JP6880460B2 - Image processing device - Google Patents

Description

The present invention relates to an image processing apparatus.

In one image processing device, a ruled line is detected based on the projection distribution in the vertical or horizontal direction of the tablespace image, and expansion / contraction processing is performed on the detected ruled line to make the ruled line a solid line (for example, a patent). Reference 1). That is, even if the ruled line in the tablespace image is a broken line or the like, the ruled line is converted into a solid line.

Japanese Unexamined Patent Publication No. 2010-182052

However, although the above-mentioned image processing apparatus converts a broken line or the like into a solid line, it is difficult to specify the type of the detected ruled line.

The present invention has been made in view of the above problems, and an object of the present invention is to obtain an image processing apparatus that specifies a type of ruled line by a relatively simple process.

The image processing apparatus according to the present invention has an edge image generation unit that detects an edge in the original image and generates an edge image, and the edge image that has been expanded by executing expansion processing on the edge image. The expansion / contraction processing unit that generates a closing image by executing the contraction processing, the rule line is specified based on the closing image, and the type of the rule line is specified based on the difference image between the closing image and the edge image. It is provided with a ruled line specifying part.

According to the present invention, an image processing apparatus for specifying a type of ruled line can be obtained by a relatively simple process.

The above or other objects, features and advantages of the present invention will be further clarified from the following detailed description along with the accompanying drawings.

FIG. 1 is a block diagram showing a configuration of an image processing device according to an embodiment of the present invention. FIG. 2 is a flowchart illustrating the operation of the image processing apparatus shown in FIG. FIG. 3 is a diagram showing an example of an original image including a table structure. FIG. 4 is a diagram showing an edge image (part) corresponding to the original image shown in FIG. FIG. 5 is a diagram showing a closing image (part) corresponding to the edge image shown in FIG. FIG. 6 is a diagram showing a difference image (part) corresponding to the edge image shown in FIG. 4 and the closing image shown in FIG. FIG. 7 is a diagram showing an image (part) obtained by performing a shrinkage process on the closing image shown in FIG.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing a configuration of an image processing device according to an embodiment of the present invention. The image processing device shown in FIG. 1 includes an image reading device 1, a storage device 2, a communication device 3, a printing device 4, and an arithmetic processing device 5.

The image reading device 1 is an internal device that optically reads a document image from a document and generates image data (here, RGB data) of the document image.

Further, the storage device 2 is a device capable of storing various data and programs. As the storage device 2, a non-volatile large-capacity storage medium such as a non-volatile memory or a hard disk drive is used. The storage device 2 can store, for example, image data of a manuscript image.

Further, the communication device 3 is a device that performs data communication with an external device. As the communication device 3, a network interface for network communication, a modem for facsimile communication, and the like are used.

Further, the printing device 4 prints the original image based on the image data of the original image that has been subjected to image processing for printing such as color conversion and halftone processing.

Further, the arithmetic processing device 5 is a computer having a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), and the like, and loads a program from the ROM, the storage device 2, and the like into the RAM, and the program is loaded. Various processing units are realized by executing the program on the CPU. In this embodiment, the arithmetic processing unit 5 realizes the reading processing unit 11 and the table detecting unit 12.

The reading processing unit 11 controls the image reading device 1 to acquire the original image, and stores the image data of the original image in the storage device 2, the RAM, or the like.

The table detection unit 12 detects the ruled lines in the original image (original image) based on the image data, and detects the table set in the original image based on the detected ruled lines.

The table detection unit 12 includes an edge image generation unit 21, an expansion / contraction processing unit 22, and a ruled line identification unit 23.

The edge image generation unit 21 detects an edge in the original image and generates an edge image. For example, the edge image generation unit 21 sets the pixel value of the pixel from the rising position to the falling position of the density as the first value (for example, 1) in the original image along the main scanning direction or the sub scanning direction. An edge image as a binary image is generated by setting the pixel values of pixels other than the above to the second value (for example, 0). The rising position and the falling position are positions where the density difference between adjacent pixels exceeds a predetermined value.

The expansion / contraction processing unit 22 executes an expansion process on the edge image, executes a contraction process (that is, a closing process) on the edge image after the expansion process, and generates a closing image.

In the expansion process, if there is a pixel whose pixel value is the first value in the range of one pixel in the main scanning direction and the sub-scanning direction for a certain pixel of interest, the pixel value of the pixel of interest is set to the first value.

In the shrinkage process, if there is a pixel whose pixel value is the second value in the range of one pixel in the main scanning direction and the sub-scanning direction for a certain pixel of interest, the pixel value of the pixel of interest is set to the second value.

The ruled line specifying unit 23 identifies the ruled line based on the above-mentioned closing image, and also executes so-called black hat processing to generate a difference image between the closing image and the edge image, and the type of the ruled line is based on the difference image. To identify. The difference image is a binary image having an exclusive OR (or its inversion value) of the pixel value of the closing image and the pixel value of the edge image at each pixel position as the pixel value.

For example, the ruled line specifying unit 23 determines that the type of the specified ruled line is a double line ruled line when a continuous linear line segment appears at the position of the specified ruled line in the difference image.

Further, when a plurality of line segments are intermittently linearly located in the difference image at the position of the specified ruled line, the ruled line specifying unit 23 determines that the type of the specified ruled line is a broken line ruled line or a chain line ruled line. To do. That is, the specified ruled line type is a type (broken line, one-dot chain line, two-dot chain line, etc.) corresponding to the arrangement pattern of the plurality of line segments.

Further, the ruled line specifying unit 23 determines that the type of the ruled line other than the ruled line determined to be any of the broken line ruled line, the chain line ruled line, and the double line ruled line among the specified ruled lines is the solid line ruled line.

Specifically, the ruled line specifying unit 23 executes a shrinking process on the closing image, and among the line segments remaining in the closing image after the shrinking process, it is determined that the type of the ruled line is a double line ruled line. The type of ruled line corresponding to a line segment other than the line segment corresponding to the ruled line is determined to be a thick line ruled line (a type of solid line segment). Then, the other types of solid line ruled lines are determined to be thin line ruled lines.

For example, the table detection unit 12 specifies one or a plurality of continuous rectangles composed of the specified ruled lines, specifies an circumscribed rectangle for the specified one or a plurality of continuous rectangles, and displays the circumscribed rectangle. Specify as an area.

Next, the operation of the image processing apparatus will be described. FIG. 2 is a flowchart illustrating the operation of the image processing apparatus shown in FIG.

When the reading processing unit 11 acquires the original image (step S1), the edge image generation unit 21 generates an edge image corresponding to the original image (step S2).

FIG. 3 is a diagram showing an example of an original image including a table structure. FIG. 4 is a diagram showing an edge image (part) corresponding to the original image shown in FIG. The table set in the original image shown in FIG. 3 includes a thick line ruled line (solid line ruled line), a double line ruled line, a broken line ruled line, a chain line ruled line, and a thin line ruled line (solid line ruled line). In FIG. 4, the upper left portion of the edge image is shown, the first value pixel described above is represented as a white pixel, and the second value pixel described above is represented as a black pixel.

Next, the expansion / contraction processing unit 22 generates a closing image corresponding to the edge image (step S3). FIG. 5 is a diagram showing a closing image (part) corresponding to the edge image shown in FIG. In FIG. 5, the first value pixel described above is represented as a white pixel, and the second value pixel described above is represented as a black pixel. For example, as shown in FIG. 5, in the closing image, a thick line segment appears at a position corresponding to a thick line ruled line and a double line ruled line, and a thin line segment appears at a position corresponding to a broken line ruled line, a chain line ruled line, and a thin line ruled line. ..

The ruled line specifying unit 23 identifies the ruled line (that is, the position of the ruled line) based on the closing image by a known method (for example, Hough transform) (step S4).

Then, the ruled line specifying unit 23 generates a difference image between the closing image and the original edge image (step S5), and in the difference image, as described above, among the types of ruled lines, the double line ruled line, the broken line, And a ruled line having any of the chain lines is specified (step S6).

FIG. 6 is a diagram showing a difference image (part) corresponding to the edge image shown in FIG. 4 and the closing image shown in FIG. In FIG. 6, the above-mentioned first value pixel is represented as a white pixel, and the above-mentioned second value pixel is represented as a black pixel. For example, as shown in FIG. 6, when a continuous line segment (a line segment 101 extending in the horizontal direction in FIG. 6) appears, the type of the ruled line specified at the position of the continuous line segment 101 is two. Identified as a double line segment. A double line ruled line having two line segments (two line segments included at the position of line segment 101) in the original image is specified as one double line ruled line. Further, when a plurality of intermittently located line segments (plurality of line segments 102 and 103 arranged in the vertical direction in FIG. 6) appear, they are specified by the positions of the plurality of line segments 102 and 103. The type of the ruled line is specified as a broken line or a chain line according to the arrangement pattern of the plurality of line segments. The type of the ruled line corresponding to the plurality of line segments 102 is determined to be a broken line, and the type of the ruled line corresponding to the plurality of line segments 103 is determined to be a chain line (dashed-dotted line).

Here, among the ruled lines specified in step S4, the ruled lines other than the ruled lines having any of the double line ruled line, the broken line, and the chain line are regarded as solid line ruled lines.

Further, the ruled line specifying unit 23 executes a shrinking process on the closing image (step S7), and among the line segments remaining in the closing image after the shrinking process, the line segment corresponding to the ruled line as the double line ruled line. The type of the ruled line corresponding to the line segment other than is determined to be a thick line ruled line (step S8), and the solid line ruled line other than the thick line ruled line is determined to be a thin line ruled line (step S9).

FIG. 7 is a diagram showing an image (part) obtained by performing a shrinkage process on the closing image shown in FIG. In FIG. 7, the above-mentioned first value pixel is represented as a white pixel, and the above-mentioned second value pixel is represented as a black pixel. For example, as shown in FIG. 7, by performing the shrinkage processing on the closing image, all the line segments corresponding to the thin line ruled lines disappear, and the line segments 111 and 112 corresponding to the thick line ruled lines and the double line ruled lines remain. .. Then, among the ruled lines corresponding to the remaining line segments 111 and 112, the ruled lines other than the double line ruled line specified as described above (that is, the ruled line corresponding to the line segment 112) (that is, the ruled line corresponding to the line segment 111) are supported. The type of ruled line to be used is specified as a thick line segment.

As described above, according to the above embodiment, the edge image generation unit 21 detects an edge in the original image and generates an edge image. The expansion / contraction processing unit 22 executes expansion processing on the edge image and performs contraction processing on the edge image after the expansion processing to generate a closing image. The ruled line specifying unit 23 specifies the ruled line based on the closing image, and specifies the type of the ruled line based on the difference image between the closing image and the edge image.

As a result, the type of ruled line can be specified by a relatively simple process.

It should be noted that various changes and modifications to the above-described embodiments will be apparent to those skilled in the art. Such changes and modifications may be made without departing from the intent and scope of the subject and without diminishing the intended benefits. That is, such changes and amendments are intended to be included in the claims.

For example, in the above embodiment, the expansion / contraction processing unit 22 may continuously perform the expansion treatment a predetermined number of times, and then execute the contraction treatment the same number of times as the expansion treatment. However, the number of expansion treatments (and the number of contraction treatments) is set to a predetermined upper limit or less. This number of times may be a fixed value or a value specified by the user. Further, this upper limit value is set to the number of times that adjacent ruled lines are not combined by the expansion process.

Further, in the above embodiment, the ruled line specifying unit 23 may continuously execute the shrinking process on the closing image a predetermined number of times.

Further, in the above embodiment, the ruled line in the original image does not have to be the ruled line in the table.

The present invention is applicable to, for example, an image forming apparatus such as a copying machine or a multifunction device.

21 Edge image generation unit 22 Expansion / contraction processing unit 23 Ruled line identification unit

Claims (5)

An edge image generator that detects edges in the original image and generates an edge image,
An expansion / contraction processing unit that executes an expansion process on the edge image and performs a contraction process on the edge image after the expansion process to generate a closing image.
A ruled line specifying unit that specifies a ruled line based on the closing image and specifies the type of the ruled line based on a difference image between the closing image and the edge image.
An image processing device comprising. When a continuous linear line segment appears in the difference image at the specified position of the ruled line, the ruled line specifying unit determines that the type of the specified ruled line is a double line ruled line. The image processing apparatus according to claim 1, wherein the image processing apparatus is characterized. When a plurality of line segments are intermittently linearly located at the specified ruled line position in the difference image, the ruled line specifying portion sets the specified ruled line type as a broken line ruled line or a chain line ruled line. The image processing apparatus according to claim 2, wherein the image processing apparatus is used for determination. In the ruled line specifying unit, among the specified ruled lines, the type of the ruled line other than the ruled line determined that the type of the ruled line is a broken line ruled line, a chain line ruled line, or a double line ruled line is defined as a solid line ruled line. The image processing apparatus according to claim 3, wherein the determination is made. The ruled line specifying portion executes a shrinking process on the closing image, and among the line segments remaining in the closing image after the shrinking process, it is determined that the type of the ruled line is a double line ruled line. The image processing apparatus according to claim 2, wherein the type of the ruled line corresponding to a line segment other than the line segment corresponding to the ruled line is determined to be a thick line ruled line.

Images