather than simply storing in the storage means the data compressed by the compression algorithm that minimizes the data length, select the data by considering both the data length and the time required to expand the data. It becomes possible to realize a data storage device to be stored in the means.

It is also possible to guarantee that the data stored in the data storage means can be expanded within the allowable time.

Further, among the compression algorithms capable of expanding within the allowable time, the data compressed by the algorithm having the shortest data length can be stored in the data storage means, and the storage medium can be effectively used. .

In the case where all the compressed data are decompressed by causing the decompression means to decompress only the compressed data having a shorter data length than the uncompressed data. Faster than
It is possible to determine which algorithm should be used to store the compressed data in the data storage means.

Further, the data selecting means selects the data which is first detected that the time measured by the expansion time measuring means is smaller than the upper limit value allowable as the expansion time, so that the data compressed by any algorithm is selected. It becomes possible to determine whether to store the data in the data storage unit at a higher speed.

Further, the data selecting means causes the decompressing means to decompress the compressed data in order from the one having the shortest data length, so that the time measured by the decompressing time measuring means is smaller than the upper limit value allowable as the decompressing time. Since it is possible to guarantee that the data first detected is the shortest data length after compression, it is possible to determine at a higher speed which algorithm should be used to store the compressed data in the data storage means.

Further, when the time measured by the expansion time measuring means exceeds the upper limit value allowable as the expansion time,
By ending the expansion of the data by the expansion means and starting the expansion of the new compressed data, it is faster than waiting for the completion of the expansion of the previous data to start the expansion of the new compressed data, It is possible to determine which algorithm should be used to store the compressed data in the data storage means.

Further, if any of the compressed data is the data to be stored in the data storage means, the uncompressed data which does not need the expansion is stored when the time measured by the expansion time measuring means is larger than the upper limit value which is allowable as the expansion time. By selecting as
It is possible to guarantee that the expansion is completed within the allowable time.

Further, when any of the compressed data has a data length counted by the data length counting means that is larger than the data length of the uncompressed data, the uncompressed data is selected to reduce the data length by the compression. On the contrary, it becomes possible to prevent the size from becoming large.

Further, since the image is optically read by the image input means and the image data generated by photoelectric conversion is compressed by the compression means, it is guaranteed that the optically read image can be expanded within the allowable time. It can be stored in the data storage means after being compressed by an algorithm.

Further, since the sound signal obtained by converting the sound to the electric signal by the sound input means is compressed by the compression means, the sound is compressed by the algorithm guaranteed to be able to be expanded within the allowable time and stored in the data storage device. It will be possible.

According to the data reading device of the present invention,
Data storing means for reading the data compressed by the selected algorithm by individually confirming that the data can be expanded within a predetermined time among a plurality of different compression algorithms and an identifier indicating the expansion method of the data It is possible to read out the data, expand the read data by the expansion means according to the identifier, convert the expanded data into information that can be recognized by the human five senses, and present the information to the human at predetermined time intervals.

Further, the data stored in the data storage means includes image data, and by displaying the image by the display means in the conversion means, the data stored in the data storage means is presented to a human as a continuous image. Can be presented.

Further, it becomes possible to display a continuous image at a constant time interval by the display means and display it as a moving image.

Further, the sound data stored in the data storage means is output to the sound generation means at a time interval equal to the sampling cycle used when storing the sound, so that the stored sound can be reproduced as a correct sound by a human. Can be presented to.

[Brief description of drawings]

FIG. 1 is a block diagram of a data storage device according to an embodiment of the present invention.

FIG. 2 is a block diagram of an image filing device in which a data storage device and a data reading device according to another embodiment of the present invention are combined.

FIG. 3 is a block diagram of a sound storage / reproduction device in which a data storage device and a data reading device according to another embodiment of the present invention are combined.

[Explanation of symbols]

 11a 1st compression means 11b 2nd compression means 11c 1st expansion means 11d 2nd expansion means 12 Data length counting means 13, 73 Expansion time measuring means 14, 22, 72 Data selection means 15 Data storage means 19, 29 image input means 22c, 72c comparison means 28 display means 29 image input means 78 sound generation means 79 sound input means

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Office reference number FI technical display location H04N 1/413 D 5/92

Claims (15)

[Claims] 1. A compression means for compressing data by a plurality of different compression algorithms, a decompression means for decompressing data compressed by each compression algorithm, and a decompression time measurement for measuring a time required for decompressing the compressed data. Means and data length counting means for counting the data length of the compressed data,
Data selecting means for selecting one of the compressed data based on the measurement result of the expansion time measuring means and the counting result of the data length counting means, and a data storage means for storing the data selected by the data selecting means. A data storage device comprising: 2. The data selecting means has a comparing means for comparing the time measured by the expanding time measuring means with an upper limit value allowable as the expanding time, and the comparison result of the comparing means and the counting result of the data length counting means. The data storage device according to claim 1, wherein the data is selected based on the data. 3. The data selecting means selects the data having the shortest data length counted by the data length counting means, out of the data whose time measured by the expansion time measuring means is smaller than the upper limit value allowable as the expansion time. The data storage device according to claim 2, which is configured. 4. The data storage device according to claim 2, wherein the data selecting means causes the decompressing means to expand only the compressed data having a shorter data length than the data before compression. 5. The data storage device according to claim 4, wherein the data selection means selects data which is first detected that the time measured by the expansion time measuring means is smaller than the upper limit value allowable as the expansion time. . 6. The data storage device according to claim 5, wherein the data selection means is configured to cause the decompression means to decompress the compressed data in order of decreasing data length. 7. The data storage device according to claim 2, wherein when the time measured by the expansion time measuring means exceeds an upper limit value allowable as the expansion time, the expansion of the data by the expanding means is terminated. 8. The data selecting means selects the data before compression when any of the compressed data is larger than the upper limit of the expansion time which is measured by the expansion time measuring means. Item 2. The data storage device according to item 2. 9. The data selecting means selects the data before compression when any of the compressed data has a data length counted by the data length counting means that is larger than the data length of the data before compression. The data storage device according to claim 1. 10. The data storage device according to claim 1, further comprising image input means for optically compressing image data generated by photoelectrically converting an image and generating photoelectric conversion. 11. The data storage device according to claim 1, further comprising a sound input unit for compressing a sound signal generated by converting sound into an electric signal by a compression unit. 12. A method of storing data compressed by an algorithm selected by individually confirming that data can be expanded within a predetermined time among a plurality of different compression algorithms and an identifier indicating a method of expanding the data. Data storage means, reading means for reading the compressed data and corresponding identifier from the data storage means, decompression means for decompressing the data read by the reading means according to the read identifier, and the decompression means. A data reading device comprising a conversion means for converting the data decompressed by the above into information that can be recognized by the human five senses at predetermined time intervals. 13. The data reading device according to claim 12, wherein the data stored in the data storage means includes image data, and the conversion means has a display means for displaying an image. 14. The data reading device according to claim 13, wherein the display means displays the image as a moving image. 15. The data reading device according to claim 12, wherein the data stored in the data storage means includes acoustic data, and the conversion means has a sound generation means for generating a sound.