JPH06331334A - Image recognition device and part-supply device using it - Google Patents

Abstract

PURPOSE:To select an article in a specific attitude accurately. CONSTITUTION:A plurality of parts are placed on a recognition table 2 with a sample area for placing, in a specific shape, samples in the same shape as that of parts P to be recognized, light is applied to them by a lighting means 4, and then the reflection light images are picked up by an image pick-up means 5. The image picked up by the image pick-up means 5 is binarized and then a means 61 for detecting features which detects the area and direction of each part image, a means 62 for setting threshold value which sets the threshold value for the binarization so that the binary-coding area of the sample part image becomes a prescribed area value, and a means 63 for updating the threshold value which allows the sample part image to be taken at a specific period and then the means for setting the threshold value to set the threshold value again are provided. The part image whose area is within the allowable tolerance of a prescribed area value is selected out of the part images detected by the means 61 for detecting the features and then a transfer means 3 picks up parts according to the position and direction of the corresponding part and then aligns the directions to a specific position and transfers them.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a device for recognizing the posture of an object to be recognized, and a component having the same posture taken out from a large number of parts scattered in various postures and sequentially supplied to other devices. It is applied to supply equipment.

[0002]

2. Description of the Related Art Conventionally, various methods for detecting the posture of an object by optical means have been proposed. In Japanese Unexamined Patent Publication (Kokai) No. 4-141325 cited as an example thereof, a predetermined position of a parts transporting track of a parts feeder is constituted by a translucent plate, an illuminating device is provided above and below the translucent plate, and image detection is performed either above or below it. A component supply device is disclosed, in which the respective means are arranged. This device captures a silhouette image of a component conveyed on a transparent plate and recognizes the posture and position of the component, or in the case of a component whose posture cannot be specified by the silhouette image, a threshold for binarization. The values are recognized in correspondence with the reflected light image to cope with the attitude recognition of various parts.

[0003]

In such a component supply device, the binarized area of an image sometimes becomes inconsistent with the actual state due to influences of fluctuations in illumination light, changes over time, scattered light incident from the outside, and the like. There were times when parts with different attitudes were supplied. An example of a method for coping with this is the method described in Japanese Patent Laid-Open No. 4-141384. That is, the calibration jig is positioned on the light-transmitting plate, the hole of the known area opened in the jig is image-recognized, and the threshold value is set so that the binarized area of the hole becomes equal to the known area. This is a method of setting so that the binarized area of the image does not largely deviate from the actuality even if the illumination light changes with time. The present invention provides a component supply device which does not use such a special jig and can cope with a change in illumination condition in a short time.

[0004]

According to the present invention, an illuminating means for irradiating an object to be recognized with light, an imaging means for picking up a reflected light image from the illuminating means, and a sample having the same shape as the object have a specific posture. A sample holding section for holding the sample is provided. And
The characteristic detection means for binarizing the reflected light image captured by the imaging means to detect the area of the image and the threshold value for binarizing the reflected light image are set in advance by the binarized area of the reflected light image from the sample. A threshold setting means for setting the area equal to the specified area value, and a threshold for causing the reflected light image of the sample to be captured by the imaging means at a predetermined cycle and causing the threshold setting means to reset the threshold. By providing the value updating means and determining whether or not the reflected light image area of the object detected by the feature detecting means is within the allowable tolerance of the specified area value, the attitude of the object is the same as that of the sample. Recognize whether or not.

Further, in the present invention, a plurality of parts are placed,
A recognition table having a sample area for mounting a sample having the same shape as the part in a specific posture on its part, a transfer means for picking out the part on a predetermined position, and an illumination means for irradiating the recognition table with light. An image pickup unit for picking up a reflected light image from the image pickup unit is provided. Then, the reflected light image captured by the imaging means is binarized, the individual component images in the imaging screen are distinguished, and the feature detection means for detecting the area and orientation of each component image, and the threshold for performing the binarization. A threshold value setting means for setting a value so that the binarized area of the reflected light image of the sample is equal to a predetermined area value; and the reflected light image of the sample is captured by the imaging means at a predetermined cycle, The threshold value setting means is provided with a threshold value updating means for resetting the threshold value, and among the component images detected by the feature detecting means, one having an area within the allowable tolerance of the specified area value is selected, The transfer means picks up the component according to the position and orientation of the corresponding component, and aligns the component at a predetermined position for transfer.

[0006]

The image of the part sample held in a specific posture is taken in from a predetermined direction at a constant cycle, and the image area of the taken sample matches the prescribed area that the part image in that posture should take. , The threshold for binarization is updated every cycle. Then, the part image is recognized with the updated threshold value, and only the part having the same posture as the sample is taken out to a predetermined position.

[0007]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the image recognition apparatus of the present invention and the component supply apparatus using the same will be described. FIG. 1 is an explanatory diagram showing the main configuration of the component supply device according to the present embodiment. The component supply device 1 includes a recognition table 2, a robot 3, a lighting unit 4, an image pickup unit 5, an image recognition unit 6, and a control unit 7. It is configured.

The recognition table 2 has a rectangular cylindrical frame body 21.
And a protrusion 22 provided at the same height on the inner surface of the frame body 21.
The transparent plate 23 is supported by the robot base 31 and is attached to the robot base 31 via a swing mechanism (not shown). The translucent plate 23 has the same area as the horizontal cross section of the inner space of the frame body 21, is fixed to the inner lid shape at a position slightly lower than the upper end surface of the frame body 21, and the component P having the same shape is provided on the upper surface thereof. It is randomly scattered from a hopper (not shown). FIG. 3 is a plan view showing this state. A partition plate 231 having an L-shaped cross section and having a light-transmitting property is provided at one corner on the upper surface of the light-transmitting plate 23.
Is attached, and the space surrounded by this and the inner wall surface of the frame 21 is defined as a sample area 230.

The illuminating means 4 is supported on the left and right inner side surfaces in FIG. 1 at a position further lower than the projecting portion 22 of the frame 21, and irradiates light from below so as to uniformly illuminate the translucent plate 23. ing. Reference numeral 41 is a reflecting plate that reflects the light of the illumination means 4 upward, but at the same time, it also functions to shield the light downward.
The image pickup means 5 is held right below the translucent plate 23 so as to face upward,
The optical axis passes vertically through the center of the transparent plate 23, and is arranged at a focal length such that the entire transparent plate can be taken into the visual field.

The robot 3 is a horizontal articulated robot, and is equipped with a chuck 31 as an end effector to take out a component P and supply it to another device (not shown) or directly attach it to a work positioned on a conveyor. . The chuck 31 moves to an arbitrary position on the translucent plate of the recognition table 2 by the turning motion of the robot arm,
The component P is sandwiched and lifted at an arbitrary θ angle by being given vertical movement and θ rotation about the vertical movement axis.

The image recognition section 6 comprises a feature detecting means 61, a threshold setting means 62 and a threshold updating means 63. The feature detection means 61 is the illumination means 4 captured by the imaging means 5.
The reflected light image by is binarized, the images of the individual components P in the imaging screen are distinguished (labeled), and the area and orientation of each component image are detected (feature extraction). This labeling and feature extraction are not the gist of the present invention, and thus detailed description thereof is omitted. The threshold setting means 62 is the feature detecting means 6
The threshold for setting binarization in 1 is set, and 2 in the reflected light image of the component P (hereinafter, referred to as the component Ps) placed in a specific posture in the sample area 230 described above. The threshold value is set so that the value-converted area is equal to the area of the surface of the component P in that posture that is in contact with the light-transmitting plate 23 (hereinafter, referred to as a prescribed area), which is calculated in advance. The threshold updating means 63 captures the reflected light image of the part Ps in the sample area at a predetermined cycle.
Then, the threshold value setting means 62 is caused to reset the threshold value.

The control unit 7 selects one of the component images detected by the characteristic detecting unit 61, the binarized area of which is within the allowable tolerance of the specified area value described above, and the characteristic detecting unit 61 detects the image of the binarized area. The center of gravity position and orientation are detected. Then, the control unit 7 causes the robot 3 to sequentially pick up the component P and supply it to another device (not shown) according to the position and orientation of the selected component in the recognition table 2, or directly onto the transport conveyor. It can be incorporated into a positioned work.

The operation of this embodiment will be described with reference to the drawings. FIG. 2 is an enlarged perspective view of a component P used in the present embodiment. As shown in FIG. 3, the component P is randomly scattered on the translucent plate 23 of the recognition table 2 from a hopper (not shown) or the like. There is. Also, sample area 2 of the recognition table
A component Ps in which the component P has a specific posture (the posture shown in FIG. 2) is placed inside the unit 30. The posture sample Ps is restrained on the transparent plate 23 by a clamp mechanism or the like (not shown).

The group of parts on the transparent plate and the posture sample Ps are
The illumination means 4 and its reflection plate 41 are illuminated from below via the translucent plate 23, and the reflected light from them is taken into the imaging means 5. FIG. 4 is a diagram showing a captured reflected light image. The captured image information is sent to the feature detection means 61 of the image recognition unit 6 and labeled. That is, since there are a plurality of images to be recognized in one screen, individual images are distinguished before proceeding to shape analysis, and shape feature extraction is performed for each distinguished image.

When the feature detecting means 61 performs feature extraction of an image, first, the threshold setting means 62 causes the feature detecting means 6 to operate.
In order to set the threshold value for binarization in 1,
As shown in the upper side of FIG. 5, the vertical axis is the number of pixels N and the horizontal axis is the density D, and a density histogram of the image in the sample area 230 is created. Then, a density value at which the cumulative distribution from the smaller density value (the area occupied by the posture sample image Vs) is equal to the predetermined area calculated in advance is found, and this is set as the threshold value d1.

The image V1 in the area labeled L1 will be described as an example. The feature detection means 61 is
The grayscale image in the L1 region is binarized with the threshold value d1 set by the threshold value setting means 62, and the binarized area V1 of the smaller density is calculated. Then, the control unit 7 determines that the part P in the region L1 has the same posture as the posture sample Ps when the area V1 is within the tolerance of the specified area described above. The feature detection means 61 also calculates the binarized area for the other similarly labeled areas, and the control unit 7 determines the posture, and the feature detection means 61 determines the center of gravity of the component only for those having the same posture. Detect position and orientation. In this way, the control unit 7 obtains the position and orientation data regarding the component having the same orientation as the orientation sample.

The control unit 7 positions the chuck 31 of the robot 3 directly above the part P in the L1 region, for example, according to the position and orientation of the acquired same-position part in the recognition table 2. Then, the chuck 31 is lowered by the necessary amount θ according to the orientation data so that the tip of the chuck 31 sandwiches the web portion of the component P, and is lowered onto the transparent plate 23 to clamp the component P. Further, the control unit 7 raises the chuck 31 holding the component P, causes the robot 3 to supply the robot 3 by aligning it in another device (not shown) in a predetermined direction, or directly incorporating the work into a workpiece positioned on the conveyor. To be

When all the parts having the same posture as the posture sample Ps on the transparent plate 23 are removed by the robot 3,
The control unit 7 drives a swing mechanism (not shown) to eject the component group in another posture, and newly supplies the component to the hopper. The control unit 7 drives the threshold value updating means 63 for each component supply cycle or every several component supply cycles, and causes the threshold value setting means 62 to reset an appropriate threshold value. For example, as shown in the lower side of FIG. 5, when there is a deviation in the density histogram of the sample area, the threshold value setting means 62 (in this case, the image density is dark as a whole) is Similarly, a threshold value d2 is set such that the area occupied by the posture sample image Vs becomes equal to the predetermined area calculated in advance. This threshold value d2 is a density value that is shifted from the threshold value d1 shown on the upper side of FIG. In this way, the component supply device 1 continues to supply the component in the specific posture by the image recognition based on the appropriate threshold value.

In the present embodiment, the illumination means and the image pickup means are fixed below the recognition table, and the reflected light image of the component is picked up,
The configuration is such that the determined parts are taken out by the robot from above the recognition table. However, the illumination means and the imaging means are attached to the robot arm to capture the reflected light image from above the parts, or the imaging means is attached to the robot arm to detect the parts. The silhouette image may be picked up from above.
Further, in the present embodiment, the one in which a component having a specific posture is taken out from a large number of component groups has been described, but the image recognition apparatus of the present invention is not limited to this embodiment, and the recognition target object, the image pickup means, and the illumination means. If the positional relationship between and is constant, it can be applied to an apparatus that determines the posture of each individual article.

[0020]

According to the present invention, the image of the part sample held in a specific posture is captured at a constant cycle, and the captured image area matches the specified area of the component in that posture.
Since the threshold value for binarization is updated every cycle, even if the gray level of the image changes due to a change with time of the illuminating means or a change of external illumination light, the same posture as that of the sample is obtained. There are few mistakes in component selection.

[Brief description of drawings]

FIG. 1 is an explanatory diagram showing a configuration of a component supply device according to an embodiment of the present invention.

FIG. 2 is a perspective view of components used in the embodiment.

FIG. 3 is a perspective view illustrating a state of a component group on a recognition table.

FIG. 4 is a diagram showing a reflected light image of the component group of FIG. 3 captured by an image capturing unit.

FIG. 5 is an explanatory diagram showing a threshold setting / updating method.

[Explanation of symbols]

 1 Parts Supply Device 2 Recognition Table 3 Horizontal Articulated Robot (Transfer Means) 4 Illumination Means 5 Imaging Means 6 Image Recognition Unit 61 Image Recognition Unit 61 Feature Detection Means 62 Threshold Setting Means 63 Threshold Updating Means 7 Control Parts P Parts V Parts Reflected light image

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁵ Identification number Office reference number FI technical display location G06F 15/62 400 9287-5L 15/64 400 J 15/70 355 8837-5L // B65G 47 / 90 Z 8010-3F

Claims (2)

[Claims] 1. A device for imaging an object to be recognized and recognizing the posture of the object, the illumination unit irradiating the object with light from a predetermined direction, and a reflected light image from the object. An image capturing unit that captures the image from a predetermined direction, a sample holding unit that holds a sample having the same shape as the target object in a specific posture, the reflected light image captured by the image capturing unit is binarized, and the area of the image And a threshold value for binarizing in the feature detecting means are set so that the binarized area of the reflected light image from the sample becomes equal to a predetermined area value. A threshold value setting unit; and a threshold value updating unit that causes the threshold value setting unit to reset the threshold value by capturing the reflected light image of the sample in the imaging unit at a predetermined cycle. Detection means The reflected light image area of the object to be detected is characterized by recognizing whether or not the posture of the target object is the same as that of the sample by determining whether or not the reflected light image area of the target object is within the tolerance of the specified area value. Image recognition device. 2. A component supply device for image-recognizing a plurality of components and supplying only a component in a specific posture to a predetermined position, wherein a plurality of components are placed and the component is part of itself. A recognition table having a sample area for placing a sample of the same shape in a specific posture, a transfer means for taking out components on the recognition table to a predetermined position, and irradiating the recognition table with light from a predetermined direction. Illuminating means, imaging means for imaging the reflected light image from the recognition table from a predetermined direction, and the reflected light image captured by the imaging means is binarized to distinguish individual component images in the imaging screen, The feature detecting means for detecting the area and orientation of each component image and the threshold value for binarizing in the feature detecting means are set so that the binarized area of the reflected light image of the sample is equal to the area value defined in advance. And a threshold value setting unit that causes the image pickup unit to capture the reflected light image of the sample in a predetermined cycle, and causes the threshold value setting unit to reset the threshold value. Among the part images detected by the updating means and the feature detecting means, one having an area within the allowable tolerance of the specified area value is selected, and the parts are transferred to the transfer means according to the position and orientation of the corresponding parts. A component supply device, comprising: a control unit that causes the component to be picked up and transferred to a predetermined position with its orientation aligned.