JPH04142676A - Binary image conversion method - Google Patents

Abstract

PURPOSE:To easily extract an object body image area by setting both the outer image area of a window set into a binary image and the body image area in the window that intersects the boundary of the window at image data of the same value to convert the image data. CONSTITUTION:Window 9 is set in input binary image, the image area a2 outside this window 9 and body image areas a3 and a4 in the window that intersect the boundary of the window are set at image data of the same value to convert image area to binary image. Accordingly, even if image areas a3 and a4 partly appear in window 9 in addition to object body image area a1, the image area a3 and a4 and image area a2 outside the window 9 are set at image data of the same value. With this, the object body image area can be extracted easily discriminating it from other image areas.

Description

[Detailed Description of the Invention] <Industrial Application Field> This invention sets a window in an input binary image and executes processing such as measuring a characteristic quantity for a specific object image area within the window. The present invention relates to an image processing apparatus for processing an image, and more particularly, the present invention relates to a binary image conversion method for converting the binary image into a new binary image convenient for performing processing aimed at.

<Prior Art> For example, when inspecting an electronic component mounted on a printed circuit board, an image of the electronic component to be inspected is captured by a camera, and the grayscale image is taken into an image processing device and binarized. Thereafter, a window is set at a predetermined portion of the binary image, and the object image area to be inspected included within the window is detected to extract feature quantities such as area per perimeter.

In this case, the window is set sufficiently large to take into account misalignment of the electronic component, so not only the image of the electronic component to be inspected but also images of surrounding electronic components may partially appear within the window. . The binary image within such a window is subjected to labeling processing, and the area of the object image area of each label is measured and compared with a threshold value to determine whether it is the object image area to be inspected. ing.

<Problems to be Solved by the Invention> However, in such a method, each process such as labeling and region discrimination becomes complicated and requires a lot of processing time.

This invention was made with attention to the above problem, and even if an image area other than the target object image area partially appears in the window, the target object image area is distinguished from other image areas. It is an object of the present invention to provide a binary image conversion method that allows easy extraction.

<Means for Solving the Problems> The present invention is a binary image conversion method for inputting a binary image and converting each pixel data, the method comprising: setting a window on the input binary image; The image area outside this window and the object image area within the window that intersects with the boundary of the window are set to pixel data of the same value to convert the binary image.

<Operation> Even if an image area other than the target object image area partially appears in the window, that image area and the image area outside the window are set to the same pixel data, so the target object image area is set to the same value pixel data. Image regions can be easily extracted by distinguishing them from other image regions.

<Embodiment> FIG. 1 shows an example of the configuration of an image processing system in which the present invention is implemented, and includes a camera 1 and an image processing device 2. As shown in FIG.

This image processing system is applied, for example, to systems that automatically inspect electronic components on printed circuit boards.
A camera 1 images an object 3, and the image signal is taken into an image processing device 2.

The image processing device 2 includes a binarization circuit 42, an image memory 5.
Conversion circuit 6. Labeling circuit 7. It includes a measurement circuit 8 and the like.

The binarization circuit 4 binarizes the gray scale image input from the camera 1 using a predetermined threshold value, and stores the binarized image in the image memory 5.

FIG. 3 (1) shows a binary image A stored in the image memory 5.
An example is shown below. In the figure, the shaded areas a to a4 are object image regions, and each pixel is given binary data of "1", for example. Further, the white portion of b is a background area, and each pixel is given binary data of, for example, "0".

The conversion circuit 6 sets a window 9 of a predetermined size at a predetermined position within the binary image A shown in FIG. 3(1). In the example shown in FIG. 3(1), the target object image area a1 located at the center is included within the window 9, and the non-target object image area a2 is located outside the window 9. In addition, each object image area as+ai that is not the target at the end position
intersect with the boundary of window 9, respectively, and the object image area a! A portion of l l a 4 appears in window 9 .

This conversion circuit 6 performs a binary data conversion process on the binary image A for which the window 9 has been set, thereby converting the binary image A into a new binary image B as shown in FIG. 3(3). Convert to That is, the image area outside the window 9 and the window 9 intersecting the border of the window 9
The pixel data of the same value (
In this case, it is set to "1") to convert the binary image from A to B.

The labeling circuit 7 performs labeling processing on the binary image B shown in FIG.
”, respectively. Measurement circuit 8 is labeled “1”
Feature quantities such as area 2 and circumference length are measured for the object image region a1 labeled with .

FIG. 2 shows a specific example of the conversion circuit 6, which includes two OR circuits 10. II and three AND circuits 12~・1
4 and two knot circuits 15° and 16.

Further, in the figure, VBI is the binary image signal of FIG. MODE is a mode signal for setting the conversion mode, and 0N10FF is a window setting signal for turning on/off the setting of window 9.

Now, considering the case where the conversion process is performed on the binary image A shown in FIG. 3(1), first the mode signal MODE and the window setting signal 0N10FF are both set to "1". As a result, the window signal WD is applied as one input to the AND circuit 13 via the AND circuit 12 and the OR circuit 10. This AND circuit 13 includes 2 in FIG. 3(1).
The binary image signal VBi constituting the value image 8 is given as the other input, and as a result, a binary image A' as shown in FIG. 3(2) is obtained as the output of the acid circuit 13. In this binary image 8', the pixel data of each pixel constituting the area outside the window 9 is all "0".

On the other hand, the mode signal MODE, the window setting signal 0N10FF, and the inverted signal of the window signal WD are inputted to the AND circuit 14, and as a result, the entire area inside the window 9 is output as pixel data of "0" , and the entire area outside window 9 is "1"
It consists of pixel data.

The outputs of each AND circuit 13 and 14 are input to the OR circuit 11, and as a result, the third
A binary image B as shown in Figure (3) is obtained. Furthermore, the fourth
The figure shows the relationship between the input and output of the conversion circuit 6 in the form of a logic table.

In the binary image B, the area composed of pixel data rN is the central object image area a.

and the data conversion area C1, that is, the image area outside the window 9, and the object image area a s + 84 within the window 9 that intersects with the window 9, and the data conversion area C is always at the origin position 0 of the binary image B. will be included. Therefore, when labeling processing is applied to this binary image B, it can be easily determined that the image area of the label that does not include the origin position O is the target object image area a1, and the feature amount for this object image area a. will be measured.

<Effects of the Invention> As described above, the present invention sets a window in the input binary image, and divides the image area outside this window and the object image area within the window that intersects the boundary of the window into pixels of the same value. data to convert the binary image, so even if an image area other than the target object image area partially appears in the window, the target object image area can be distinguished from other image areas. It is possible to separate and easily extract the data, and it is possible to easily perform various processing such as labeling and region discrimination, thereby speeding up the processing.

[Brief Description of the Drawings] Fig. 1 is a block diagram showing the configuration of an image processing system according to the present invention, Fig. 2 is an electric circuit diagram showing a specific example of a conversion circuit, and Fig. 3 is a block diagram showing the configuration of an image processing system according to the present invention. FIG. 4 is an explanatory diagram showing the value image conversion method, and FIG. 4 is an explanatory diagram showing the input/output relationship of the conversion circuit. 1...Camera 2...Image processing device 6...
... Conversion circuit patent applicant Omron Co., Ltd.

Claims (1)

[Claims] A binary image conversion method for inputting a binary image and converting each pixel data, the method comprising: setting a window on the input binary image; A binary image conversion method comprising converting a binary image by setting pixel data of the same value to an object image area within the window that intersects with the boundary of the window.