KR980013461A - Adjustment device of adjustment parts and method of detecting flaws - Google Patents

Abstract

An image of a component to be adjusted is acquired from a CCD camera to perform median filtering, smoothing the image to remove noise, performing a binary image process, and performing an opening process A labeling process is performed in which a connecting portion is removed from an image of a part through a plurality of blob regions and a blob region is numbered in order, and the outline of the blob region is detected, a moment is calculated in each pixel of the detected outline, The center of gravity of each blob image is found by finding two similar blob images, and the center of gravity (C) is calculated by using the calculated center of gravity and set as the center of the adjusted blob.

Description

Adjustment device of adjustment parts and method of detecting flaws

[0001] The present invention relates to an adjusting device for adjusting parts and an adjusting flaw detecting method, and more particularly to a adjusting device for adjusting parts of an adjusting part capable of accurately capturing an image of a part by using a vision system, And an adjusting groove detecting method.

It is equipped with functions such as human vision to give a high degree of adaptability to industrial robots and to perform complex tasks. For example, a device such as a printed circuit board tester may have a vision system to adjust the adjustment part to inspect the circuit for failure or the like.

When adjusting a conventional adjusting part, as shown in FIG. 1 (a), since it is a component which is not sensitive to the angle of the adjusting bit, the center detecting is easy by the template matching method and insertion into the adjusting piece with the adjusting bit is easy.

However, as shown in FIG. 1 (b), in the case of an image having a sharp angle, a templating matching method can not be applied and a new angle detecting method is required. When adjusting an adjusting part, There is a risk of loss of time and damage to the part during rotation of the adjustment bit, because the adjustment bit must be rotated until the adjustment bit is inserted into the adjustment defect unless the position is correctly adjusted and inserted into the adjustment defect.

In addition, when the parts are assembled, since the position is not uniform due to self-insertion by the automatic inserter, the color of each part is different, and therefore there are problems such as error in image processing due to noise image due to the surrounding environment .

According to an aspect of the present invention, there is provided an image processing apparatus including a camera unit for capturing an image of an object to be adjusted and acquiring and outputting a video signal, an image processor for processing and outputting a video signal input from the camera unit, And removing the connection portion from the image of the component through the opening process, removing a blob image near the corner where the blob region is numbered in order, and the blob region , Calculates the moment at each pixel of the detected outline, finds two blob images having similar moments, calculates the center of gravity of each blob image, and calculates the center of the adjusted blob using the calculated center of gravity And outputting a control signal for detecting and adjusting the component to be adjusted; and a controller An XY table that receives a control signal and moves in the X and Y axis directions; and an adjustment unit that is mounted on the XY table and receives a control signal from the control unit and inserts and adjusts an adjustment bit in the adjustment piece of the adjustment target component do.

A method according to the present invention includes the steps of acquiring an image of a part to be adjusted from a camera and performing median filtering, removing noise by smoothing the image, performing binarization image processing, And performing a labeling process of attaching numbers to the blob areas in order, detecting an outline of the blob area, calculating moments in each pixel of the detected outline, Calculating two blob images, calculating the center of gravity of each blob image, calculating a center point C using the calculated center of gravity, and setting the center point as the center of the adjusted blob.

Figures 1 (a) and (b) show the form of adjustment fits on the part.

FIG. 2 is a block diagram showing a configuration of an adjusting device for adjusting parts according to the present invention.

FIG. 3 is a flowchart showing an adjusting flaw detecting method of an adjusting part according to the present invention.

4 (a) - (f) are image processing states in each step for performing the method according to the present invention.

DESCRIPTION OF THE REFERENCE NUMERALS

1: control unit 2: image processing unit

3: X-Y Table 4: Camera section

5: Adjustment means

Hereinafter, the present invention will be described in detail with reference to the drawings.

FIG. 2 is a block diagram showing a configuration of an apparatus for detecting an adjustment of an adjustment part according to the present invention.

The apparatus according to the present invention includes a CCD camera unit 4 for capturing an image of an adjustment object 61 and acquiring and outputting a video signal, an image processing unit for processing and outputting a video signal input from the CCD camera unit 4 (2), a binary image processing is performed by receiving the video signal outputted from the image processing unit (2), the connection part is removed from the image of the part through the opening process, and the edge The near blob image is removed, the outline of the blob area is detected, the moment is calculated at each pixel of the detected outline, two blob images having similar moment values are found to find the center of gravity of each blob image, A control unit (1) for outputting a control signal for detecting the center of the adjustment flaw and adjusting the adjustment target component by using a center of gravity; An XY table 3 that receives input and moves in the X and Y axis directions, and an XY table 3 that is mounted on the XY table 3 and receives a control signal from the control unit 1, And adjusting means 5 for inserting and adjusting the bit 7.

FIG. 3 is a flowchart showing an adjusting flaw detecting method of an adjusting part according to the present invention. 4 (a) - (f) show the image processing state in each step for performing the method according to the present invention.

In step 301, an image of the part to be adjusted is obtained from the CCD camera of the vision system. In step 302, due to the poor surrounding environment, the moving image includes a lot of noise. Therefore, in order to remove the noise of the image, median filtering is performed to smooth the image to remove the noise. The state of the image from which noises are removed by median filtering is shown in FIG. 4 (a).

The background is set to "0 ", and the meaningful area is set to" 1 "by setting the data representing the brightness of each pixel to a value larger than the threshold value at" 1 " Processing is performed. An image obtained by performing the binarization image processing is shown in FIG. 4 (b). As shown in the figure, when the binarized image processing is performed, pixels having luminance lower than the threshold value luminance value are centered on the threshold value, and pixels having luminance values larger than the threshold value luminance value are displayed in white.

In step 304, the joint is removed from the image of the part using the opening to remove the minor blob area. An image of the result of performing the opening process in FIG. 4 (c) is shown, and the centered adjustment mark (indicated by reference numeral 33 shown in FIG. 4 (b) Lt; / RTI >

In step 305, "1" is assigned to the binarized video, and the blob areas connected thereto are numbered in order. The large blobs at the edge of the image in step 306 (the part marked with "1" and "2" in FIG. 4 (d)) are virtually unnecessary blob images and are therefore removed. FIG. 4 (e) shows an image with the unnecessary area removed.

In step 307, a predetermined arithmetic operation is performed with the spatial filter table, the Sovell operator, and the like and each pixel of the image to detect an outline (see FIG. 4 (f)). In step 308, a moment is calculated Handan. In step 309, two blob images resembling each other are detected regardless of angles by detecting a blob image having the same moment value calculated in the above step.

In step 310, the center of gravity of each blob image ("③", "④" in FIG. 4 (e)) is calculated and the center of gravity C calculated in step 311 is used to calculate the center point As a reference for detecting the adjustment flaws of the adjustment component.

As described above, according to the present invention, the adjustment time can be shortened by accurately recognizing the angle with the position of the adjustment component, and the adjustment of the idle degree prevents the adjustment component from being damaged by the adjustment bit.

Claims (2)

An adjustment device for adjusting an adjustment component mounted on a printed circuit board, comprising: a camera section for picking up an image of an adjustment target component and acquiring and outputting a video signal; an image processing section for processing and outputting a video signal input from the camera section; A binary image processing unit that receives the image signal output from the image processing unit, performs a binary image process, removes a connected portion from an image of the component through an opening process, removes a blob image close to an edge, The outline of the blob area is detected, and the moment is calculated at each pixel of the detected outline. Two blob images having similar moment values are found, and the center of gravity of each blob image is calculated. Using the calculated center of gravity, And outputs a control signal for adjusting the component to be adjusted, An XY table that receives a control signal from the XY table and moves in the X and Y axis directions, and an adjustment unit that is mounted on the XY table and receives a control signal from the control unit to insert and adjust an adjustment bit in the adjustment piece of the adjustment target part And an adjustment device for adjusting the adjustment component. A method for detecting an adjustment flaw in an adjustment component, the method comprising: acquiring an image of a component to be adjusted from a camera and performing median filtering to remove noise by smoothing the image; performing a binarized image process; A step of removing a connected portion from the image of the blob region and labeling the blob region in order; executing a labeling process of numbering the blob region in order, detecting an outline of the blob region, calculating a moment at each pixel of the detected outline, Calculating two blob images resembling each other, calculating the center of gravity of each blob image, calculating a center point (C) using the calculated center of gravity, and setting the center point as the center of the adjusted blob A method of detecting an adjustment flaw in a component. ※ Note: It is disclosed by the contents of the first application.