JPH02187883A - Document reader - Google Patents

Abstract

PURPOSE:To obtain a character reader to input the area of a character string to be read by deciding a black area group to be read and a reading sequence from an indicated black area group and its indicated sequence. CONSTITUTION:A display means 11 displays the circumscribed rectangle group of black areas detected in a black area detecting means 2 on a display device such as a display. An operator indicates rectangles corresponding to black areas to be read out of the rectangle group displayed on the display device by using the indicating device of an indicating means 12. In a read area deciding means 13, all the rectangles existing between the rectangles indicated by the operator are detected as the rectangles corresponding to the areas to be read. Further, the means 11 displays a picture in which the detected rectangle group is reversely displayed, on the display device. In the means 13, the black area group corresponding to the reversely displayed rectangle group is made into an area group to be read, and the read object black areas existing from a reading start black area in the direction of a reading completion black area are successively transferred to a character segmenting and recognizing means.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention optically scans a document and cuts out characters in a character string from a document image obtained by photoelectric conversion. The present invention relates to a document reading device, and particularly relates to character string extraction processing.

[Prior art] Figure 7 shows, for example, "Character string extraction from printed documents" (Proceedings of the 1986 Institute of Electronics, Information and Communication Engineers Autumn National Conference D-20)
2) is a block diagram showing a conventional document reading device as described in 2).

In the figure, 1 is a document image input means for optically scanning a document printed on paper or the like and photoelectrically converting it to obtain a document image;
2 scans the document image obtained by the document image input means 1;
A black area detecting means 3 detects an area consisting of connected components of black pixels as a black area, and a black area detecting means 3 scans the black area group obtained by the black area detecting means 2 and detects a section consisting of a group of text character strings in the same column. A crotch area detection means for detecting the region; 4 is a text string detection means for detecting the area of the text string in each row area obtained by the crotch area detection means 3; 5 is a text string detection means for detecting the text string area detected by the text string detection means 4 This is a character extraction/recognition means that extracts and recognizes individual characters in each body character string area. The document image input means 1 is realized by an image scanner, and the black area detection means 2 to the character cutting/recognition means 5 are realized by a computer system equipped with an image memory.

The diagrams shown in FIGS. 8 to 11 are diagrams showing specific operational examples of the conventional document reading device shown in FIG. 7.

6 in FIG. 8 is a document image input by the document image input means 1, 7 in FIG. 9 is a rectangle group obtained from the black area group obtained by the black area detection means 2, and 8 in FIG. 10 is a crotch area. The crotch area obtained by the detection means 3, 9 in FIG. 11, is a group of rectangles determined from the text string obtained by the text string detection means 4.

10 in FIG. 12 is a document image 6 that is displayed on a display device such as a display, and is readable by specifying two points, the upper right end and the lower left end, of the reading range using a pointing device such as a mouse. This is a screen of a display device displaying the results of partially specifying a range.

Next, the operation will be explained.

The document image input means 1 optically scans the document and performs photoelectric conversion to obtain a binary document image 6 in which the text portion is black pixels and the background portion is white pixels.

The black area detection means 2 performs image processing such as labeling on the manually generated binary document image 6.
A black area consisting of connected components of black pixels is detected, and a rectangle group 7 is obtained by finding a circumscribing rectangle of each detected black area. At this time, ruled lines are also removed.

The crotch area detecting means 3 scans the rectangle group 7 obtained by the black area detecting means 2, and determines whether the width of the rectangle is close to the width of the text string to be read and the upper and lower ends of the rectangles are adjacent to each other. If the upper and lower ends of the rectangles are approximately equal to each other, a region obtained by combining the black regions corresponding to these rectangles is detected as the crotch region 8.

The text character string detection means 4 scans the rectangle group 7 obtained by the black area detection means 2 for each crotch region 8 obtained by the crotch region detection means 3, and corresponds to the rectangle group 9 determined to be the main text string. The black areas are arranged in J-1 order.

In the character segmentation/recognition means 5, the text character string detection means 4
Individual character patterns are extracted and recognized for each black area that is determined to be a main text string.

[Problems to be Solved by the Invention] Conventional document reading devices are configured as described above, so if the crotch area detection means or body character string detection means incorrectly detects the crotch area or body character string, it is difficult to correct the crotch area or body character string. There was a problem that it was time-consuming. For example, if a part of a heading or figure is incorrectly determined to be a text string, or if a text string cannot be detected, it is necessary to correct all characters within the specified reading range, and the recognition results are corrected. There was a problem in that it took a lot of time to delete and insert characters.

Furthermore, nil Ba of all detected text string areas
Since the results are output, there is a problem in that if only a part of the results is needed, the unnecessary parts must be deleted. One way to deal with this is to display the document image on a display device, specify the reading range, and process only the document image within the specified range, as shown in Figure 12. Since it is necessary to indicate an accurate position on the document image, it is necessary to use a display device with a high resolution, which poses a problem in that the device becomes expensive.

The present invention has been made to solve the above-mentioned problems, and an object of the present invention is to provide a document reading device that can easily input the region of a character string to be read.

[Means for Solving the Problem] A document reading device according to the present invention includes a display means for displaying a group of black areas detected by a black area detecting means, and a black area to be read from among the displayed group of black areas. an instruction means for instructing;
The apparatus includes reading area determining means for determining the black area group to be read and the reading order from the designated black area group and the designated order.

[Function] The document reading device according to the present invention displays the black area group obtained by the black area detection means on a display device such as a display,
By using a pointing device such as a mouse to indicate a black area to be read from among a group of black areas displayed on a display device, the operator can easily determine the black area to be read, that is, the character string.

[Example] An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block configuration diagram showing a document reading device according to an embodiment, and the same or equivalent parts as in the conventional example shown in FIG.

In the figure, reference numeral 11 displays a circumscribing rectangle of the black area detected by the black area detection means 2 on a display device such as a display, and 12 indicates a black area to be read from among the rectangles displayed by the display means 11. Instructing means for instructing a corresponding rectangle with an instructing device such as a mouse, and reading area determining means 13 for determining the black area to be read and the reading order based on the group of rectangles indicated by the instructing means 12 and the instruction order thereof. . Note that the display means 11. The instruction means 12 and the reading area determination means 13 are realized by a computer system equipped with a display device such as a display and an instruction device such as a mouse.

The diagrams shown in FIGS. 2 to 6 are diagrams showing specific operational examples of the document reading device shown in FIG. 1.

7 in FIG. 2 is a group of rectangles in which circumscribed rectangles of the black area detected by the black area detection means 2 are displayed on the display device by the display means 11;
3, the operator uses the instruction means 12 to specify a rectangle corresponding to the black area to start reading from among the rectangle group 7 displayed on the display device 2, and the display means 11 displays the specified rectangle in reverse. 4, the operator uses the instruction means 12 to specify a rectangle corresponding to the black area to be read from among the rectangles displayed on the two display devices, and the display means 11 displays the specified rectangle. Highlighted screen, 5th
16 in the figure shows that the reading area determining means 13 determines a group of black areas to be read based on the operator's instructions on the screen 14 and the screen 15, and the display means 11 displays a display device corresponding to the black area determined as the reading area. The double-sided display of the displayed rectangle inverted, 17 in FIG.

Next, the operation will be explained.

The display means 11 displays the circumscribed rectangle group 7 of the black area detected by the black area detection means 2 on a display device such as a display.

The operator uses the pointing device of the pointing means 12 to designate a rectangle that corresponds to the black area to be read out of the rectangular group 7 displayed on the display device. In this instruction method, first, when the operator uses the instruction means 12 to specify a rectangle displayed on the display device corresponding to the black area to start reading, the display means 11 displays a screen 14 that displays the specified rectangle in reverse video. Display on a display device. Next, the operator points out a rectangle displayed on the display device corresponding to the black area to be read by the instruction means 1.
When an instruction is given using the instruction device No. 2, the display means 11 displays a screen 15 in which the instructed rectangle is displayed in reverse on the display device. These instructions are given by pointing an arbitrary point within the rectangle to be pointed out using a mouse or the like. Thereafter, the reading area determining means 13 detects all rectangles existing between the rectangles designated by the operator as rectangles corresponding to the reading area. Then, the display means 11 displays a screen 16 in which the detected rectangle group is displayed in reverse on the display device. The reading area determining means 13 uses the black area group corresponding to the inverted rectangle group 17 as the reading area group, and extracts characters from the black area to be read in the direction from the reading start black area to the reading end black area.
The information is sequentially transferred to the recognition means 5.

In the above embodiment, the instruction means specifies the reading area within the reading range by specifying two rectangles, but it may also specify the rectangles corresponding to the reading area one by one according to the reading order. .

Further, although the display means displays only the circumscribed rectangle of the black area detected by the black area detection means, document images within the circumscribed rectangle or within the black area may be displayed in a superimposed manner.

Furthermore, the black area detection means may be any means that can detect areas where black pixels are densely packed on the document image, such as means for determining the black area using peripheral distribution values of black pixels obtained by scanning the document image. Well-known means may also be used.

[Effects of the Invention] As described above, according to the present invention, the display means, the instruction means, and the reading area determining means are provided after the black area detection means, and the black area group obtained by the black area detection means is displayed. Since the reading area is determined by specifying the black area to be read from among the displayed black area group, it is possible to partially and easily specify the reading area.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing a document reading device according to an embodiment of the present invention, FIGS. 2 to 6 are diagrams showing an example of the operation of the embodiment, and FIG. 7 is a block diagram showing a conventional document reading device. 8 to 12 are block diagrams showing an example of the operation of a conventional document reading device. 1 is a document image input means, 2 is a black area detection means, 5 is a character cutting/recognition means, 11 is a display means, 12 is an instruction means, and 13 is a reading area determination means.

Claims (1)

[Claims] By optically scanning a document and photoelectrically converting it into a document image, areas where black pixels are concentrated are detected as black areas, and characters in a character string are extracted and recognized from the detected black areas. In a document reading device that reads a document, a display means for displaying a detected black area group, an instruction means for indicating a black area to be read from among the displayed black area group;
1. A document reading device comprising a reading area determining means for determining a black area group to be read and a reading order based on a designated black area group and the designated order.