JP3170988B2 - Slot position recognition method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a position recognizing method for a visual angle recognizing apparatus for a screw tightening robot, and more particularly, to a position of an upper hole which is a long hole and a lower hole which is a screw hole from an image taken by an image pickup means. And a method for recognizing a slot position for determining whether or not a screw can be tightened.

[0002]

2. Description of the Related Art In recent years, a position recognition method for detecting the position of an object from an image picked up by an image pickup device has been widely used for a visual angle recognition device of a robot in various production facilities.

A conventional example in which the position of a circular object is recognized from among the position recognition methods used above will be described.

FIG. 6 shows the principle of a conventional position recognition method, and shows a basic configuration for detecting the center position of a circular object. In FIG. 6, reference numeral 20 denotes an image pickup means for picking up an image of an object to output a video signal, reference numeral 21 denotes a gray-scale image storage means for storing the video signal in image data, and reference numeral 22 scans the gray-scale image data by a contour scanning window described later. The scanning means 23 is a position calculating means for calculating the position of the object using the scanning data.

FIG. 7 shows a contour scanning window A. In FIG. 7, reference numeral 24 denotes a scanning contour equal to the contour of the circular object, 25 denotes a density correlation line orthogonal to the scanning contour and measuring the image density inside and outside the scanning contour, and contour scanning window A denotes this density correlation line. Is provided in a plurality of windows.

The position calculating means 23 calculates the absolute value of the difference between the measured density total value at a measurement point inside the contour of the object and the measured density sum value at a measurement point outside the contour of the object on the density correlation line 25. It is.

[0007] The image data of the circular object imaged by the imaging means 20 and stored by the grayscale image storage means 21 is scanned by the contour scanning window 22 and is calculated by the position calculation means 23 for each density correlation line. The position of the contour scanning window in which the number of density correlation lines larger than the predetermined threshold is maximized can be detected as the position of the circular object.

[0008]

However, in the above-mentioned conventional method, when the hole of the upper plate of the screw hole of the object is a long hole, the position of the upper hole cannot be detected. The screw was tightened upon detection, but when the upper and lower holes overlapped, screw tightening could be poor or the object could be damaged.

The present invention has been made to solve the above-mentioned conventional problems, and even if the upper plate of an object is a long hole, it is possible to accurately detect its position regardless of the inclination of the long hole, An object of the present invention is to provide a long hole position recognition method for judging an overlap between upper and lower holes and preventing a screw tightening failure beforehand.

[0010]

According to the present invention, there is provided a method for recognizing the position of a long hole, the method comprising the steps of: The image is scanned by a contour scanning window provided with a plurality of density correlation lines orthogonal to the circular contour including the semicircle, and the measured density total value at the inner measurement point and the outer measurement from the circular contour are measured by the circular contour. The position of the contour scanning window in which the number of density correlation lines in which the absolute value of the difference from the measured density total value of the point is greater than a predetermined threshold is maximized is detected as the position of one semicircular portion of the object, and the position is detected. And detecting the position of the contour scanning window in which the number of density correlation lines larger than a predetermined threshold value within the range designated by the maximum is maximized as the position of the other semicircular portion.

In order to solve the above-mentioned problems, the method of recognizing the position of a long hole according to the present invention detects the positions of both ends of the long hole and determines the position of the upper hole from the two positions and the detected position of the prepared hole. It is characterized by judging the presence or absence of overlap with a pilot hole.

[0012]

According to the method for recognizing a long hole according to the present invention, even if the object is a long hole, its position can be obtained irrespective of the inclination of the long hole. Can be detected, and the overlap with the upper hole, which is a long hole, can be determined, and defective screw tightening can be prevented.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A long hole position recognition method according to one embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing a main part of a position recognition apparatus using the long hole position recognition method of the present invention.

In FIG. 1, an image pickup means 1 picks up an image of an object and outputs a video signal, a gray-scale image storage means 2 stores the video signal as gray-scale image data, and a scanning means 3 uses the contour scanning window to generate the video signal. Scan the grayscale image data,
The position calculating means 4 calculates the position of the object using the scanning data obtained by scanning as described above. Further, the shift amount judging means 5 judges whether the positions of the upper hole and the lower hole obtained by the image pickup means 1, the grayscale image storage means 2, the scanning means 3, and the position calculation means 4 overlap.

FIG. 2 shows grayscale image data, in which 6 is a long hole of the upper hole object, and 7 is a lower hole of the lower hole object which normally performs screw tightening.

FIG. 3 shows a contour scanning window A used in this embodiment, and reference numeral 8 denotes a scanning contour having a diameter equal to the diameter of the semicircle at both ends of the image of the long hole 6 of the upper hole object. . 9a to 9h are orthogonal to the scanning contour 8 and
Are a plurality of density correlation lines for measuring the image densities inside and outside the area.

FIG. 4 shows the details of the density correlation line 9c among the density correlation lines 9a to 9h of the contour scanning window A used in this embodiment, and 10c denotes the scanning contour position of the density correlation line 9c, and 11c and 11c. Reference numerals 12c and 13c denote a predetermined number of inner image density measurement points constituting the inner density correlation line B;
Reference numerals c, 15c, and 16c denote a predetermined number of outer image density measurement points constituting the outer density correlation line C.

Next, the position recognition operation in this embodiment will be described with reference to FIGS. First, an object including a long hole is imaged by the imaging unit shown in FIG. 1 to obtain a grayscale image, and stored in the grayscale image storage unit 2.

In this gray scale image, there are a long hole 6 of the upper hole object and a prepared hole 7 of the lower hole object as shown in FIG.

In this case, the operation of this embodiment for detecting the position of the long hole will be described below. First, the scanning means 3 shown in FIG. 1 starts scanning rightward from the upper left of the grayscale image stored by the grayscale image storage means 2 using the contour scanning window A shown in FIG. Is calculated by the position calculation means.

For each density correlation line, the measured density total value 11c + 12c + 13c of the inner density correlation line B composed of a predetermined number of inner image density measurement points and the measured density of the outer density correlation line C composed of a predetermined number of outer image density measurement points. The absolute value of the difference from the total value 14c + 15c + 16c is calculated, and the position of the contour scanning window where the absolute value is larger than the predetermined density threshold value and the number of density correlation lines is maximized is determined by one semicircle at both ends of the elongated hole. Is detected as the position.

The position of the contour scanning window at which the number of density correlation lines larger than a predetermined density threshold is maximum within a specified range from the detected position of one semicircle is defined as the position of the other semicircle. To detect.

As described above, the positions of both ends of the elongated hole can be recognized. Next, in FIG. 5, reference numeral 17a denotes a semicircular position at one end of both ends of the detected elongated hole, and 17b denotes a semicircular position at the other end of the elongated hole. Reference numeral 18 denotes a prepared hole 7 for a prepared hole object.
Is the center position, and 19 indicates a straight line connecting 17a and 17b. D is a region of a semicircular portion of the long hole, and E is a region of the long hole 6 of the upper hole object other than the region D.

The elongated hole 6 of the upper hole object and the prepared hole 7 of the prepared hole object
When the prepared hole 7 of the prepared hole object exists in the area D, the distance between the straight line 19 and the prepared hole center position 18 is calculated. When the prepared hole 7 of the prepared hole object exists in the region E, the position 17a of the prepared hole included in the existing portion
Alternatively, the distance between 17b and the center position 18 of the pilot hole is calculated.

As described above, it is possible to determine the presence or absence of an overlap based on the amount of displacement between the long hole object and the low hole object.

[0027]

As described above, according to the present invention, even if the object to be screwed is a long hole, the position of the long hole can be detected from the boundary pattern information of a part of the long hole. Further, it is possible to calculate the amount of deviation of the upper and lower holes from the position of the upper hole and the position of the prepared hole, determine the overlap, and prevent screw tightening failure.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of an embodiment of a position recognition device using a long hole position recognition method of the present invention.

FIG. 2 is a diagram showing an example of a grayscale image used in the long hole position recognition method of the present invention.

FIG. 3 is a diagram showing a configuration of a contour scanning window used in the position recognition device shown in FIG. 1;

FIG. 4 is a partial detailed view of the contour scanning window shown in FIG. 3;

FIG. 5 is a detailed explanatory diagram of the grayscale image shown in FIG. 2;

FIG. 6 is a block diagram showing a configuration of a position recognition device using a conventional position recognition method.

FIG. 7 is a diagram showing a configuration of a contour scanning window used in the position recognition device shown in FIG. 6;

[Explanation of symbols]

A outline scanning window B inner density correlation line C outer density correlation line D Both ends of a long hole semicircle E Portion excluding the region of the long hole 1 Imaging means 2 Grayscale image storage means 3 Scanning means by contour scanning window 4 Position Calculation means 5 Deviation amount determination means 9a, 9b, 9c, 9d, 9e, 9f, 9g, 9h Density correlation line 10c Scanning contour position 11c, 12c, 13c Inner image density measurement point 14c, 15c, 16c Outer image density measurement point 17a , 17b The center position of the semicircle forming both ends of the long hole

────────────────────────────────────────────────── ─── Continued on the front page (72) Inventor Shinya Nakao 1006 Kazuma Kadoma, Kadoma, Osaka Prefecture Matsushita Electric Industrial Co., Ltd. (56) References JP-A-63-124181 (JP, A) JP-A-63- 234103 (JP, A) JP-A 1-244077 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) G01B 11/00-11/30 102 G06T 1/00 G06T 7/00 G06T 7/60

Claims (1)

(57) [Claims] 1. An object in which both ends of a long hole are semicircular is imaged by an imaging means to obtain a grayscale image, and a scanning contour equal to a circular outline including a semicircular portion on the grayscale image and this scanning are obtained. The grayscale image is scanned by a contour scanning window provided with a plurality of density correlation lines for measuring image densities inside and outside of the scanning contour, which are orthogonal to the contour, and each of the density correlation lines includes a predetermined number of inner image density measuring points. Calculate the absolute value of the difference between the measured density total value of the density correlation line and the measured density total value of the outer density correlation line composed of a predetermined number of outer image density measurement points, and calculate the absolute value of the absolute value greater than a predetermined density threshold value. The position of the contour scan window at which the number of correlation lines is maximized is detected as the position of one semicircular portion of the object, and the number of density correlation lines larger than a predetermined threshold within a specified range from that position is determined. The position of the maximum contour scanning window is detected as the position of the other semicircle, and
A long hole position recognizing method characterized by judging an overlap between a prepared hole and an elongated hole which is an upper hole from two positions and a detected position of the prepared hole.