JPH05108865A - Picture processing method and device thereof - Google Patents

Abstract

(57) [Summary] (Correction) [Purpose] To provide a method of storing conditions such as designation of an area when performing character recognition by image processing. [Arrangement] A document for character recognition is set on a scanner, and when image information is input, it is displayed on a CRT as binary image data, for example. Step 1.1. The character recognition frame designation step 1.2 is performed using the input device to designate the processing range. The coordinate information of the recognition frame is sent to the character recognition table.
When there are a plurality of designated recognition frames, the input device is used to determine whether the recognition mode is set for each of the designated N recognition frames or whether the recognition mode is common to all the recognition frames. Give instructions. If it is determined in step 1.3 that the individual setting mode is set, an instruction is issued to set the recognition mode for each frame, and if it is determined that the collective setting mode is set, the setting condition is set as a mode common to all the frames, Store in memory. When various settings are completed, "recognition execution" is instructed by the input device from the displayed menu. Step 1.
6.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing method and apparatus capable of recognizing characters by designating a plurality of portions as recognition target portions for input image information.

[0002]

2. Description of the Related Art Conventionally, when performing character recognition on input image information, there has been a method for designating whether or not to perform word matching on all designated areas to be recognized.

[0003]

In the prior art, once the word matching execution mode is set, word matching is performed on all designated recognition frames. However, the number of words is huge and using a word dictionary requires considerable memory. Further, it takes a considerable time to refer to the word of the character of interest from among them and compare and select it with the generated word. Therefore, when the operator designates a plurality of recognition frames, it is possible to avoid wasting memory and time if a mode can be selected whether or not to perform word matching for each frame.

[0004]

In order to solve the above-mentioned problems, the present invention inputs image information, specifies an area in which character recognition is performed in the input image information, and specifies in each of the specified areas. Provided is an image processing method characterized by storing conditions for performing character recognition correspondingly.

In order to solve the above problems, the present invention provides an input means for inputting image information, an area designating means for designating an area for character recognition in the image information input by the input means, and the area designating. There is provided an image processing device, comprising: a storage unit that stores a condition for performing character recognition corresponding to each area designated by the unit.

In order to solve the above problems, preferably, the condition for performing the character recognition is whether or not word matching is performed.

[0007]

FIG. 3 is a block diagram showing the arrangement of the image processing apparatus of this embodiment.

In FIG. 3, reference numeral 3.1 is an input device for instructing various information and functions such as a keyboard and pointing device (PD), and instructing coordinates, and 3.2 is a character from image information to be recognized. Cut out and draw the features,
It is a character recognition device that derives a character string as a recognition result by matching with a recognition dictionary. Reference numeral 3.3 denotes a CPU that controls the entire apparatus, and determines and controls the processing shown in the flowchart of FIG. 1, for example, according to a program stored in the memory 3.5. Reference numeral 3.4 is a scanner for inputting, for example, binary image data from a document, and 3.5 stores a program for controlling this apparatus, and stores input data and data for displaying on a display unit. A program for controlling the processing of the flowchart of FIG. 1 is also stored in this memory 3.5. Reference numeral 3.6 denotes a display unit such as a CRT, which displays image data of input images and recognition results.

This embodiment will be described with reference to the flowchart showing the processing of this embodiment in FIG.

First, a document for character recognition is scanned by the scanner 3.
4 and input the image information. The input image information is displayed on the CRT 3.6 as binary image data, for example (step 1.1). On the displayed image data, the range (recognition frame) in which character recognition processing is performed is designated using PD 3.1 (step 1.2). The coordinate information of the designated recognition frame is sent to the character recognition device. If all the recognition frames have been designated in step 1.2 and there are a plurality of designated recognition frames, does the operator set the recognition mode for each of the designated N recognition frames for each frame? The input device 3.1 is used to instruct whether or not the common mode is set. For the instruction, the "individual setting mode" icon displayed on the CRT 3.6 as the main menu is displayed on the PD3.
1 and click PD3.1 on the target recognition frame, a window of "individual setting mode" as shown in FIG. 2 is displayed on the CRT 3.6, and word matching is performed in that frame. Various settings such as whether or not to select a dictionary used when performing word matching can be performed (step 1.3).
And step 1.4). Information on whether the setting mode is the individual setting mode or the collective setting mode in step 1.3 and the setting condition of each setting mode are stored in the memory 3.5.

If it is determined in step 1.3 that the mode is the collective setting mode, the setting conditions are set as a mode common to all frames and stored in the memory 3.5.

When the various settings are completed, the character recognition process is performed by instructing the "recognition execution" icon from the main menu displayed on the CRT 3.6 with the input device 3.1 (step 1.6).

The recognition result is stored in the memory 3.5 and the CR
Displayed at T3.6.

In the present embodiment, as shown in FIG. 2, the presence or absence of word matching can be determined for each frame, and at the same time, the word dictionary table used for the matching can be selected. Here, among the plurality of word dictionary tables, if the prepared one is the basic dictionary and the one registered by the user himself is the user dictionary, as shown in FIG. 2, the operator selects a dictionary for each recognition frame. You can select a combination. For example, if the sentence you want to match a word is specialized and there are many words that are not in the basic dictionary, or if you want to repeat a similar sentence repeatedly, you can use the user dictionary only. Only the dictionary is enough. If the two are properly mixed, the user & basic dictionary may be used. This makes it possible to use the memory more efficiently and reduce unnecessary collation errors.

[0015]

As described above, according to the present invention,
When multiple recognition frames are specified in one image information, the recognition mode can be set for each frame, so it can be set according to the characteristics of each recognition frame, unnecessary processing can be omitted, and processing efficiency can be improved. There is.

[Brief description of drawings]

FIG. 1 is a flowchart showing the overall processing of the present invention.

[Figure 2] Screen display example of individual settings

FIG. 3 is a simplified system configuration for implementing the present invention.

Claims (4)

[Claims] 1. Inputting image information, designating an area for character recognition in the inputted image information, and storing conditions for character recognition corresponding to each of the designated areas. Image processing method. 2. The image processing method according to claim 1, wherein the condition is whether to perform word matching. 3. An input means for inputting image information, an area designating means for designating an area in which character recognition is performed in the image information input by the input means, and each area corresponding to each area designated by the area designating means. And a storage unit that stores conditions for character recognition. 4. The condition stored in the storage means is whether or not to perform word matching.
The image processing device according to.