JP3087788B2 - Component position detection method and device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a robot for handling an object, for example, which detects the position and angle of the object,
The present invention relates to a method and an apparatus for detecting a position of a component capable of obtaining such information as control information.

[0002]

2. Description of the Related Art When controlling an automatic assembling apparatus using a robot or the like, it is necessary to detect the position and angle of an object on which the robot is to act. As a method for detecting the position and the angle of such an object, a method is known in which the coordinates of the center of gravity obtained from the shape of the target object are set as the positions, and the direction of the principal axis of inertia passing through the center of gravity is set as the angle. However, in the case where the objects are overlapped or partially missing when viewed from the robot side, accurate position detection of the objects cannot be performed by this method.

[0003] Therefore, features of the shape of an object are extracted from the difference between an original image captured by a device such as a robot and an image subjected to dilation / reduction processing, and the position of the object is determined from the positional relationship between the centers of gravity. And a method of detecting the direction are being considered. According to this method, when recognizing the shape of the object,
Even if a part of the object is missing, if a plurality of features of the shape can be obtained, the position and direction of the object can be detected.

[0004]

When the position and angle of an object are detected by the above-described method, it is better to extract as many features of the shape of the object as possible. In the above-described method using the difference of the above, only one of the concave portion and the convex portion of the shape of the target object is obtained in the collective processing, while the feature such as the hole portion of the target object needs to be obtained separately by a labeling process or the like. there were.

SUMMARY OF THE INVENTION The present invention has been made in view of the above-described conventional example, and detects a feature amount of a shape of an object at a time to detect a position and an angle of the object. It is an object to provide a method and an apparatus.

[0006]

In order to achieve the above object, a position detecting device according to the present invention has the following arrangement. That is, a component position detecting device that photoelectrically reads the shape of a component and detects the position of the component, the edge extracting unit extracting an edge portion from image information of an object, and the edge extracted by the edge extracting unit. After expanding the part , the original
And extracted by the edge extracting means.
Detection means for detecting irregularities moiety included in the edge portion by comparing said edge portion that is, an extraction means for extracting an edge of said detected irregularity portion by the detecting means, is extracted by the extraction means and inversion processing means for obtaining a closed image area indicating inversion of the uneven portion of the edge was a feature quantity extracting means for calculating a feature value representing the size and positional relationship of the closed image region obtained by said inverting means , comparison for comparing a memory means for storing feature amount when placing the said object in a predetermined position, the feature amount stored in the storage means, and a feature amount calculated by the feature calculating unit Means, and a position / angle calculating means for detecting a position and / or an orientation of the object according to a comparison result by the comparing means.

In order to achieve the above object, a position detecting method according to the present invention includes the following steps. That is, a component position detecting method for photoelectrically reading the shape of a component and detecting its position, comprising the steps of photoelectrically reading an object and extracting an edge portion from image information thereof; after the partial expanded treated, and reduced to the original size, the
Compare with the edge part extracted in the extraction step.
Step and a step of extracting an edge of said detected irregularity portion thereof, wherein the uneven portion by inverting the extracted edge detecting irregularities moiety included in the edge portion by the
The step of obtaining a closed image area indicating the minutes and thus the inverted
Calculating a feature value representing the size and positional relationship of the closed image area stores the characteristic amount when the object is placed at a predetermined position, a feature amount that is the stored, is calculated And a step of detecting the position and / or orientation of the object according to the result of the comparison.

[0008]

In the above arrangement, the edge portion is extracted from the image information of the object, the extracted edge portion is expanded, then reduced to the original size, and the edge portion extracted by the edge extracting means is extracted. A closed image area showing the uneven portion by detecting the uneven portion included in the edge portion, extracting the edge of the detected uneven portion, and inverting the edge of the extracted uneven portion. Obtain, calculate the feature amount representing the size and positional relationship of the obtained closed image area, compare the feature amount when the object is placed at a predetermined position, and the calculated feature amount, An operation is performed to detect the position and / or orientation of the object according to the comparison result.

[0009]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described below in detail with reference to the accompanying drawings.

FIG. 1 is an external perspective view of a component conveying device provided with the position / angle detecting device of the present embodiment.

In this position / angle detecting device, a target component is first put into component separating devices 100 and 101.
These component separation devices place the components that have been input separately on a table in the component separation device and raise the table. This table is provided with several concave portions. By raising and tilting the table, components other than the components fitted in the concave portions are dropped out of the table. By such an operation, the component separating device 10
0, 101 can place the part in a fixed area on the table. The component separation work by the component separation devices 100 and 101 is performed alternately.

The components separated by the component separating devices 100 and 101 in this way are supplied to the cameras 102 and 10 respectively.
3, and the image signal is sent to the position / angle detection device 104. The position / angle detection device 104 detects the position and orientation of the component on the table based on the image signal by the processing shown in the flowchart of FIG. A robot controller 105 operates the robot 110 in accordance with the information indicating the position and orientation from the position / angle detector 104, and stores the components on the table in the magazine 106 in a predetermined direction.

FIG. 2 shows the position / angle detecting device 104 shown in FIG.
FIG. 2 is a block diagram showing a schematic configuration of the embodiment.

The position / angle detecting device 104 is a cable 1
07 and 108 are connected to two cameras 102 and 103, respectively, for taking in image signals representing the shape of the component. These cameras 102, 1
The analog image signal output from the CCD 03 is A / D
It is input to the conversion unit 210. In this embodiment, only one A / D converter 210 is provided.
The image signals from the two cameras 102 and 103 are switched as necessary to be converted into digital signals, and are converted into digital signals.
Output to

The position / angle detection device 104 has a communication unit 217 so that data communication can be performed with the robot control device 105. It is configured so that communication can be performed. Next, the configuration of the position / angle detection device 104 will be described in detail.

A microcomputer 200 controls the operation of the entire position / angle detecting device 104 according to a control program stored in a control program unit 201, and is connected to all other components via a bus 218. Have been. In the present embodiment, the cameras 102 and 10
The A / D converter 210 divides the image information fetched from 3 into 8-bit (“0” to “255”) 256-level brightness, and fetches a part of the image data into the frame memory 211. . The image information thus taken into the frame memory 211 is converted by the binarizing unit 202 into two levels of “1” and “0” at a required brightness level. The binarized image information is stored in the frame memory 211 again.

Reference numeral 203 denotes an edge extracting unit, which is a binarizing unit 2
Based on the image information binarized by 02 and stored in the frame memory 211, it has a function of extracting an outline of the image. Reference numeral 204 denotes an expansion processing unit which performs expansion (enlargement) processing of image information stored in the frame memory 211. Reference numeral 205 denotes a reduction processing unit, which is a frame memory 2
A reduction process is performed on the image information stored in 11. Reference numeral 206 denotes an inversion processing unit, which is a frame memory 211.
Has the function of inverting the “1” and “0” data of the image information stored in the “0” and “1” data, respectively. 207
Is a labeling unit, and 2 is added to data consisting of “0” and “1”.
The image information that has been converted into a value and stored in the frame memory 211 is divided for each closed area of “1” or “0”,
It has a function to label each area. 208
Denotes a CRT for displaying an image signal read by the cameras 102 and 103 or an image subjected to various conversions;
Reference numeral 09 denotes a keyboard for inputting various instructions and the like.

Reference numeral 211 denotes a frame memory.
02, 103 and a memory for temporarily storing image information taken in through the A / D converter 210 and image information processed in the position / angle detection device 104.
212, a coordinate conversion unit which is a CCD of the cameras 102 and 103;
The conversion coefficient of the relationship between the coordinate position of the robot and the coordinate position of the robot is stored in advance, and when the position data sent from the robot control device 105 is used for processing in the position / angle detection device 104, the position The data is transferred to the camera 102,
It is converted to the coordinate system of the CCD 103. When sending the position data calculated by the position and angle detection device 104 to the robot control device 105, the robot has a function of converting the position data into position data of a robot coordinate system. An area setting unit 213 can set an area to be captured by the cameras 102 and 103. Thereby, the cameras 102 and 103
With this CCD, it is possible to capture only image information of a necessary limited area. As a result, processing can be performed in a shorter time than processing image information from the entire area of the CCD.

Reference numeral 214 denotes a feature value calculation unit having a function of calculating a feature value such as the size of a closed region, its center of gravity, and the positional relationship between the image information processed up to the labeling. A feature amount storage unit 216 stores in advance feature amounts for discriminating a part and its position and angle. As a result, the feature amount obtained by the feature amount calculation unit 214 and the feature amount storage unit 216 by the feature amount comparison unit 215.
And determines whether the object is a determined part by comparing with the feature amount stored in the storage unit, and obtains the position and angle of the object.

FIG. 3 is a flowchart showing processing in the object position / angle detecting device 104 according to the present invention. Hereinafter, the processing shown in the flowchart of FIG.
This will be described in detail with reference to FIG.

First, in step S1, an image of a target component is captured by the camera 102 or 103, and image information of the component is fetched and stored in the frame memory 211. Here, an image area to be read by the camera 102 or the camera 103 is determined by the area setting unit 213, and the coordinate position of the image information thus read is converted by the coordinate conversion unit 212 into the coordinate values of the CCDs of the cameras 102 and 103. . Next, the process proceeds to step S2, where the image information fetched in step S1 and stored in the frame memory 211 is
The image data is binarized by the value conversion unit 202 and stored in the frame memory 211. FIG. 4A is a diagram showing an example of the binarized image information. Here, the area of the object 40 having the convex portion 41, the concave portion 42, and the opening 43 is indicated by data “1”, and the outside of the object is indicated by data “0”.

Then, the process proceeds to a step S3, wherein the edge extracting unit 2
In step 03, edges of image information binarized and stored in the frame memory 211 are extracted. The result of the edge extraction is shown in FIG. 4B, and this edge indicates the edge portion of the area where the data of FIG. 4A is "1".

Next, the process proceeds to step S4, where the expansion processing unit 20
In step 4, expansion processing of the edge extracted in step S3 is performed, and the edge shape whose edge portion has been smoothed by this expansion processing is reduced to the original size by the reduction processing unit 205 in step S5. As shown in FIG. 4C, the concave portion 42 is formed by the reduced edge data whose edges are smoothed and the original edge data shown in FIG.
And the convex portion 41 are detected.

Then, the process proceeds to a step S6, wherein the edge extracting unit 2
04, the image thus processed (FIG. 4C)
Is extracted. Next, in step S7, the inversion processing unit 206 performs the inversion processing of the edge portion, thereby dividing each feature (here, the convex portion 41, the concave portion 42, and the opening 43) into a closed region (FIG. 4). (E)).

Next, the process proceeds to step S8, where the labeling unit 2
07, a labeling process for labeling each of these areas is performed. In this labeling process, the frame memory 211 is scanned in the main scanning direction to search for black dots, and the first black dot found is labeled, for example, as “a”. Next, it is checked whether or not a dot next to the black dot is black.
Set the area of. If a pixel previously labeled as "a" is found in this area, the center pixel of that area is also labeled as "a". By repeating such an operation, a continuous area of one block is extracted as the same label. If there are a plurality of continuous areas in this image, these areas are naturally recognized with different labels such as “b” and “c”.

Then, the process proceeds to a step S 9, wherein the feature amount calculation unit 2
In step S13, the sizes of the regions 44 to 46, which are the characteristic portions labeled in step S8, the positions of the respective centers of gravity, and the positional relationship between the regions are calculated as characteristic amounts.
Next, the process proceeds to step S10, where the feature amount stored in advance in the feature amount storage unit 216 as reference data is compared with the feature amount obtained by the feature amount calculation unit 214. As a result,
When the feature amount is different from the feature amount stored in the feature amount storage unit 216, an error display is performed because, for example, there is a possibility that the parts overlap or a part of the parts is missing.

On the other hand, when the values substantially coincide with the characteristics stored in the characteristics storage 216, the flow advances to step S12, and based on the position of the center of gravity of each region, the positional relationship, and the like obtained in step S9. The position and angle (direction) of the target component are determined.

In the present embodiment, the target image is first binarized. However, the multi-valued image may be subjected to edge detection or expansion / reduction processing.

Also, in this embodiment, a component transfer device has been described as an embodiment of the position / angle detection device, but it is needless to say that the present invention can be applied to other component assembly devices.

The present invention may be applied to a system composed of a plurality of devices or an apparatus composed of one device. Needless to say, the present invention can also be applied to a case where the present invention is achieved by supplying a program for implementing the present invention to a system or an apparatus.

As described above, according to the present embodiment, it is possible to extract characteristic amounts of the concave portion, the convex portion, and the opening of the target component.

[0032]

As described above, according to the present invention, there is an effect that the feature amount of the shape of an object can be extracted at a time, and the position and angle of the object can be detected.

[Brief description of the drawings]

FIG. 1 is an external perspective view of a component transfer device according to the present embodiment.

FIG. 2 is a block diagram illustrating a schematic configuration of a position / angle detection device according to the present embodiment.

FIG. 3 is a flowchart illustrating a process for determining a component position / angle in the position / angle detection device of the present embodiment.

FIG. 4 is a diagram illustrating a process of processing image information obtained by reading a component in the position / angle detection device according to the embodiment;

[Explanation of symbols]

 102, 103 Camera 104 Position / angle detection device 105 Robot control device 200 Microcomputer 201 Control program unit 202 Binarization unit 203 Edge extraction unit 204 Expansion processing unit 205 Reduction processing unit 206 Inversion processing unit 207 Labeling unit 211 Frame memory 213 Area setting unit 214 Feature amount calculation unit 215 Feature amount comparison unit 216 Feature amount storage unit

──────────────────────────────────────────────────の Continued on the front page (51) Int.Cl. ⁷ identification code FI H05K 13/08 G06F 15/62 405B (58) Fields investigated (Int.Cl. ⁷ , DB name) B23P 19/00 303 B23P 21 / 00 305 G01B 11/00 G06T 7/00 H05K 13/04 H05K 13/08

Claims (2)

(57) [Claims] 1. A component position detecting device that photoelectrically reads a shape of a component and detects a position of the component, comprising: an edge extracting unit that extracts an edge portion from image information of an object; after the extracted edge portion and dilation processing, and reduced to the original size, the edge extraction means
By comparing with the edge part extracted by
Detection means for detecting irregularities moiety included in the edge portion, an extraction means for extracting an edge of said detected irregularity portion by the detecting means, the concave by inverting the edges extracted by the extraction means
And inversion processing means for obtaining a closed image area showing a protruding portion, the feature value extraction means for calculating a feature value representing the size and positional relationship of the closed image region obtained by inverting means, said object predetermined A storage unit for storing a feature amount when placed at a position; a comparison unit for comparing the feature amount stored in the storage unit with the feature amount calculated by the feature amount calculation unit; And a position / angle calculating means for detecting a position and / or an orientation of the object in accordance with a comparison result. 2. A component position detecting method for photoelectrically reading the shape of a component and detecting its position, comprising the steps of photoelectrically reading an object and extracting an edge portion from image information thereof; After expanding the extracted edge part , the original size
The edge extracted in the extracting step
Closed showing the steps of detecting the uneven portion included in the edge portion by comparing a partial, a step of extracting an edge of said detected irregularity portion thereof, the uneven portion by inverting the extracted edge
And obtaining an image area, thus calculating a feature value representing the size and positional relationship of the inverted said closed image area stores the characteristic amount when the object is placed in a predetermined position, Comparing the stored feature value with the calculated feature value; and, depending on the comparison result, the position and / or position of the object.
Detecting a direction ; and a method for detecting the position of a component.