KR20210085717A - Position calibrating method for semiconductor factory - Google Patents

Abstract

In a position calibrating method using a position calibrating device for a semiconductor factory, first, a picture of a subject arranged on a line of the semiconductor factory is captured using a camera, and a two-dimensional image signal in a still state is generated. Then, a two-dimensional black-and-white image signal is generated from the two-dimensional image signal using a black-and-white imaging unit. Thereafter, a two-dimensional binary data form image signal is generated from the two-dimensional black-and-white image signal by using a binarization unit. Next, quadrangular-shaped images are detected from the two-dimensional binary data-shaped image signal by using a quadrangle detection unit. Then, a quadrangle that satisfies a preset condition among the detected quadrangle is recognized as a quadrangle image of an object to be processed and extracted by using a quadrangle extraction unit. Thereafter, calibrating data is output by calculating the position and inclination of the extracted quadrangular-shaped image using the calibrating unit.

Description

POSITION CALIBRATING METHOD FOR SEMICONDUCTOR FACTORY

The present invention is to provide a position correction method for a semiconductor factory. More particularly, it relates to a method of providing corrected position information to an automated robot by analyzing an image of a subject used in a semiconductor factory.

With industrialization, the application of automated processes is increasing. Works previously performed by humans are now being performed by industrial robots. This trend is widely used in factories that require precise and fine machining.

In particular, in the case of semiconductor factories, the range in which industrial robots are used is wide because fine processing that is invisible to the naked eye is required.

In the use of industrial robots, it is important to accurately position the object to be processed in the processing part of the robot.

However, the position or angle of the object to be processed is likely to be shifted due to external impact or vibration generated by itself during the process. If the position or angle of the object to be processed is misaligned, it may cause processing defects and cause disruptions in product production.

In the case of Republic of Korea Patent No. 10-1850835, "Method for estimating indoor position of a mobile robot using a ray tracing method," after acquiring image data using a camera, an outline is extracted, and the pexel distance data and actual distance information are A technique for estimating the position of a camera by sequential calculation has been developed. However, Korean Patent Registration No. 10-1850835 has a problem in that it becomes difficult to control the robot when the type or shape of the object to be processed is changed because it simply estimates the distance of the camera.

In the case of Republic of Korea Patent No. 10-1471103, "vision-based gas insulated switchgear grounding switch/disconnector driver position recognition method", a technology for generating position information by correcting the distortion of the captured image and extracting the position of the feature point was developed has been However, even in the case of Korean Patent Registration No. 10-1471103, there is a problem in that the control of the robot becomes difficult when the type or shape of the object to be processed is changed.

Republic of Korea Patent No. 10-1850835, "Method for estimating indoor position of mobile robot using ray tracing method" Republic of Korea Patent No. 10-1471103, "vision-based gas insulated switchgear grounding switch/disconnector driver position recognition method"

The present invention has been devised to solve the above problems, and an object of the present invention is to provide a method of providing corrected position information to an automated robot by analyzing an image of a subject used in a semiconductor factory.

The present invention is a means for solving the above problems, in a position correction method using a position correction device for a semiconductor factory, wherein the position correction device for a semiconductor factory includes a camera disposed on a line of the semiconductor factory, and the camera A capture unit connected to the capture unit, a black and white unit connected to the capture unit, a binarization unit connected to the black and white unit, a rectangle detection unit connected to the binarization unit, a rectangle extraction unit connected to the rectangle detection unit, and connected to the rectangle extraction unit and an information processing member having a correction unit to be In the position correction method using the position correction device for a semiconductor factory, a picture of a subject arranged on a line of the semiconductor factory is first captured using the camera to generate a two-dimensional image signal in a still state. Next, a two-dimensional black-and-white image signal is generated from the two-dimensional image signal using the black-and-white imager. Thereafter, an image signal in a two-dimensional binary data format is generated from the two-dimensional black-and-white image signal by using the binarization unit. Subsequently, the square-shaped images are detected from the two-dimensional binary data-type image signal by using the square detection unit. Then, by using the quadrangle extraction unit, a quadrangle meeting a preset condition among the detected quadrangle is recognized and extracted as a quadrangle image of an object to be processed. Thereafter, the position and inclination of the extracted rectangular image are calculated using the correction unit and output as correction data.

In an embodiment, the generating of a two-dimensional image signal in a still state by capturing a photo of the subject includes: generating image data of the subject using the camera; and using the capture unit to convert the image data into a two-dimensional image signal of the still life.

In an embodiment, the detecting of the rectangular images from the two-dimensional binary data format image signal includes selecting regions where black (0) and white (1) intersect among the two-dimensional binary data format image signals. detecting the connected line; and extracting an image in which the four detected lines are connected and detecting the image as a rectangle.

In one embodiment, the step of recognizing and extracting the rectangular image of the object to be processed includes analyzing the two-dimensional coordinates of the four vertices of the rectangular image to determine the center of the four vertices as the object to be processed. estimating the location of the; and averaging the inclinations of adjacent vertices among the four vertices to calculate the inclination of the object to be processed.

As described above, according to the present invention, the position and inclination of the object to be processed can be measured by extracting the object to be processed into a rectangular shape that meets the preset conditions, rather than simply estimating the distance of the camera. have.

In addition, by first changing the two-dimensional color image signal into a two-dimensional black-and-white image signal with a small data capacity and secondarily changing it to a two-dimensional binary data form image signal, the information processing time is reduced and the square Errors in black and white images are reduced compared to color images in extraction, positioning, and gradient calculation. Therefore, it can be easily applied even in a process requiring rapid operation.

Therefore, even if the type or shape of the object to be processed is changed in the semiconductor factory, precise processing is possible by accurately measuring the position and inclination of the object, and defects are reduced.

1 is a block diagram illustrating a position correction device for a semiconductor factory according to an embodiment of the present invention.
FIG. 2 is a flowchart illustrating a position correction method using the position correction device for a semiconductor factory shown in FIG. 1 .
3 to 8 are images illustrating a position correction method for a semiconductor factory shown in FIG. 2 .

Before describing various embodiments of the present invention in detail, it is to be understood that the application is not limited to the details of the construction and arrangement of components described in the following detailed description or shown in the drawings. The invention is capable of being embodied and practiced in other embodiments and of being carried out in various ways. Also, device or element orientation (eg "front", "back", "up", "down", "top", "bottom") The expressions and predicates used herein with respect to terms such as ", "left", "right", "lateral", etc. are used merely to simplify the description of the invention, and the associated apparatus Or it will be appreciated that it does not simply indicate or imply that an element must have a particular orientation.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. Prior to this, the terms or words used in the present specification and claims should not be construed as being limited to their ordinary or dictionary meanings, and the inventor should properly understand the concept of the term in order to best describe his invention. Based on the principle that it can be defined, it should be interpreted as meaning and concept consistent with the technical idea of the present invention.

Therefore, the configuration shown in the embodiments and drawings described in the present specification is only the most preferred embodiment of the present invention and does not represent all of the technical spirit of the present invention, so at the time of the present application, various It should be understood that there may be equivalents and variations.

1 is a block diagram illustrating a position correction device for a semiconductor factory according to an embodiment of the present invention.

Referring to FIG. 1 , a position correction device for a semiconductor factory includes a camera 70 , a capture unit 80 , an information processing member 100 , and an output member 200 .

The camera 70 is disposed on a line of a semiconductor factory and mounted on an automated facility (not shown) to generate image data of the subject 50 .

The capture unit 80 is electrically connected to the camera 70 , and generates a two-dimensional image signal in a still state receiving image data from the camera 70 . In this embodiment, the capture unit 80 receives a still image, not a moving image, from the camera 70 . In another embodiment, the capture unit 80 may receive a moving picture from the camera 70, and extract a still image in a preset time unit from among the authorized moving images to generate a two-dimensional image signal in a time unit.

The information processing member 100 is connected to the capture unit 80 and corrects the position and inclination of the object to be processed from the two-dimensional image signal applied from the capture unit 80 .

In this embodiment, the information processing member 100 includes a black and white unit 110 , a binarization unit 120 , a rectangle detection unit 140 , and a correction unit 150 .

The black-and-white imager 110 generates a two-dimensional black-and-white image signal from the two-dimensional color image signal received from the capture unit 80 .

The binarization unit 120 generates an image signal in the form of two-dimensional binary data from the image signal of the two-dimensional black-and-white form applied from the black-and-white unit 110 .

After the black-and-white conversion unit 110 primarily converts the two-dimensional color image signal into a two-dimensional black-and-white image signal, the binarization unit 120 converts the two-dimensional black-and-white image signal into a two-dimensional binary data form image. The reason for the secondary change to the signal is that the data capacity is reduced, the subsequent information processing time is reduced, and the error of the black-and-white image is reduced compared to the color image in rectangle extraction, positioning, and slope calculation. Therefore, it can be easily applied even in a process requiring rapid operation.

The rectangle detector 130 detects images in the form of a rectangle on the image signal in the form of two-dimensional binary data received from the binarizer 120 . In the present embodiment, a line connecting regions where black (0) and white (1) intersect is detected among image signals in the form of two-dimensional binary data, and an image in which four of these lines are connected is detected as a rectangle.

The quadrangle extraction unit 140 compares the rectangle-shaped images received from the rectangle detection unit 130 with a preset condition, and extracts a rectangle-shaped image that meets the preset condition. Specifically, conditions such as area, aspect ratio, and the like of the object to be processed are set in advance and input to the rectangle extraction unit 140 , and the rectangle extraction unit 140 receives the rectangular shape received from the rectangle detection unit 130 . Among the images, a rectangular image most similar to a preset condition is extracted as an image of an object to be processed. For example, the rectangle extraction unit 140 may extract a rectangular image in the form of two-dimensional position coordinates in which the upper left corner is (0, 0) with respect to four vertices.

The correction unit 150 analyzes a rectangular image that meets a preset condition applied from the rectangle extraction unit 140, and calculates the exact position and inclination of the object to be processed.

The output member 200 uses the correct position and inclination of the object to be processed by the correction unit 150 to determine the position and inclination of the object to be processed by an automated processing machine (not shown) of the semiconductor factory. Output the correction data.

Hereinafter, a position correction method using a position correction device for a semiconductor factory according to an embodiment of the present invention will be described.

FIG. 2 is a flowchart illustrating a position correction method using the position correction device for a semiconductor factory shown in FIG. 1 , and FIGS. 3 to 8 are images illustrating the position correction method for a semiconductor factory shown in FIG. 2 .

FIG. 3 is an image illustrating a step of capturing a picture in the method for correcting a position for a semiconductor factory shown in FIG. 2 .

1 to 4 , a picture of a subject 50 is first captured to generate a two-dimensional image signal in a still state regarding the subject 50 (step S110).

Specifically, image data of the subject 50 transported by a conveyor (not shown) or the like is generated using the camera 70 disposed on the line of the semiconductor factory. Next, by using the capture unit 80 connected to the camera 70, the image data is changed to a two-dimensional image signal in a still state.

FIG. 4 is an image illustrating a black-and-white step in the position correction method for a semiconductor factory shown in FIG. 2 .

1, 2, and 4, a two-dimensional black-and-white image signal is generated from a two-dimensional image signal in a still state (step S120).

In this embodiment, a two-dimensional color image signal is converted into a two-dimensional black-and-white image signal by using the black and white unit 110 connected to the capture unit 80 .

FIG. 5 is an image illustrating a black-and-white image in the position correction method for a semiconductor factory shown in FIG. 2 .

Referring to FIGS. 1, 2 and 5 , a two-dimensional black-and-white image signal is then binarized to generate an image signal in a two-dimensional binary data form (step S130).

In the present embodiment, a two-dimensional black-and-white image signal is binarized into an image signal in a two-dimensional binary data form by using the binarization unit 120 connected to the black and white unit 110 .

FIG. 6 is an image illustrating a binarization step in the position correction method for a semiconductor factory shown in FIG. 2 .

Referring to FIGS. 1, 2, and 6, images in a square shape are continuously detected from an image signal in the form of two-dimensional binary data (step S140).

In the present embodiment, the rectangle-shaped images are detected from the two-dimensional binary data type image signal by using the rectangle detection unit 130 connected to the binarization unit 120 . In this embodiment, the line connecting the regions where black (0) and white (1) intersect is first detected among the image signals in the form of two-dimensional binary data, and then an image in which four detected lines are connected is extracted to detect as a rectangle.

FIG. 7 is an image illustrating a step of detecting a quadrangle in the position correction method for a semiconductor factory shown in FIG. 2 .

Referring to FIGS. 1, 2, and 7 , a quadrangle meeting a preset condition among the detected quadrilaterals is recognized as an object to be processed, and a quadrangle image is extracted (step S150).

In this embodiment, using the rectangle extraction unit 140 connected to the rectangle detection unit 130, the received rectangular images and a preset condition are compared. Specifically, conditions such as area, aspect ratio, and the like of the object to be processed are set in advance and input to the rectangle extraction unit 140 , and the rectangle extraction unit 140 receives the rectangular shape received from the rectangle detection unit 130 . Among the images, a rectangular image most similar to a preset condition is extracted as a rectangular image of an object to be processed. For example, when the contour of the object to be processed is not an exact rectangle, it can be extracted as a rectangle image by approximating it to be as close to a rectangle as possible.

For example, the rectangle extraction unit 140 extracts a rectangular image in the form of two-dimensional position coordinates with the upper left corner as (0,0) and the lower right corner as (640, 640) with respect to four vertices. can do.

FIG. 8 is an image illustrating the steps of checking the position of the rectangle extracted in the position correction method for a semiconductor factory shown in FIG. 2 and calculating the slope.

Thereafter, the position of the extracted rectangle is checked and the slope is calculated (step S160).

In this embodiment, by using the correction unit 150 connected to the rectangle extraction unit 140 to analyze a rectangular image corresponding to the object to be processed, the exact position of the object to be processed and the inclination Calculate.

For example, by analyzing the two-dimensional coordinates of the four vertices of a rectangular image, the center of the four vertices is calculated as the position of the object to be processed, and the inclinations of the adjacent vertices are averaged to become the processing target. Calculate the inclination of the object.

Finally, by using the output member 200 connected to the correction unit 150, the correct position and inclination of the object to be processed are output as correction data of an automated processing machine (not shown) of a semiconductor factory.

According to the embodiment of the present invention as described above, rather than simply estimating the distance of the camera, the object to be processed is extracted into a rectangular shape that meets a preset condition, and the position and inclination of the object to be processed are measured. can

In addition, by first changing the two-dimensional color image signal into a two-dimensional black-and-white image signal with a small data capacity and secondarily changing it to a two-dimensional binary data form image signal, the information processing time is reduced and the square Errors in black and white images are reduced compared to color images in extraction, positioning, and gradient calculation. Therefore, it can be easily applied even in a process requiring rapid operation.

Therefore, even if the type or shape of the object to be processed is changed in the semiconductor factory, precise processing is possible by accurately measuring the position and inclination of the object, and defects are reduced.

The position correction method for a semiconductor factory of the present invention has industrial applicability that can be used in a semiconductor factory, a precision processing factory, a fine processing factory, and the like.

As described above, although the present invention has been described with reference to limited embodiments and drawings, the present invention is not limited thereto, and the technical spirit of the present invention and the following by those of ordinary skill in the art to which the present invention pertains. Of course, various modifications and changes are possible within the scope of equivalents of the claims to be described.

50: subject 70: camera
80: capture unit 100: information processing member
110: black and white image 120: binary image
130: rectangle detection unit 140: rectangle extraction unit
150: correction unit 200: output member

Claims (4)

A camera disposed on a line of a semiconductor factory, a capture unit connected to the camera, a black and white unit connected to the capture unit, a binarization unit connected to the black and white unit, a rectangle detection unit connected to the binarization unit, the rectangle detection unit In the position correction method using a position correction device for a semiconductor factory comprising an information processing member having a rectangle extraction unit connected to, and a correction unit connected to the rectangle extraction unit,
generating a two-dimensional image signal in a still state by using the camera to capture a photograph of a subject disposed on a line of the semiconductor factory;
generating a two-dimensional black-and-white image signal from the two-dimensional image signal using the black and white unit;
generating an image signal in a two-dimensional binary data format from the two-dimensional black-and-white image signal using the binarization unit;
detecting square-shaped images from the two-dimensional binary data-type image signal using the square detector;
Recognizing and extracting a rectangle that meets a preset condition among the detected rectangles as a rectangle image of an object to be processed using the rectangle extraction unit; and
and checking the position of the extracted rectangular image by using the correction unit, calculating a slope, and outputting the extracted rectangular image as correction data. According to claim 1, wherein the step of generating a two-dimensional image signal in a still state by capturing a picture of the subject,
generating image data of the subject using the camera; and
and converting the image data into a two-dimensional image signal of the still state by using the capture unit. According to claim 1, wherein the step of detecting the rectangular images from the image signal in the form of two-dimensional binary data,
detecting a line connecting regions where black (0) and white (1) intersect from among the two-dimensional binary data type image signals; and
and extracting an image in which the four detected lines are connected and detecting the image as a rectangle. The method of claim 1, wherein the step of recognizing and extracting a rectangular image of the object to be processed comprises:
estimating the center of the four vertices as the position of the object to be processed by analyzing the two-dimensional coordinates of the four vertices of the rectangular image; and
and calculating the inclination of the object to be processed by averaging the inclinations of adjacent vertices among the four vertices.

Images