JP2507344B2 - Data registration method and data retrieval method in image file system - Google Patents

Description

The present invention relates to a data registration method and a data search method in an image file system. Particularly, the present invention relates to a data registration method and a data search method suitable for displaying a part of a searched image at high speed.

[Prior Art] In recent years, a document image file system (electronic file) using a large-capacity optical disk has been attracting attention as a new document management means. Since the optical disc has a large data storage capacity and can record a large amount of data such as image data, it can be used as a storage means for a document, a design drawing, a contract, and other document image information. Usually, when searching these document images, it is preferable to use indexes such as predetermined document names, classification names, and keywords. However, if the index becomes complicated, the work required to store the document images with these added will increase,
Also, it is often difficult to remember these indexes when searching. Therefore, in practice, data is often stored by adding only a simple index such as a classification name.
Further, it is routinely performed to specify a simple index to search for an image stored by adding a complicated index. In this case, in order to search for the target document image, a plurality of candidate data searched by designating the index such as the classification name are sequentially displayed on the display, and the operator visually confirms the displayed contents to extract the target document. There is a need to.

As a method of selecting the document image described above, for example, as described on pages 6 to 7 of the operation manual (manual number 60-10-001-20) of the optical disk file system of Hitachi, There is a method of using the next page key. This method displays a plurality of image data on the display one by one each time a key is pressed. As an improvement plan of this method, for example, when the page break key is pressed, a method of automatically page breaking the search data until the next key that means a stop command is pressed and sequentially displaying it on the display image is also considered. To be

[Problems to be Solved by the Invention] As described above, when the document images that have been index-searched (primary search) are sequentially displayed and the target document is extracted (secondary search) from them by visual search, the display is performed. It is desired to speed up the page break operation within a range where it can be determined whether or not the document is the target document and to shorten the time required for the secondary search. However, in the conventional search system, the document data that is primarily searched is read from the document image file one page (one record) at a time and sequentially displayed, and the read speed of each image data depends on the file device. Since it is restricted by the performance, the display contents cannot be updated in a cycle shorter than the time required to read the data per document.
There was a problem that a waiting time was generated before the next screen.

In addition, in visual search, it is not necessary to display each document in full, for example, title, author, abstract, etc.
In many cases, the search purpose can be achieved by displaying only one area in the document. For example, if the operator remembers the position where the contents that specify the target document are recorded, the operator will pay attention to the position to determine the necessity of the display document. The display need only be able to display a specific partial area designated by the operator. If only the specified partial area can be read from the file in this way, the time required to read the data will be shortened, and the display screen can be updated in a short time, thus reducing the time required for the secondary search. Can be shortened. However, in the image data file system, in order to effectively use the memory area to store a large number of images in one file medium, it is usual to store the original image as a record in which data is compressed by encoding. In this case, in the record on the data-compressed file medium, since the information is distributed at a density different from that of the original image, even if half the information is read from one record, this is half a page of the original document. Was not always reproduced, and it was not possible for the operator to selectively read only the image data of the partial area desired to be displayed from the file.

An object of the present invention is to provide a data registration method capable of selectively reading out partial image data at a specified position on an original image from a file device that stores image data in a data compressed form by encoding. To provide.

[Means for Solving Problems] In order to achieve the above object, in the present invention, at the time of registration of image data, it is determined where the partial area on the original image is located on the image record recorded in the file. Ask,
This position information is stored in an auxiliary file corresponding to the image file.

The first invention of the present application is an image file system capable of converting a plurality of original images into data-compressed records, sequentially registering them on a file medium, and searching for a desired original image from the registered plurality of original images. , Each of the original images is divided into a plurality of partial areas according to a unified standard, and the step of sequentially compressing and coding each partial area while obtaining the position on the record corresponding to the boundary of each partial image is obtained by compression coding. A step of registering the compressed encoded data as a record in the image file, and a step of recording information indicating the position on the record corresponding to the obtained boundary of each partial area in the auxiliary file corresponding to the image file. It is characterized in that a search from the registered original image can be performed by displaying a partial area that matches a predetermined criterion.

In addition, the second invention of the present application is an image file capable of converting a plurality of original images into data-compressed records, sequentially registering them on a file medium, and searching for a desired original image from the plurality of registered original images. In the system, each original image is divided into a plurality of partial areas according to the unified standard, and each partial area is sequentially compression-encoded while obtaining the position on the record corresponding to the boundary of each partial image. The compression-encoded data is registered as a record in the image file, the information indicating the position on the record corresponding to the boundary of each obtained partial area is recorded in the auxiliary file corresponding to the image file, and the registered original image is recorded. When searching from, the first partial area that meets a predetermined criterion is displayed on the display means, and if the user requests display of the second partial area that is another partial area, the first partial area is displayed. When the size of the second partial region is large as a result of the comparison, the second partial image is displayed when the first partial region is displayed. It is determined whether or not it protrudes below the means. If the result of the determination protrudes, the lower end of the display means and the second
It is characterized in that the lower ends of the partial areas are displayed so as to coincide with each other.

[Action]

According to the present invention, since the position on the record corresponding to the boundary of the partial area on the original image is stored, when the operator designates the partial area displayed on the display image, the position on each record from the image file is changed. Only part of the image data corresponding to the partial area can be selectively read. Therefore, as compared with the case of reading all the data of the record from the image file, the time required for reading and transferring the data and the time required for the conversion process for restoring the compressed data to the original data are shortened, and the display is reduced. It is possible to speed up the cycle of image update (page break operation).

[Example] An example of the present invention will be described below.

FIG. 1 is an overall configuration diagram of an image file system according to the present invention, in which 1 is a scanner for inputting image data, 2 is a controller of the scanner 1, and 3A is a compression encoded form of the image data that has appeared. Image file to be stored, 3B is an index file that stores index data prepared for searching each image data, 4 is a controller that writes and reads data to and from the image file 3A and the index file 3B, and 5 is A display device for displaying character data such as image data and commands, 6 is a controller for controlling the display device 5, 7 is a keyboard for inputting character data such as commands and index data, and 8 is the entire system (CPU) that controls the CPU, 9 stores the program executed by CPU8 Memory, 10 is a work memory for storing various variables or tables used by the CPU 8, 11A and 11B are buffer memories for temporarily storing the encoded image read from the image file 3A, 13 Is an image memory for storing the original image data input from the scanner 1, 14 is a dedicated processor for encoding the original image data, 15 is a dedicated processor for restoring the encoded image data, and 16 is Bit map memory corresponding to the display contents of the display image, and 17 indicates a bus.

In this configuration, the memories 9 to 13 and 16 do not have to be physically different devices, and may be allocated as different storage areas on the same memory device. Further, as will be described later, it is possible to reduce the total memory amount by sharing some areas in this area. The image file 3A and the index file 3B may be configured by different file media, but since these are paired with each other,
Here, it is assumed that different areas on one optical disc 3 are assigned to the files 3A and 3B, respectively. Therefore, the file controller 4 is shared by the two files 3A and 3B. The file controller 4 is an optical disc 73
The drive device of is included. Also the encoding processor
Although the 14 and the restoration processor 15 are different in circuit, they are the same.
Since they are often realized on LSI, they are drawn adjacent to each other in the figure. Various types of image data encoding algorithms performed by the encoding processor 14 are conventionally known. For example, "FA
Article entitled "Signal Processing of Images for X, OA" (1982)
The run length coding method described on pages 61 to 75 of the above can be used. However, the gist of the present invention does not change depending on the encoding algorithm, and any other algorithm may be used.

FIG. 2 is a diagram showing an image file area 3A and an index file area 3B on the optical disc 3. On the optical disc 1, one spiral track TR is formed continuously from the center side of the disc toward the outside. This track TR has a plurality of tracks TR- with the reference line L as a boundary.
It is divided into 0 to TR-m + n, and the track numbers of 0 to m + n are given in order from the inner one. Further, each track is divided into a plurality of sectors SE-0 to SE-p, which are the minimum units of a data block to be read and written, and each sector has a read / write signal when the optical disc 1 is rotated in the R direction. 0 / p / sector numbers are given in the order in which the heads pass. In this embodiment, the 0th track TR-0
The sector to the last sector of the m-th track TR-m is used as the index file area 3B for storing the index which is the code data, and the m + 1-th track TR-m + 1
From the 0th sector to the last sector of the (m + n) th track TR-m + n are set as the image file area 3A for storing the document image data. The address of each sector can be uniquely determined by the track number and the sector number. Here, the sector number is assigned in order from the 0th sector of the 0th track, for example, the jth sector of the i-th track is assigned. [I
-(P + 1) + j] sector is defined.

FIG. 3 shows original image data 20 stored in the image memory 13.
And the image data (record) 30 compressed by the encoding stored in the buffer memory 11A. In the present invention, the original image data is divided into P partial areas, and each partial area is encoded. In this example, P = 4,
The original image data 20 includes four rectangular partial areas 21 of the same size.
Divided into ~ 24. Each partial area 21 in the original image
Since 24 contains different amounts of information, if the image data is encoded for each partial area, the encoded image data
The partial areas 31 to 34 on the area 30 have different sizes according to the amount of data. In the figure, B (i) indicates the data size of the i-th partial area on the encoded image. For practical purposes, it is convenient to represent this data size by the number of bytes and the number of bits of the terminal byte, but in this embodiment, for the sake of convenience of explanation, it will be represented by the total number of bits. Further, A (i) attached on the encoded image 30 indicates the start address of the i-th partial area, and this address is the sector number in the image file area 3A described above and the sector address. Although it is more practical to represent by the byte position and the position of the terminal bit, in this embodiment, the sector number A '(i) and the bit position a (i) in the sector are used for the sake of explanation later. .

In the embodiment, as described above, the original image data 20 is divided into a plurality of partial areas, each partial area is encoded,
When the partial area 2i on the original image is designated, the boundary position of each encoded partial area 3i, here the start address A (i), can be specified so that the partial area 3i on the encoded image corresponding to this can be specified.
And data size B (i) as index file 3B
It is characterized by keeping a record above. The index record 40 recorded in the index file 3B is configured as shown in FIG. 4, for example. In the figure, 41 is an index field, and 42-i and 43-i are sector numbers A '(i) on the image file 3B in which the i-th partial area is recorded.
And a field for recording the bit position a (i) in the sector, 44-i is the length B of the encoded data of the i-th partial area.
A '(i) is a field for recording (i).
And a (i) specify the start address of the first partial area on the image file.

If the index file is configured in such a manner, when the index-searched image data is sequentially output to the display device 5, for example, the operator specifies display of only the upper half of each image data. In this case, only the coded image name areas 31 and 32 corresponding to the partial areas 21 and 22 corresponding to the upper half on the original image can be selectively read from the image file 3A. By this selective data reading, the time required for reading each image data from the image file can be shortened, so that the update cycle of the display content on the display image can be remarkably shortened.

Next, with reference to the flow chart shown in FIG. 5, the formation of the above-mentioned image file and index file, and the control operation of the image filing system for performing image retrieval using this will be described. The program shown in this flowchart is stored in the memory 9 and executed by the CPU 8. First, in step 102, the optical disc 3 is mounted on the file controller 4, and in step 104, the address (index file pointer Pb) on the optical disc where the next index record should be stored and the next image data should be stored. The address (image file pointer Pa) on the optical disk is obtained by searching the optical disk 3 and stored in the work memory 10.

In step 106, determine the input command from the operator,
If it is a registration command, the input image data from the scanner 1 is written in the image memory 13 in step 108, and the image data is transferred to the bitmap memory 16 in the next step 110 and displayed on the display image. The operator determines the quality of the image displayed on the display 5, such as the tilt and position,
Judgment of light and shade, etc., and input the result (Step 11
2). If there is no problem with the input image, in step 114, the index data such as the document name and the classification name which are the search keys of the image data are input from the keyboard 7. These index data are input to the work memory 10.
In this embodiment, it is assumed that only a character string indicating the classification name of a document such as “TOKKYO (patent)”, “RONBUN (paper)”, “HOUKOKUSYO (report)” is input as an index. Next, in step 116, the number of divisions P of the original image described above is designated by a command input from the keyboard 7. The division number P may be determined in advance as a fixed value on the system side, and in this case, the input operation in step 116 is unnecessary.

In steps 118 to 126, the original image data 20 input to the image memory 13 is divided into a plurality of partial areas by the division number P,
The encoded image data is obtained in the buffer memory 11A, and the index memory described in FIG.
The sequence for forming a record is shown. First, in step 118, the value of the divided area designating parameter i is initialized (i = 1), and then the value of the image file pointer Pa obtained in step 104 is set in A (1). At this point, the pointer Pa points to the head bit position of the specific sector, and a (1) = 0. In step 120, the i-th partial area 2i of the original image is encoded. This encoding processing is performed by the encoding processor 14, and the encoded image data area 3i is stored in the buffer memory 11A.
Immediately after the above (i-1) th partial area (the first partial area 31 is written from the beginning of the buffer memory 11A). In step 122, the size B (i) of the partial area 3i of the encoded image is stored in the work memory together with the address A (i).
In the step 124, the storage address A of the next i + 1-th partial area 3i + 1 is added.
(I + 1) is calculated as A (i + 1) = A (i) + B (i). A (i + 1) is the sector address A '
Expressed separately as (i + 1) and bit position a (i + 1), the above address is It is calculated by a (i + 1) = (a (i) + B (i)) mod K. Here, the symbol [] is a Gaussian symbol indicating truncation into an integer, K is the number of bits per sector, mo
d represents the remainder of division.

In step 126, the value of parameter i is incremented by 1 and i
By repeating steps 120 to 126 until the number of divisions exceeds P, the coded image 30 of FIG. 3 and the index record of FIG. 4 are obtained in the buffer memory 11A and the work memory 10, respectively.

Data symbol capacity per sector on optical disk is 100
Bytes, that is, K = 800 (bits), file
When the sector address pointed to by the pointer Pa is "50", the sizes of the coded partial areas B (a) to B (4) are "1200", "4800", "2400". The index record 40 of the image of "2400" is given the partial area addresses A (1) to A (4) with the values shown in FIG.

Encoded image data 30 obtained in the buffer memory 11A
Image file area 3 on this disc in step 128
The index record 40, which is recorded in the area of a plurality of sectors starting from the address A (1) of A and is formed in the work memory 10, is the index file pointer in the index file area 3B on the optical disk in step 130. It is recorded in the sector at the address pointed by Pa. In step 132, the values of the image file pointer Pa and the index file pointer Pb are updated to the values indicating the next recording start sectors, respectively, in preparation for the registration process of the next image data, and the control sequence returns to step 106. Waits for a new command input from the operator.

Next, a control operation when the operator inputs a command for designating an image search will be described.

FIG. 6 is a diagram schematically showing the search function in the file system of the present invention, in which 20,20 'and 20 "respectively indicate the image data searched. The shaded portions in the figure are displayed on the screen. (A) shows the entire image, (B) shows the upper half of the image (first and second partial areas), and (C) shows the upper part of the image. 1/4 (first partial area) is displayed, (D) is the second and third partial areas in the center of the image, and (E) is 1/4 of the top of the image. The second partial area starting from the position is displayed The selection of each of the above display states can be arbitrarily designated from the keyboard 7 shown in Fig. 7, and the sequential display of images can also be performed in the forward direction. Either of the opposite directions can be specified arbitrarily.

In FIG. 7, 71 is a command input key for displaying the next image by turning one page forward, and 72 is a command input key for displaying the next image by turning one page backward. , 73 is a command input key for continuously turning forward and sequentially displaying pages until the stop key 70 is pressed, and 74 is a command for sequentially turning backward and sequentially displaying pages. Indicates a key for entering. The display size can be changed as shown in Fig. 6 (A).
The ten-keys 75 to 78 corresponding to the numbers shown between-(B) and (B)-(C) are used, and the display position is changed by (B)-(D), (C)-(E in FIG. ) The cursor keys 79 and 80 corresponding to the arrow shown between are used. Here, the number j
The size of the image displayed by the ten-key pad with is set to (1-j / P) times the total size, and the value of P is "4". Further, the amount of change in the display position changed by the cursor keys 79 and 80 is set to an amount corresponding to 1 / P times the original image as a whole. In the keyboard shown in Fig. 7, the key group 82
Is used for inputting character strings such as indexes.

The above search function is realized by the control sequence shown in FIGS. 5B to 5D which is executed when the search command is input. First, in step 134, the index of the desired image data, for example, "TOKKYO" is input to the keyboard 7.
Enter from. The input index is the work memory 10
Stored in a predetermined area in the index file 3B in step 136, and the record having the above index is searched. This search is performed by sequentially reading the data in the index file 3B into the work memory 10 and matching the index character strings. The set of index records searched is shown in the work memory 10 in FIG. It is stored as a record table 80. Each record has an index field as in Fig. 4.
81 and address fields 82-i, 83- for each partial area
i and a size field 83-i. l
Indicates the number of data items extracted by the index search.

Next, in step 138, a variable "S" indicating the size of the display image is displayed.
IZE ”and the variable“ POS ”indicating the start position of the display partial area are initialized. Here, “SIZE” and “POS” are 1 for the entire image.
With / P as the unit, the initial value of "SIZE" is "P", "PO"
It is assumed that "S" is "O", that is, the entire surface of the image is displayed as shown in FIG. In step 140, the image designation parameter j to be displayed next is initialized to “O”, and
11A is designated as the buffer memory for storing the image read next from the image file 3A. Step 142
Waits for a command input from the operator, and checks the validity of the input command in step 144. Here, the forward one page display command C2 corresponding to the key 73, the backward continuous display command C3 corresponding to the key 74, the backward one page display command C4 corresponding to the key 72, and the display size are changed. Only the end commands corresponding to keys 75 to 78, keys 79 and 80 for changing the display position, and key 81 are valid. Further, the forward display command is input while the final image having the image number 1 is already displayed, and the backward display command is input when the first image having the image number 1 is already displayed. Is also invalid like the input of the undefined command. In step 148, the display command is forward (C1
Or C2) or the reverse direction (C3 or C4) is determined, and step 150 or 152 is selectively executed to determine the designated parameter number j of the image to be read next.

In step 154, it is checked whether or not a command for changing the display size or a command for changing the display position has been input, and if it has been input, the variables SIZE and POS are changed.

FIG. 9 is a detailed flowchart of the subroutine corresponding to step 154 above. Numeric keypad 7 in step 802
5 to 78, that is, it is judged whether or not the command for changing the size of the display has been input, and if it has been input, step 80
Execute the sequence from 4 to 814. After confirming in step 804 that the specified new size is not equal to the current size, in step 806 it is determined whether the new size is less than the current size. If smaller, step 80
At 8, the portion which is currently displayed and is no longer required to be displayed is cleared on the bitmap memory 13. In step 810, it is determined whether or not the image should be projected downward when the image is resized to the designated size at the current display position.
In the case of protruding, the position POS is moved upward until the lower end of the display unit matches the lower end of the image. Finally, step 8
At 14, change the size SIZE.

If the size change command has not been input in step 802, step 816 determines whether or not the command for moving the display position downward has been input. If this command has been input, in step 818, after confirming that the current position is not at the bottom end, in step 820, the portion of the current display portion that is not required to be displayed after the display position is moved is displayed on the bitmap memory 13. clear. Finally, in step 820, the value of the position POS is incremented and the display partial area is moved downward.

If the downward movement command has not been input, the process proceeds to step 824, and it is determined whether the upward movement command has been input.
If this command is entered, step 826
Run ~ 830. In steps 824 to 830, the display position is moved upward by the same idea as in steps 816 to 822. In this subroutine, after the processing of each command is completed, the process returns to the first step 802 so that the command for changing the size of the display area and the command for changing the display position are accepted in an arbitrary order, and whether the next command has been input. The determination as to whether or not to do so is repeated.

Returning to FIG. 5 (B), in the next step 156, the eighth
Referring to the record of the image number j in the table in the figure, the divided image data from the Aj '(POS) sector to the "SIZE" number of Aj' (POS + SIZE-1) sectors of the optical disk are read out and read. Store in buffer memory. 11A and 1
The buffer memory of 1B to be read is determined according to the state at that time, as will be described later, but in the initial state, it is read to 11A set in step 140. After confirming the completion of the reading of the image data in step 158, in step 160, the buffer memories 11A to 11B from which the image data should be read next are switched to the buffer memories 11A to 11B (next time from 11B to 11A). Below, step 142
When the command input in step 3 is the forward display command, steps 164 to 186 shown in FIG. 5C are executed, and when it is the reverse display command, steps 188 to 210 shown in FIG. 5D are executed.

In the case of the forward display command, j and l are compared in step 164, and if the latest image read is not the final data of the image number l, steps 166 to 182 are executed. In step 166, the data of the image number j on the buffer memory 11A or 11B is restored by the restoration processor 15 and stored in the bitmap memory 13. The image data to be restored is B _j from the a _j (POS) th bit in the buffer memory.
(POS) + B _j (POS + 1) + ・ ・ ・ ・ ＋ B _j (POS + SIZE-
1) It is a bit. After the subroutine of FIG. 9 is executed again in step 168, in step 170, the (j + 1) th image is read into a buffer memory different from the buffer memory from which the jth image was read. The restoration process in step 166 and the reading process in step 170 are performed in parallel by using the bus 17 in a time-division manner. After confirming the end of the restoration process and the read process in steps 172 and 174, it is confirmed in step 176 whether the stop command 70 has been input by the operator. If it has been input, the process returns to the command input waiting state of step 142. If the data has not been input, first in step 178, the buffer memory from which the image data is to be read is already stored in 11A.
11B when using 11B and vice versa.
Switch to. Next, at step 180, it is judged whether it is a forward continuous display command or a forward 1-page display command. In the former case, step 182 is executed, and steps 164 and thereafter are repeated. In the latter case, the process returns to the command input waiting state of step 142. If it is determined in step 164 that the image read is the final data, the next image data read is not necessary, so only steps 184 and 186 corresponding to steps 168 and 172 are executed, and the process returns to the command waiting state of step 142.

The sequence of steps 188 to 210 executed when the backward display command is selected is the same as steps 164 to 186 described above except that the image reading order is reversed. The description is omitted.

As can be seen from the above description of operation, since registration and search are performed in different modes, the memory that is required only during registration 13
Can be also physically used as the memory 11A or 11B. In the above embodiment, the image file system capable of both registration and retrieval has been described, but the partially readable image file medium to which the present invention is applied can be removed from the system and transferred to another system. Therefore, there is a system only for registration processing, and the search function is not always necessary here. Further, in the embodiment, the original image is divided into equal-sized partial areas by the number of divisions P, but it is divided into predetermined unequal-sized partial areas, or divided according to the size of the original image. The position may be changed. Further, in the above embodiment, the information indicating the size and position of each partial area is stored in the index file as the information indicating the division state, but only either the size information or the position information is stored. Alternatively, the position or size may be calculated during the search. Further, in the above embodiment, one index record is provided for one image, but it may be provided for each index record partial area.

In the above embodiment, since one image is displayed for one image, the display position of each image on the display 5 is basically unchanged even if the position or size of the partial image designated for display changes. is there. However, it is also possible to change the display position of each image as follows and display a plurality of images in one image. FIG. 10 shows the display contents of the display 5 when two pieces of partial image data cut out in the state of FIG. 6 (B) or (D) are displayed on each screen. The numbers circled here represent the image numbers, and in this example, the corresponding partial areas of the next image are displayed alternately above and below one image. FIG. 11 shows the display contents of the display 5 when the partial image data cut out in the state of FIG. 6 (C) or (E) is displayed on each pixel by four items. Here, the corresponding partial image of the next image is cyclically shown in the display area obtained by dividing the image into 1/4. These display formats can be easily realized by sequentially changing the addresses of the bitmap memory 16 storing the restored image in steps 166, 184, 190 and 208 in the flowchart of FIG.

[Effects of the Invention] As is apparent from the above description, according to the present invention, a part of image data encoded and stored in a file can be selectively read out and can be restored and displayed.
When displaying a plurality of images sequentially, all the images can be displayed in a short time. As a result, when the operator remembers the approximate position of the characteristic portion of the target image to be searched, the image can be quickly selected as the target from a large number of images.

[Brief description of drawings]

FIG. 1 is a block diagram showing the overall configuration of an image file system, and FIG. 2 is a storage medium 1 that constitutes an image file.
FIG. 3 is a diagram showing an example, FIG. 3 is a diagram for explaining a correspondence relationship between a partial area on the original image data 20 and a partial area on the record 30 recorded in the file, and FIG. 4 is a record of the index file. .Figure showing one embodiment of the format, fifth
FIGS. 6A to 6D are flowcharts showing an embodiment of a file system control program according to the present invention, and FIG.
FIG. 7 is an explanatory diagram of a data search function in a file system to which the present invention is applied, FIG. 7 is a diagram showing an embodiment of a keyboard, and FIG. 8 is an index of image data to be displayed.
Explanatory drawing of index table consisting of records, No. 9
FIG. 10 is a flowchart showing an example of a program for changing the display position and size of image data, and FIGS. 10 and 11 are diagrams showing other embodiments of the display format for explaining the present invention. Explanation of code 3A: Image file, 3B: Index file, 20: Original image data, 21 to 24: Partial area, 30: Compressed data (encoding)
Record, 31-34: Encoded partial area, A (1) -A
(4): Address.

Claims (6)

(57) [Claims] 1. An image file system capable of converting a plurality of original images into data-compressed records, sequentially registering them on a file medium, and searching for a desired original image from the plurality of registered original images, Each of the original images is divided into a plurality of partial areas according to the unified standard, and the step of sequentially compression-encoding each partial area while obtaining the position on the record corresponding to the boundary of each partial image, and the compression-encoding There is a step of registering the compression-encoded data as a record in the image file, and a step of recording the information indicating the position on the record corresponding to the obtained boundary of each partial area in the auxiliary file corresponding to the image file. However, it is possible to perform a search from the registered original image by displaying a partial area that meets a predetermined criterion and performing the search. System data registration method. 2. Information indicating the position on the record corresponding to the boundary of each of the partial areas is recorded in the auxiliary file in association with index information for searching an image of a desired type from the image file. A data registration method in an image file system according to claim 1. 3. The information indicating the position on the record corresponding to the boundary of each partial area is the start address and data size on the record of the compression encoded data of each partial area. A data registration method in the image file system according to claim 1. 4. The data registration method in an image file system according to claim 1, wherein the unified standard is that each of the original images is divided into partial areas at the same position. . 5. The data registration method in an image file system according to claim 1, wherein the unified standard is to equally divide each of the original images by the same number of divisions. 6. An image file system capable of converting a plurality of original images into data-compressed records, sequentially registering them on a file medium, and searching for a desired original image from the plurality of registered original images, Each original image is divided into multiple partial areas according to a unified standard, and each partial area is sequentially compression-encoded while obtaining the position on the record corresponding to the boundary of each partial image, and compression-encoding obtained by compression-encoding The data is registered as a record in the image file, and the information indicating the position on the record corresponding to the obtained boundary of each partial area is recorded in the auxiliary file corresponding to the image file. When searching for, the first partial area matching the predetermined criterion is displayed on the display means, and when the user requests the display of the second partial area which is another partial area, The size of the first partial area is compared with the size of the second partial area, and if the result of comparison is that the second partial area is large, the second partial image is displayed at the position where the first partial area is displayed. When it is displayed, the image file is characterized in that it is projected below the display means, and if the result of the judgment is projected, the lower end of the display means and the lower end of the second partial area are displayed in coincidence with each other. -Data retrieval method in the system.