JPH11296676A - Image data classification method and image data registration method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To efficiently detect the features of inputted image data and to classify them into corresponding formats. SOLUTION: This method is provided with a process (S101) for inputting the image data, the process (S102-S103) for detecting feature information corresponding to the feature detection item from the image data corresponding to the priority order of the predetermined feature detection item and collating the detected feature information with the feature information on the format registered beforehand and the process (S104-S106) for outputting the format in the case that the detected feature information and the format of the feature information extracted by the collation are in one-to-one correspondence.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image data classification method and an image data registration method for inputting image data in which various images such as characters, tables and photographs are mixed and performing predetermined classification and registration.

[0002]

2. Description of the Related Art In recent years, image data in which a variety of images are mixed is input, and an OCR (optical character reading device) is input.
An image processing system that reads data by using a filing system or stores and manages the image data by a filing system is considered. In this image processing system, a process of classifying input image data based on predetermined characteristics is performed.

That is, in an image processing system, when inputting image data with an image input device such as an image scanner or a facsimile, a type of image to be input is designated in advance by an operator, and a sheet for determining the type of image is used. (Eg, a batch slip), and a classification ID (identification code) is attached to the sheet for image input.

Further, recently, the features of ruled lines of image data to be classified are extracted, and the characteristics are registered and classified in advance as the format of the image, and when new image data is input, the image is deleted. Features of the ruled line of the data,
It is also considered that the format is specified by comparing it with the features of the ruled lines of the registered format, and classification is performed.

[0005]

However, the method for specifying the format based on the characteristics of the ruled line has the following problems. That is, when image data is input by an image input device such as an image scanner or a facsimile, the ruled line is formed as image data depending on various conditions such as the thickness, color, and density of the ruled line of the input sheet based on the resolution. There is a problem that the feature of the ruled line cannot be sufficiently detected because it may not appear in a perfect pattern.

In this case, if the features of the ruled lines of the input sheet are similar, it is difficult to distinguish them. In general, forms used in business such as office work, accounting, and exchange have similar characteristics of the ruled lines, and the format of the form cannot be accurately specified only by the characteristics of the ruled lines.

Therefore, in addition to the characteristics of the ruled lines, other characteristic items such as the size, color, copy-forgery-inhibited pattern, character recognition result of the characteristic region, pattern matching, black dot detection result, character detection result, and character type detection result are included in the form. It is also conceivable to specify the format correctly by adding and combining.

[0008] By performing feature detection of input image data using many features as described above, similar image data can be accurately classified, but the more features are registered, the longer the feature detection takes. However, it takes a lot of time to specify the format. Also, by registering many features, it becomes necessary to secure a large memory capacity.

[0009]

SUMMARY OF THE INVENTION The present invention relates to a method for classifying image data and a method for registering image data which have been made to solve such problems. That is, the image data classification method of the present invention includes a step of inputting image data, and detecting feature information corresponding to the feature detection item from the image data in accordance with a predetermined priority order of the feature detection item. Comparing the detected feature information with the feature information in a pre-registered format; and detecting the detected feature information and the format of the feature information extracted by the matching.
Outputting the format when the correspondence is one-to-one.

According to the present invention, characteristic information of image data is detected in accordance with a predetermined priority order of characteristic detection items, and the detected characteristic information is compared with characteristic information of a format registered in advance. If the detected feature information and the format of the feature information extracted by collation correspond one-to-one, the format is output. For this reason, it is not necessary to detect the feature information corresponding to the feature detection item having the priority order after the format is output, and the format can be specified efficiently.

The image data registration method according to the present invention comprises the steps of: inputting image data; and detecting feature information corresponding to the feature detection item from the image data in accordance with a predetermined priority order of the feature detection item. A step of comparing the detected characteristic information with the characteristic information of a format registered in advance, and, if the detected characteristic information is not extracted from the characteristic information of the format,
Registering the detected feature information as new feature information of the corresponding feature detection item.

In the present invention, the feature information of the image data is collated in accordance with a predetermined priority order of the feature detection items, and the collation is performed, and a format in which the previously detected feature information is registered in advance. If the feature information is not extracted from the feature information, the feature information is registered as new feature information of the corresponding feature detection item, so the feature information corresponding to the feature detection item with high importance is preferentially registered. become able to.

[0013]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the image data classification method and the image data registration method of the present invention will be described below with reference to the drawings. First, an embodiment of the image data classification method of the present invention will be described. FIG. 1 is a flowchart illustrating an image data classification method according to the present embodiment.
FIG. 2 is a hardware configuration diagram for realizing an image data classification method.

That is, as shown in FIG. 2, the hardware for realizing the image data classification method in this embodiment mainly includes an image input unit 1, an image data storage memory 2,
It comprises a feature priority information storage memory 3, a feature detection unit 4, a feature storage memory 5, a format registration memory 6, a format classification unit 7, and a format classification result storage memory 8.

The image input unit 1 is composed of a scanner or a facsimile that reads an image of a form or a sheet to be classified, and has a function of storing the read image data in the image data storage memory 2.

The image data storage memory 2 includes an image input unit 1
Is a memory for storing the image data input in step (1). The feature priority information storage memory 3 is a memory for storing correspondence between feature detection items, which are items for classifying image data to be classified, and their priorities. The priority order differs depending on the variation in the characteristics of the image data to be classified, the difficulty level of the feature detection (processing time, resources), and the like.

The feature detecting section 4 is provided so as to detect feature information by the number of feature detection items for which the priority order is set. For example, when the form size is set as a feature detection item, the feature detection unit 4 uses the image data input by the image input unit 1 and stored in the image data storage memory 2 to determine the size of the form (vertical and horizontal images). Of the size)
The detected value is output as feature information. Similarly, the feature detection unit 4 detects the feature information for each set feature detection item.

The feature storage memory 5 is a memory for storing feature information detected by the feature detection unit 4, and is prepared corresponding to the number of feature detection items. The format registration memory 6 is a memory in which characteristic information detected in advance from image data to be classified is registered, and characteristic information of a format serving as a reference for performing classification is registered.

The format classification unit 7 compares the feature information of the image data stored in the feature storage memory 5 with the feature information of the format registered in advance in the format registration memory 6 and selects one or a plurality of formats. It has the function to do.

A format classification result storage memory 8
Is a memory for storing information of the format selected by the format classification unit 7.

Next, an image data classification method according to this embodiment will be described with reference to the flowchart of FIG. In addition,
In the following description, reference numerals not shown in FIG. 1 refer to FIG. In the present embodiment, when classifying image data, it is assumed that four formats shown in FIGS. 3 to 3 are stored as formats stored in the format registration memory 6.

Here, the format has a different form size than the format, the format has a different form title than the format, and the format has a different ruled line type than the format. .

First, image data (form image) to be classified is input by the image input unit 1 and stored in the image data storage memory 2 (step S101). Next, the feature detection item having the highest priority is acquired from the feature priority information storage memory 3 (step S102), and the feature information of the feature detection item is obtained by the feature detection unit 4 (step S103).

The feature information of the feature detection item having the highest priority of the image data detected by the feature detection unit 4 is stored in the feature storage memory 3.

Next, the feature information stored in the feature storage memory 3 is compared with the feature information of each format stored in the format registration memory 6 by the format classification unit 7 (step S104). By this comparison, it is determined whether there is only one format (one-to-one) having the feature information matching the feature information of the image data (step S).
105).

If there is only one format that matches the characteristic information, it is determined that the input image data is that format, and the determined format is output to the format classification result storage memory 8 (step S106).

On the other hand, if there are two or more formats (1: n) having the feature information matching the feature information of the image data, it is determined whether or not there is a feature detection item having the next priority. Step S107) In some cases, the processing of steps S102 to S105 is repeated. That is, feature information is detected from the input image data for the feature detection item having the next priority, and the format classification unit 7 corresponds to the feature information and the feature detection item of each of the two or more previously selected formats. Is compared with the obtained feature information.

The characteristic information of the image data is compared with the characteristic information of the format in accordance with the priority assigned to the characteristic detection item. When one format is finally determined, the format is stored as a format classification result. Output to the memory 8. If one format is not determined even if the feature information of all the feature detection items is checked, a plurality of formats selected at that time are output to the format classification result storage memory 8 as unclassifiable (see FIG. 3). Step S108).

Thus, it is possible to detect the characteristic information of the image data in accordance with the order of priority and determine the format efficiently. That is, since the feature information of the image data is detected in the order of the feature detection items having the higher priority, it is not necessary to detect the feature information of the feature detection items having the lower priority when the format is determined. It is possible to quickly classify waste without performing collation.

Next, a specific example of this embodiment will be described. FIG. 4 is a diagram showing an example of correspondence between feature detection items and their priorities,
FIG. 5 is a diagram illustrating an example of image data to be classified.
That is, as shown in FIG.
“Form size”, “form title”, and “ruled line feature” are set, and priorities are set in this order. When inputting image data as shown in FIG. The "form size", which is a feature of the above, is detected.

In the form shown in FIG. 5, since the form size matches the format shown in FIG. 3, (1 to n), one format cannot be determined. Therefore, the second feature “form title” having the next priority is detected.

The "form title" can be obtained by recognizing a character at a predetermined position (coordinate) of the input image data. In the form shown in FIG. 5, "payment slip" can be recognized as "form title".

Then, among the formats shown in FIG. 3 which have been previously selected, a search is made for one whose “form title” is “payment slip”. Here, the "form title" of the format is "payment slip", and since the other formats are different, only one format is selected (one-to-one).

That is, in this case, one format can be selected by detecting the first and second features without detecting the "ruled line feature" which is the third feature. You will be able to determine the format.

Next, an embodiment of the image data registration method of the present invention will be described. FIG. 6 is a hardware configuration diagram for realizing the image data registration method according to the present embodiment, and FIG. 7 is a flowchart illustrating the image data registration method according to the present embodiment. This image data registration method newly registers feature information of a format serving as a reference for classification in the image data classification method described above.

As shown in FIG. 6, the hardware for realizing the image data registration method in the present embodiment mainly includes an image input unit 1, an image data storage memory 2, a feature priority information storage memory 3, a feature detection unit 4, Feature storage memory 5,
It comprises a format registration memory 6, a format classification unit 7, a format feature determination unit 9, and a format registration unit 10.

The image input unit 1 comprises a scanner or facsimile for reading an image of a form or a sheet to be registered, and has a function of storing the read image data in the image data storage memory 2.

The image data storage memory 2 includes an image input unit 1
Is a memory for storing the image data input in step (1). The feature priority information storage memory 3 is a memory for storing correspondence between feature detection items, which are items for obtaining feature information of image data to be registered, and their priorities. The priority order differs depending on the variation in the characteristics of the image data to be registered, the difficulty of feature detection (processing time, resources), and the like.

The feature detecting section 4 is provided so as to detect feature information by the number of feature detection items for which the priority order is set. For example, when the form size is set as a feature detection item, the feature detection unit 4 uses the image data input by the image input unit 1 and stored in the image data storage memory 2 to determine the size of the form (vertical and horizontal images). Of the size)
The detected value is output as feature information. Similarly, the feature detection unit 4 detects the feature information for each set feature detection item.

The feature storage memory 5 is a memory for storing the feature information detected by the feature detection unit 4, and is prepared corresponding to the number of feature detection items. The format registration memory 6 is a memory in which characteristic information detected in advance from image data to be registered is registered, and characteristic information of a format serving as a reference for classification and new registration is registered.

The format classification unit 7 compares the feature information of the image data stored in the feature storage memory 5 with the feature information of the format registered in the format registration memory 6 in advance, and determines whether there is a corresponding format. It has a function to make a judgment.

When the format classifying section 7 determines that there is no corresponding format, the format feature determining section 9 has a function of determining feature information detected so far as feature information in the format of the image data. .

The format registration unit 10 has a function of registering the feature information determined by the format feature determination unit 9 in the format registration memory 6 as a format of the image data.

Next, the image data registration method according to the present embodiment will be described with reference to the flowchart of FIG. In addition,
In the following description, reference numerals not shown in FIG. 7 refer to FIG. Further, in the present embodiment, when registering image data, it is assumed that the format registration memory 6 has already stored the four formats shown in FIGS.

Here, the format has a different form size than the format, the format has a different form title than the format, and the format has a different ruled line type than the format. .

First, image data (form image) to be registered is input by the image input unit 1 and stored in the image data storage memory 2 (step S201). Next, the feature detection item having the highest priority is obtained from the feature priority information storage memory 3 (step S202), and the feature detection section 4 obtains feature information of the feature detection item (step S203).

The feature information of the feature detection item having the highest priority of the image data detected by the feature detection unit 4 is stored in the feature storage memory 3.

Next, the feature information stored in the feature storage memory 3 is compared with the feature information of each format stored in the format registration memory 6 by the format classification unit 7 (step S104). By this collation, it is determined whether or not there is the same feature information as the feature information of the image data (step S205). If not, the feature information of the image data is additionally registered as new feature information in the format registration memory 6 (step S205). Step S206).

On the other hand, if there is the same feature information as that of the image data, it is determined whether or not there is a feature detection item having the next priority (step S207). Step S205 is repeatedly performed. That is, feature information is detected from the input image data for the feature detection item having the next priority,
The format classification unit 7 determines whether or not there is the same feature information as the feature information.

If the same feature information does not exist, the feature information of the image data is newly registered in the format registration memory 6. Further, feature information of image data is detected for feature detection items in all priorities, and when each feature information is already registered in the format registration memory 6, the feature detection information registered in the format registration memory 6 is detected. Since the format cannot be determined only by the feature information of the item, the user is prompted to add a new feature detection item (step S208).

Thus, the feature information of the image data can be detected in accordance with the priority order, and new feature information can be registered efficiently. That is, the feature information corresponding to the feature detection item can be preferentially registered according to the priority order.

Next, a specific example of this embodiment will be described. FIG. 8 is a diagram showing an example of image data (form) to be registered. Note that in this specific example, the priority shown in FIG. 4 is used as the priority of the feature detection items.

First, image data as shown in FIG. 8 is input, and "form size", which is the first feature, is detected. FIG.
Since the form size does not match any of the formats shown in FIG. 3, this "form size" is additionally registered in the format registration memory 6 as new feature information.

In this example, since it is possible to determine that there is no feature information matching the format only by detecting the “form size” as the first feature, only “form size” is added as new feature information of the format. I will register. Thereby, according to the priority set in advance,
New feature information corresponding to feature detection items with high importance can be quickly registered.

When the user sets a new feature detection item when registering a new format, the processes of steps S201 to S208 shown in FIG. 7 are executed for the added feature detection item. do it.

In the registration process shown in FIG. 7, if the feature information of the image data does not match the feature information of the already registered format in accordance with the priority of the feature detection item, the feature information of the image data is newly registered. The feature detection items with lower priorities are not registered. However, by repeating the registration process shown in FIG. 7 an arbitrary number of times, the feature information items in the feature detection items corresponding to the priorities for the number of times are not registered. Can be registered.

In each of the above-described embodiments, an example has been described in which classification and registration are performed on a form. However, the present invention is not limited to this, and other types of image data (including photographs and line drawings) are included. The present invention can be applied even when classifying and registering. For example, a business card composed of various forms may be read, and a company name or a department name may be classified as a feature detection item, or a business card with a photo may be classified.

[0058]

As described above, according to the image data classification method and the image data registration method of the present invention, the following effects can be obtained. That is, priorities are assigned to a plurality of feature detection items, and the feature information of the image data is detected according to the priorities, so that the format of the image data can be efficiently determined without detecting useless feature information. become. In addition, feature information of a feature detection item having a high degree of importance required for format determination can be quickly registered, and feature information can be efficiently stored with only a necessary data amount when a feature of a new format is registered. It becomes possible.

[Brief description of the drawings]

FIG. 1 is a flowchart illustrating an image data classification method according to an embodiment.

FIG. 2 is a hardware configuration diagram for realizing the image data classification method of the present embodiment.

FIG. 3 is a diagram showing an example of a stored format.

FIG. 4 is a diagram showing an example of correspondence between feature detection items and their priorities.

FIG. 5 is a diagram showing an example of a form to be classified.

FIG. 6 is a hardware configuration diagram for realizing the image data registration method of the embodiment.

FIG. 7 is a flowchart illustrating an image data registration method according to the embodiment.

FIG. 8 is a diagram showing an example of a form to be registered.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Image input part 2 Image data storage memory 3 Feature priority information storage memory 4 Feature detection part 5 Feature storage memory 6 Format registration memory 7 Format classification part 8 Format classification result storage memory 9 Format characteristic determination part 10 Format registration part

Claims (4)

[Claims] A step of inputting image data; detecting characteristic information corresponding to the characteristic detection item from the image data in accordance with a predetermined priority order of the characteristic detection item; A step of collating feature information of a format registered in advance; and a step of outputting the format when the detected feature information and the format of feature information extracted by the collation correspond one-to-one. An image data classification method, comprising: 2. When the detected feature information and the format of the feature information extracted by the collation correspond one to n, detection of feature information corresponding to a feature detection item having the next priority order and its detection are performed. The matching between the detected feature information and the feature information of the format extracted by the previous matching is repeated until the detected feature information and the format of the feature information extracted by the matching correspond one-to-one. 2. The method according to claim 1, wherein: 3. A step of inputting image data, detecting feature information corresponding to the feature detection item from the image data in accordance with a predetermined priority order of the feature detection item, and detecting the detected feature information and A step of matching feature information of a format registered in advance, and if the detected feature information is not extracted from the feature information of the format by the matching, the detected feature information corresponds to the extracted feature information. Registering new feature information of the feature detection item as new feature information. 4. When the detected feature information is extracted from the feature information of the format by the collation, detection of feature information corresponding to a feature detection item having the next priority order and detection of the detected feature information and 4. The image data registration method according to claim 3, wherein the matching with the feature information of the format extracted in the previous matching is repeatedly performed until the detected feature information is not extracted from the feature information of the format. .