JPH10118976A - Image capturing position setting method and setting device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To secure recognition precision over a wide range so as to stably recognize workpieces heaped on a palette at random by finding a coordinate value in the height direction of the workpieces when an image based on a three-dimensional visual sensor is recognized and determining whether the workpieces are positioned within a measurable area or not. SOLUTION: In a wrist 2a of a robot 2, a three-dimensional visual sensor 3 provided with a projector projecting slit pattern light and a CCD camera arranged slantingly to the light projecting direction of the projector is installed. An image of the three-dimensional visual sensor 3 is inputted to an image processing device 4. On the basis of the image processing device 4, an image inputting position setting device lowers the three-dimensional visual sensor 3 to the position, in which sufficient recognition precision can be secured, on the basis of the three-dimensional coordinate value recognized by the three- dimensional visual sensor 3 so as to perform measurement for image recognition of workpieces W step loaded on a pallet 1. In this way, recognition precision over a wide range can be secured, and stable recognition can be carried out.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image capturing position setting method and an image capturing position setting method for setting an image capturing position at the time of image recognition when handling articles stacked on a pallet, for example.

[0002]

2. Description of the Related Art When depalletizing articles (work pieces) stacked on a pallet by holding them one by one by a robot hand, the image of the work piece is usually recognized by a visual sensor and processed. Recognition of the handling position of the work is performed.

Conventionally, the following method has been known as a method of capturing an image of a workpiece by the visual sensor. A device that fixes a visual sensor at a predetermined position. For example, as disclosed in Japanese Patent Application Laid-Open No. 7-291450, when the depth of field is deep, a camera device (visual sensor) is disposed above and below, respectively, so that the measurement range is substantially extended. What you did. For example, as disclosed in Japanese Patent Application Laid-Open No. 6-241719, a distance sensor controls the distance between a workpiece and an optical system to be constant.

[0004]

However, when the visual sensor is fixed at a predetermined position as described above, the depth of field is limited and the measurement error due to the angle of view is included. However, there is a problem that it is not possible to ensure the recognition accuracy. In addition, when a plurality of optical systems are arranged as described above, the system as a whole requires a large space and is expensive, and there is a problem in terms of maintainability because it is necessary to set illumination in each optical system. Further, in the method according to the above, when a laser displacement meter is used as the distance sensor, the distance between the sensor and the work is limited to a maximum of about 200 mm due to the limitation of the laser output, so that it can be used in practice. When an ultrasonic displacement meter is used as the distance sensor, the distance between the sensor and the work conforms to the conditions, but the work is randomly stacked because the in-plane range is narrow. There is a problem that accurate measurement cannot be performed when there is a work having a different height only in the part.

SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances and aims to solve the above-mentioned problems. Therefore, the recognition accuracy can be ensured in a wide range, and the work can be stably recognized even on a work randomly stacked on a pallet. It is an object of the present invention to provide an image capturing position setting method and a setting device that can perform the setting.

[0006]

In order to achieve the above-mentioned object, an image capturing position setting method according to a first aspect of the present invention recognizes an image of a workpiece by a three-dimensional visual sensor including a projector and a camera. In this method, a coordinate value in the height direction of the work is obtained at the time of recognizing an image, and the coordinate value is set within a measurable area set by the angle of view of the camera and the irradiation range of the projector. It is characterized in that it is determined whether or not a work is contained.

According to the present invention, since the image of the work is recognized based on the coordinate values in the height direction of the work recognized by the three-dimensional visual sensor, it is possible to always ensure stable recognition accuracy. It is possible to change the measurement height arbitrarily, and it is possible to ensure the recognition accuracy of the work in a wide range. Further, by using the three-dimensional visual sensor in this manner, three-dimensional coordinate values of the entire measurement area can be obtained, and even when the workpieces are randomly stacked, the accurate distance between the sensor and the workpiece can be obtained. The distance can be measured.

In the present invention, when the coordinate value in the height direction of the work reaches a threshold value set near the lower limit of the measurable area, the measuring position of the three-dimensional visual sensor is lowered by a fixed amount to thereby measure the measurable area. It is preferable to ensure By doing so, it is possible to always stably recognize a work stacked high on a pallet even if the load height changes due to depalletizing. Here, it is preferable that the lowering amount of the measurement position of the three-dimensional visual sensor is equal to or less than the height dimension of the measurable area in order to avoid a collision between the three-dimensional visual sensor and the work.

Next, the image capturing position setting device according to the second invention comprises: (a) a three-dimensional visual sensor having a projector and a camera; and (b) a workpiece when the three-dimensional visual sensor recognizes an image of the workpiece. Determining a coordinate value in the height direction of the camera, and determining from this coordinate value whether or not the workpiece is within a measurable area set by the angle of view of the camera and the irradiation range of the projector. It is a feature.

The present invention is an apparatus embodying the image capturing position setting method according to the first aspect of the present invention. In setting an image capturing position, an image of, for example, a slit pattern light emitted from a projector to a work is captured by a camera. Thus, when the work is recognized, the determination unit determines whether or not the work is within the measurable area of the three-dimensional visual sensor from the coordinate values in the height direction of the work.
In this way, stable recognition accuracy can always be ensured, and the work recognition accuracy can be ensured over a wide range by arbitrarily changing the measurement height.

In the present invention, furthermore, a detecting means for detecting that the coordinate value in the height direction of the work has reached a threshold value set near a lower limit of the measurable area, and receiving an output of the detecting means. It is preferable to provide a sensor moving unit that secures a measurable area by lowering the measurement position of the three-dimensional visual sensor by a predetermined amount, and in this case, the amount of decrease in the measurement position of the three-dimensional visual sensor is the measurable amount. The amount is equal to or less than the height dimension of the area.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, specific embodiments of an image capturing position setting method and a setting device according to the present invention will be described with reference to the drawings.

FIG. 1 is a system configuration diagram of an image capturing position setting device according to an embodiment of the present invention.

The present embodiment is applied to an operation of depalletizing a work W such as a cardboard box placed in a stacked state on a pallet 1 by a robot 2. The wrist 2a of the robot 2 is equipped with a three-dimensional visual sensor 3 including a projector that projects slit pattern light, and a CCD camera that is inclined with respect to the projection direction of the projector. Thus, an image based on the slit light projected on the surface of the work W from the projector of the three-dimensional visual sensor 3 is captured by the CCD camera and input to the image processing device 4. In the image processing device 4, the handling position of the work W by the robot 2 is calculated based on the input data, and the calculation result is output to the robot control device 5. In the robot control device 5, a control signal is transmitted to the robot 2 based on input data from the image processing device 4, thereby controlling the position of the wrist 2a.

Next, a processing flow for setting an image capturing position of the work W in the image processing apparatus 4 will be described with reference to a flowchart shown in FIG.

S1: Initially set the image capturing position 1.
As an example, as shown in FIG. 3, when the measurable area A of the three-dimensional visual sensor 3 is an area inside a cube whose one side is 500 mm, the cube 1200 mm below the measuring head of the three-dimensional visual sensor 3 Set the three-dimensional coordinate system in. In the handling of the work W loaded on the pallet 1, if the maximum height position of the work W is known, the image capturing position 1 is set so that the uppermost work W enters the measurable area A. I just need. In the measurable area A, if the work W exists in this area, the three-dimensional coordinate value of the work W can be measured. If the work W exists in other areas, an accurate three-dimensional coordinate can be obtained. Unable to get coordinate values.

S2: The wrist 2 of the robot 2 moves to the image capturing position 1 (solid line position in FIG. 1) set by the initial setting.
Move a. S3 to S5: The image of the work W is captured at the image capturing position 1, the three-dimensional coordinate value of the image capturing position 1 is calculated, and the work W is recognized. This recognition processing is performed using, for example, the processing of the distance image obtained by irradiating the slit pattern light from the projector onto the surface of the work W and the processing of the grayscale image obtained by fully opening the slit of the projector. Is preferred.

S6-S7: A process for recognizing the handling position of the work W is performed, and three-dimensional coordinate values of the handling position are taken in. Thereafter, the obtained coordinate values are converted from the sensor coordinate system to the robot coordinate system. It should be noted that the above-described processes in step S3 and subsequent steps can be performed in parallel with the operation of placing the workpiece W held by the hand unit of the robot 2 at a predetermined position.

S8: It is determined whether or not a signal (height change flag) for changing the measurement height of the three-dimensional visual sensor 3 is ON. When the process of step S8 is first performed, the process proceeds to the next step S9 because the height change flag is in the OFF state. S9: It is determined whether the z coordinate of the handling position is larger than the height threshold B. Here, the height threshold value B is determined by referring to FIG.
Is set near the lower limit of the stable measurement area C (= measurable area A) at the image capture position 1 and set within a range overlapping with a stable measurement area D at the image capture position 2 described later. Have been.

S10: When the z coordinate of the handling position is above the height threshold B, the wrist 2a of the robot 2 is moved to the handling position, and the workpiece W is gripped by the hand part of the robot 2. S11: It is determined again whether the height change flag is ON, and if not, the process returns to step S2.

S12-S13: When the z-coordinate of the handling position reaches the height threshold B in the judgment of step S9, the work W on the pallet 1 gradually decreases,
Since the height direction coordinate value of the three-dimensional coordinate value exceeds a value sufficient to ensure recognition accuracy, the robot controller 5
, A command to change the measurement height is output. Upon receiving the height change command, the robot controller 5 sets a height change flag in the memory. S14: If the height change flag is ON,
The wrist 2a of the robot 2 is moved to the image capturing position 2 (the chain line position in FIG. 1), and the measurement is performed by lowering the wrist 2a to a position where the recognition accuracy of the workpiece W can be ensured. Thereafter, the process returns to step S3. Here, the amount of reduction from the image capture position 1 to the image capture position 2 is determined by the measurable area A
(In this embodiment, 500 mm). Thus, collision of the three-dimensional visual sensor 3 with the workpiece W due to lowering of the three-dimensional visual sensor 3 can be avoided.

As described above, according to the image capturing position setting device of the present embodiment, in the image recognition of the works W stacked on the pallet 1 based on the three-dimensional coordinate values recognized by the three-dimensional visual sensor 3, Since the measurement is performed by lowering the three-dimensional visual sensor 3 to a position where the recognition accuracy can be sufficiently ensured, the recognition accuracy can be ensured in a wide range, and stable recognition can always be performed. Further, since the three-dimensional visual sensor 3 is used, it is possible to accurately obtain three-dimensional coordinate values of the entire measurement area, and even when the workpieces W are randomly stacked, the three-dimensional visual sensor 3 is used. There is an advantage that an accurate distance between the workpiece and the workpiece W can be measured.

In this embodiment, an example in which the present invention is applied to a depalletizing operation has been described. However, it goes without saying that the present invention can be applied to other work recognizing devices such as handling of works between conveyors.

[Brief description of the drawings]

FIG. 1 is a system configuration diagram of an image capturing position setting device according to an embodiment of the present invention.

FIG. 2 is a flowchart illustrating a processing flow for setting an image capturing position of a work;

FIG. 3 is an explanatory diagram of an initial setting process of an image capturing position.

[Explanation of symbols]

 Reference Signs List 1 pallet 2 robot 3 3D visual sensor 4 image processing device 5 robot control device A measurable area B height threshold C stable measurement area at image capturing position 1 D stable measurement area at image capturing position 2 W work

Claims (6)

[Claims] An image capturing position setting method for recognizing an image of a workpiece by a three-dimensional visual sensor including a projector and a camera, wherein a coordinate value in a height direction of the workpiece is obtained at the time of recognizing the image. An image capturing position setting method, comprising: determining from a value whether or not a workpiece is within a measurable region set by an angle of view of a camera and an irradiation range of a projector. 2. When the coordinate value in the height direction of the workpiece reaches a threshold value set near the lower limit of the measurable area, the measurable area is secured by lowering the measurement position of the three-dimensional visual sensor by a certain amount. The method for setting an image capturing position according to claim 1. 3. The image capturing position setting method according to claim 2, wherein the amount of decrease in the measurement position of the three-dimensional visual sensor is equal to or less than the height of the measurable area. 4. A three-dimensional visual sensor having a projector and a camera, and (b) obtaining a coordinate value in a height direction of the work when recognizing an image of the work by the three-dimensional visual sensor. An image capture position setting device, comprising: a determination unit configured to determine whether a workpiece is within a measurable area set by a camera angle of view and an irradiation range of a projector based on a value. 5. A detecting means for detecting that a coordinate value in the height direction of the workpiece has reached a threshold value set near a lower limit of the measurable area, and a three-dimensional image receiving means for receiving an output of the detecting means. 5. The image capturing position setting device according to claim 4, further comprising: a sensor moving unit that secures a measurable area by lowering a measurement position of the visual sensor by a predetermined amount. 6. The image capturing position setting device according to claim 5, wherein the amount by which the measurement position of the three-dimensional visual sensor is lowered is equal to or less than the height of the measurable area.