JPH11353482A - Image processing method and device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To separate an object image into the areas having optional slants by invalidating the slant image data on every other line, generating the valid line data in narrow-paper-tablet shape and returning the valid image data consisting of the valid line data to its original position though the affine transformation. SOLUTION: The image data which are temporarily stored in a frame memory 18 are turned to be slanted at a prescribed angle by a 1st affine transformer 20. A line selection means 22 invalidates one or plural lines on every other lines prescribed among the image data which are supplied from the transformer 20. The separated lines are validated by the invalidated lines, and the valid image data consist of the validated lines. A 2nd affine transformer 24 turns the valid image data outputted from the means 22 up to its original position. A labeling processor 26 gives the labels to the image data outputted from the transformer 24 in every separated area.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing technique for separating an object image into a plurality of arbitrary areas by image processing.

[0002]

2. Description of the Related Art Separating a target object image into a plurality of areas is an important technique for speeding up and simplifying subsequent necessary arithmetic processing. For example, when obtaining the size, shape, and the like of the target object, the target object image is separated into regions of an arbitrary shape in advance, and then the circumscribed coordinates of each separated region are obtained. Also, when obtaining the barycentric line of the target object image, the processing can be speeded up by performing various processing after separating the image into a plurality of regions.

[0003]

Generally, when evaluating arbitrary regions of an object having various shapes, graphic data obtained by combining fan-shaped data having various radii of curvature with various rectangular data is used. Approximation processing. However, the extraction of an arbitrary region and its evaluation increase the load due to the system configuration. Further, the obtained approximate value was not accurate enough to be put to practical use.

[0004] Japanese Patent Application Laid-Open No. Hei 9-114989 discloses an image processing apparatus for separating a target object image into a region horizontal to a scanning line. According to this device, the target object image can be separated into a plurality of regions at a high speed, and by processing data for each of the separated regions, the center of gravity of the target object can be measured at a high speed. However, depending on the feature quantity to be measured, such as the shape and size of the target object image, it is not sufficient to separate only the area parallel to the scanning line.

Accordingly, it is a first object of the present invention to provide an image processing method capable of separating a target object image into a region having an arbitrary inclination.

A second object of the present invention is to provide an image processing apparatus capable of separating a target object image into a region having an arbitrary inclination.

[0007]

In order to achieve the above object, in an image processing method according to a first aspect of the present invention, target analog image data is converted into digital image data; The digital image data binarized by the affine transformation is tilted at an arbitrary angle on the frame; and the slanted image data is invalidated at a fixed line interval, whereby strip-shaped effective line data is converted. Generating; returning the effective image data composed of the effective line data to the original position by affine transformation, thereby separating the digital image data into slanted strip regions.

An image processing apparatus according to a second aspect of the present invention includes an A / D converter for converting target analog image data into digital image data; and a binarization process for binarizing the digital image data. A frame memory for storing the binarized image data; a first affine converter for inclining digital image data stored in the frame memory at an arbitrary angle on a frame by affine transformation; A line selecting means for generating strip-shaped effective line data by invalidating the inclined image data at every fixed line; and returning the effective image data composed of the effective line data to the original position by affine transformation A second affine converter for separating the digital image data into slanted strip regions.

As described above, according to the present invention, after the digital image data is inclined at an arbitrary angle, for example,
The image is separated in a direction parallel to the scanning line. Thereafter, the separated image data is returned to the original position. Thereby, the target object image can be separated into a region inclined at an arbitrary angle from the scanning line. As a result, evaluation of various shaped objects,
Measurement can be performed easily and at high speed.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described in detail with reference to the following examples. In this embodiment, the present invention is applied to the calculation of the barycentric line of the target object image.

[0011]

FIG. 1 shows the configuration of an image processing system 10 according to an embodiment of the present invention. This image processing system 10
A television camera 12 for capturing an object image to be processed, an A / D converter 14 connected to an output terminal of the television camera 12, and an A / D converter 14 connected to an output terminal of the A / D converter 14.
A binarization processor 16, a frame memory 18 connected to an output terminal of the binarization processor 16, a first affine converter 20 connected to an output terminal of the frame memory 18,
And a second affine converter 24 connected to the output terminal of the line selecting means 22 and an output terminal of the second affine converter 24 connected to the output terminal of the affine converter 20. Labeling processor 26
And a feature amount extractor 28 connected to the output terminal of the labeling processor 26.

As a means for capturing an object image, a device other than the television camera 12 can be used. A / D
The converter 14 converts analog image data captured by the television camera 12 into digital image data. The binarization processor 16 binarizes the digital image data output from the A / D converter 14. The binarized data output from the binarization processor 16 is stored in the frame memory 18.
The image data temporarily stored in the frame memory 18 is rotated by the first affine converter 20 so as to be inclined by a predetermined angle.

The line selecting means 22 invalidates one or more lines every predetermined number of lines of the image data supplied from the affine converter 20, and activates the lines separated by the invalidated lines. It is said.
The valid lines form valid image data. The second affine converter 24 is connected to the line selecting unit 22.
Is rotated to the original position.

The labeling processor 26 labels the image data output from the affine transformer 24 for each separation area. The area data subjected to the labeling process is supplied to the feature amount extractor 28, and a predetermined feature amount is extracted for each area. As shown in FIG. 2, for example, the feature quantity extractor 28 includes a circumscribed rectangular coordinate measuring device 28a and a centroid calculator 28b.
And can be composed of Bounding rectangular coordinate measuring instrument 28
a is the circumscribed rectangle 4 of each area data with a label
Measure the coordinates of the point. The center-of-gravity calculator 28b calculates the center-of-gravity coordinates of each area based on the coordinates (four points) of each area data.

Next, the operation of calculating the center of gravity of the image processing system configured as described above will be described with reference to FIGS. First, before starting the process, data of each unit in the system is initialized (S1). next,
The analog signal of the target object image captured by the television camera 12 is supplied to the A / D converter 14 and converted into a digital signal (S2).

The image data converted into a digital signal is
The image is binarized by the binarization processor 16 (S3). The binarized data is temporarily stored in the frame memory 18 (S4). The image data stored in the frame memory 18 is rotated by the first affine converter 20 so as to be inclined by a predetermined angle (S5) (see FIG. 4). The affine-transformed image data is supplied to the line selection unit 22. The line selecting means 22 fetches the line data (S6), and when determining that the pixel is a line to be invalidated, replaces the signal of the pixel with an invalid signal (S7). Perform this operation until the last line,
At the point in time when the capture of the last line is completed, the effective image data (see FIG. 4) constituted by the effective signal is rotated to the original position by the second affine converter 24 (S
8) (see FIG. 4).

The output image data of the second affine converter 24 is subjected to a labeling process by a labeling processor 26 (S9). As a result, the surfaces constituting the target object image are separated. With respect to the image data labeled by the labeling processor 26, the coordinates of four points of the circumscribed rectangle are obtained for each region by the circumscribed rectangle coordinate measuring device 28a (S10). Thus, the position of the surface forming the target object image is detected.

The center-of-gravity calculator 28b calculates the coordinates of the center of gravity (the center of the circumscribed rectangle) of the surface detected as described above (S11) and outputs it (S11). Perform the above operation 1
By repeating in a pipeline for the number of frames, the barycentric line of the target object image in which the barycenters obtained by the above-described operation are connected is calculated. In this embodiment, the target object image is 1
Although two cases have been described, the same processing can be performed even when there are a plurality of target object images.

[0019]

As described above, according to the present invention, there is an effect that the target object image is divided into regions inclined at an arbitrary angle, and various feature amounts of the image can be measured easily and at high speed.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of an image processing system according to an embodiment of the present invention.

FIG. 2 is a block diagram showing a configuration of a main part of the embodiment.

FIG. 3 is a flowchart illustrating an overall operation of the embodiment.

FIG. 4 is an explanatory diagram visually (imaging) the operation of the embodiment.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 Image processing system 12 Television camera (imaging means) 14 A / D converter 16 Binary processing unit 18 Frame memory 20 First affine converter 22 Line selecting means 24 Second affine converter 26 Labeling processor 28 Features Quantity extractor 28a circumscribed rectangular coordinate measuring instrument 28b center of gravity calculator

Claims (4)

[Claims] A step of converting the target analog image data into digital image data; a step of binarizing the digital image data; and a step of converting the binarized digital image data by affine transformation into an arbitrary angle on a frame. Generating a strip-shaped effective line data by invalidating the skewed image data at regular intervals; and performing affine transformation on the effective image data composed of the effective line data. Separating the digital image data into slanted strip-shaped regions by returning to the original position. 2. A step of performing a labeling process on each of the separated areas after returning the effective image data to an original position by affine transformation; and calculating a center of gravity of each of the labeled areas. 2. The image processing method according to claim 1, further comprising: calculating a barycentric line of the target image data from a barycentric position of each region. 3. An A / D converter for converting target analog image data into digital image data; a binarization processor for binarizing the digital image data; and storing the binarized image data. A frame memory; a first affine converter for tilting the digital image data stored in the frame memory at an arbitrary angle on the frame by affine transformation; and invalidating the tilted image data at regular intervals. A line selecting means for generating strip-shaped effective line data; and returning the effective image data composed of the effective line data to the original position by affine transformation, thereby converting the digital image data into an inclined strip region. An image processing apparatus, comprising: a second affine converter for separating. 4. A labeling processor for performing a labeling process for each of the separated areas after returning the effective image data to the original position by the second affine transformer; and each of the labeled areas. 4. A center-of-gravity calculator for calculating a center of gravity of each area, wherein a center-of-gravity line of the target image data is calculated from a center of gravity of each area.
An image processing apparatus according to claim 1.