DE202007002260U1 - Optoelectronic device - Google Patents

Abstract

Optoelectronic Device (1) with a camera (5), several, this in different positions associated lighting devices (2a to 2d) and with a Evaluation unit (11), wherein for the detection of objects (6) in the Evaluation unit (11) at least one difference image from two images is formed, which when activated different lighting devices (2a to 2d) are generated in the camera (5), and wherein object features and / or object coordinates from those determined in the difference image Shade derived from objects (6).

Description

The The invention relates to an optoelectronic device.

Optoelectronic Devices of the type in question provide image processing Systems are and have a camera for taking pictures. For object detection can also be a lighting device be provided to illuminate the objects to be detected.

Known Optoelectronic devices of this type are designed as so-called sectional image sensors. The lighting device becomes a specific line pattern generated. First, will then the object-free area to be detected is registered with the camera. Then the area is registered with the camera when a Object located in this area. For object detection then becomes compared this image with the image of the object-free space. there is the offset of the line pattern caused by the object is, registered and evaluated in an evaluation unit.

adversely in such optoelectronic devices is that the generation suitable line pattern means for structuring the illumination device emitted light presupposes what the design effort of the optoelectronic device increases. Furthermore, it is disadvantageous that the line patterns depending on the object shape and object position only on one Part of the object surface can be projected so that a reliable and complete Object detection not possible is.

Of the Invention is based on the object, an optoelectronic device of the type mentioned above, by means of which as possible gerin constructive effort a reliable and accurate detection allows objects becomes.

to solution This object, the features of claim 1 are provided. advantageous embodiments and appropriate training The invention are described in the subclaims.

The Optoelectronic device according to the invention includes a camera, several, this in different positions associated lighting devices and an evaluation unit. For the detection of objects in the evaluation is at least a difference image formed from two images, which when activated different lighting devices are generated in the camera. From the shadows of objects determined in the difference image Derived object features and / or object coordinates.

By the detection according to the invention the shadow of objects can detailed information about whose shapes and positions are obtained in space.

Essential here is that for this shadow analysis generates differential images of two images be in the camera for various active lighting devices are obtained. There when taking individual pictures each different and in different positions to the camera lying lighting devices become activated, becomes one after the triangulation principle working Optoelectronic device obtained. By the evaluation of the Shadows in the difference image can Thus, not only object outlines are determined, but also the Positions, that is the positions and orientations of objects can be determined.

The Camera itself consists of a matrix-like arrangement of receiving elements. The The difference image of two pictures becomes thereby through the formation of the differences the reception signals of the receiving elements of the won both pictures.

The Lighting devices can be formed in the simplest case as point light sources. alternative can line-shaped Lighting devices are used. Generally, the arrangement the lighting devices relative to the camera to the respective Application adapted.

Especially the optoelectronic device according to the invention is advantageous adapted for detection of objects in a rectangular container. The camera and lighting equipment are located above of the container open at the top. In the simplest case, the lighting devices are in the form of Point light sources are arranged above diagonally opposite corners of the container. Particularly advantageous is above each side wall of the container a parallel to this running line-shaped illumination device intended. By activation of two opposite each other Lighting devices, a difference image is formed in each case. To increase The accuracy of the detection advantageously becomes the sum of both difference images educated. With the optoelectronic device can not only the layers, that is Positions of objects in the container but also, where appropriate, gaps between the individual objects. By the inventive shadow analysis can thereby even very narrow columns are captured.

The The invention will be explained below with reference to the drawings. It demonstrate:

1 : Block diagram of the optoelectronic device.

2 : Arrangement of the optoelectronic device at a commission place with a container.

3 : Arrangement of the optoelectronic device in a first container ventilation with illumination from the right.

4 : Arrangement of the optoelectronic device in a second container filling with illumination from the right.

5 : Silhouettes for the arrangement according to 4 ,

6 : Arrangement of the optoelectronic device according to 4 with lighting from the left.

7 : Silhouettes for the arrangement according to 6 ,

8th : Silhouette for the difference image of the single images according to 5 and 7 ,

9a : Additive silhouette for the bin arrangement according to 4 and 6 ,

9b : First section of the silhouette according to 9a ,

9c : Second part of the silhouette according to 9a ,

10 : Schematic representation for the detection of small objects in a container with the optoelectronic device according to 1 ,

11 : Diagram for calculating object heights of objects in a container.

12 : Beam path for the arrangement according to 2 with empty container.

13 : Beam path for the arrangement according to 2 in an object arranged in the container.

14a : Arrangement according to 3 with lighting with a barcode pattern.

14b : Camera registered illumination pattern according to the arrangement 14 without container.

14c : Camera registered illumination pattern according to the arrangement 14 with empty container.

14d : Camera registered illumination pattern according to the arrangement 14 with filled container.

15a : Arrangement according to 3 with sinusoidally varying illumination.

15b : Camera registered illumination pattern according to the arrangement 15 without container.

15c : Camera registered illumination pattern according to the arrangement 15 with empty container.

15d : Camera registered illumination pattern according to the arrangement 15 with filled container.

1 schematically shows the structure of an embodiment of an optoelectronic device 1 , The optoelectronic device 1 has a predetermined number of lighting devices 2a to 2d on. The of the lighting devices 2a to 2d emitted transmitted light rays 3 serve to illuminate an object scene 10 , The of the object scene 10 as reception light beams 4 back reflected transmitted light rays 3 be in a camera 5 registered. The camera 5 has a matrix-like arrangement of receiving elements. The camera is preferred 5 designed as a CMOS or CCD sensor. The lighting equipment 2a to 2d lie in different positions to the camera 5 and are from an evaluation unit 11 individually activated by an image processing unit implemented on a microprocessor. The lighting equipment 2a to 2d can be formed in the simplest case of point light sources such as LEDs or laser diodes. In the present case are line-shaped lighting devices 2a to 2d consisting of a series arrangement of individual point light sources, in particular light-emitting diodes, provided. In the present case, the optoelectronic device 1 four identical lighting devices 2a to 2d on. The evaluation of the camera 5 register pictures of the object scene 10 takes place in the evaluation unit 11 , The object information obtained is via an interface 9 output.

When Object information becomes object features such as object outlines as well Object coordinates as information about the sizes and locations of objects won.

The extraction of the object information takes place in the form of a shadow analysis of the objects. In order to determine such shadows of objects, a difference image of two images is formed which is at different activating illumination directions 2a to 2d in the camera 5 be won. The term shadow is always below the visual shadow of the camera 5 meant. The difference image is obtained by the fact that in the evaluation 11 the differences of the receive signals of the receiving elements of the two images present as gray values are calculated. From this difference image, the shadows of the objects are determined, for which purpose, in a known manner, a grayscale adjustment and increase in contrast in the evaluation unit 11 be performed.

The shadow of an object is generally of illumination shadow 12 formed, that is zones, in which by shading of objects no transmitted light of the respective lighting devices 2a to 2d arrives. Furthermore, the shadows are also from the field of view of the camera 5 formed, that is zones of the light field of the camera 5 that are shadowed by objects.

It is essential that a difference image is always generated from two images that are at different activated lighting devices 2a to 2d are obtained to eliminate by contrasts or the like resulting contrasts. Due to the different positions of the lighting devices 2a to 2d Thus, a device operating on the principle of triangulation is obtained for the camera, that is, by evaluating the shadow structures, information about the positions of the objects is obtained.

To increase the detection reliability of the optoelectronic device 1 Advantageously, the sums of different difference images are formed. Due to the different arrangements of lighting devices 2a to 2d to the camera 5 Thus object contours can be completely captured.

In 2 is a preferred application of the optoelectronic device 1 according to 1 shown.

2 shows a picking place with a container 7 , which is equipped with transport goods, in rectangular packaging, hereinafter referred to as objects 6 referred to, are stacked in layers. A conveyor belt moves the container 7 in z-direction to the picking station, where the container 7 remains in rest position. About this rest position is the optoelectronic device 1 with the camera 5 and the lighting devices 2a to 2d , One after the other the container will be 7 with the objects 6 through the lighting devices 2a to 2d lit and in each case a picture with the camera 5 added. For each image, at least one object edge provides a shadow. By subtracting two images each, ie generating a difference image, the shadows are distinguished from other object structures, such as a print. By adding differentiated difference images, the outline of the object 6 determined and determined from the shadow width, the object height. After picking up the image and calculating the room coordination, a picking robot pans 8th between the optoelectronic device 1 and the container 7 and lifts that through the optoelectronic device 1 determined object 6 at. After that can be another loading or unloading by other objects 6 in or out of the free spaces of the container 7 respectively.

3 shows a longitudinal section through the arrangement according to 2 , being in the container 7 only one object 6 lies. How out 3 is only the lighting device 2c activated. With this, the interior of the container is illuminated obliquely from above. The field of vision 5a the camera 5 captures the entire container surface. By the obliquely coming from above lighting with the lighting device 2c creates an object shadow 12 of the object 6 , which is exploited for object detection. Furthermore, created by the shading of the transmitted light beams 3 the lighting device 2c through the left container wall and by shading the field of view 5a the camera 5 a container overhang shadow through this container wall 14 ,

The 4 to 8th illustrate the principle of object detection for the device according to 2 , The show 4 and 6 each the container 7 with a particular container filling comprising objects 6 . 6a . 6b . 6c ,

To generate a first image, as in 4 represented, the lighting device 2c activated. In this case, for the object 6 an object shadow 12 receive. For the spaces between the objects 6a . 6b . 6c become slit shadows 13a . 13b . 13c receive. In addition, the container projection shadow 14 receive.

To generate a second image is that of the lighting device 2a opposite lighting device 2c activated ( 6 ). Here are the spaces between the objects 6a . 6b . 6c the object shadows 13a . 13b receive. In addition, the container projection shadow 14 receive ( 7 ).

From the two pictures according to the 5 and 7 the difference image is generated. This difference image, in which the shadows outside the container 7 is eliminated in 8th shown. This silhouette already provides accurate information about the locations and sizes of the objects 6 in the container 7 ,

To get complete information about the contours and positions of the objects 6 in the container, in addition to the difference image according to 8th in which the difference of the individual images with activated lighting devices 2a . 2c was formed, a difference image of individual images, in which the difference of the individual images with activated lighting devices 2 B . 2d is formed, generated.

The sum of both difference pictures is in 9a shown. The container overhang shadows 14 define the location of the container 7 , the object shadows 12 the positions of the objects 6a . 6b . 6c ,

To prepare the detection of the column between the objects 6a . 6b . 6c will the in 9b marked area 6 ' formed the objects 6a . 6b and 6c includes, but the associated shadows 12 safely excludes. This area is rotated in the image evaluation so that the object edges are orthogonal.

9c shows the area 6 ' the objects 6a . 6b and 6c , which is scanned column by row or line by line. The column or line with a very low gray value marks the slit shadows 13a and 13b , from whose position the gap between two objects 6a . 6b is determined.

10 shows the arrangement according to 2 with the object 6 smallest dimension, which of the optoelectronic device 1 is still detectable. The object 6 with the minimum dimensions is located in the lowest filling height in a container corner. It must be ensured that the shadow 14 and the object shadow 12 still distinguishable. This arrangement determines the location of the lighting devices 2a to 2d in which the object 6 with a silhouette analysis is still detectable.

With the optoelectronic device 1 according to 1 can according to the arrangement 2 with the silhouette analysis in the 4 to 9 is explained, the object heights and thus the filling heights of the container 7 be determined.

In 11 the relevant dimensions are defined and the reference to the shadows 12 . 13 and 14 produced. Starting from the known container height y_7 is using the Behälterüberstandsschattens 14 the distance y_14 determined from the edge of the container to the current highest filling level. From the width x_12 of the object shadow projected to the baseline 12 the object height y_6 is determined.

For the calibration of object heights depending on the shadow widths, as in 12 shown, the container overhang shadow 14 with empty container 7 be recorded.

x14':

der auf die Grundfläche projizierte Schatten 14' des leeren Behälters 7,

y7:

die Höhe des Behälters 7

x_Basis:

der Abstand zwischen Kamera 5 und Sender 2

y5:

die Höhe der Kamera 5 über der Grundfläche

It applies the relationship x14 '= y7 x Base / (y5 - y7) is there

x14 ':

the shadow projected onto the base 14 ' of the empty container 7 .

y7:

the height of the container 7

x _basis :

the distance between camera 5 and transmitter 2

y5:

the height of the camera 5 over the base

13 shows the calculation of the object height from the shadow length 12 ,

y6:

die Höhe des Objektes 6

x12:

der auf die Grundfläche projizierte Schatten 12 des Objektes 6,

x14:

der auf die Grundfläche projizierte Behälterüberstandsschatten 14,

x_Basis:

der Abstand zwischen Kamera 5 und Sender 2.

It applies the relationship y6 = x12 x14 / x Base is there

y6:

the height of the object 6

x12:

the shadow projected onto the base 12 of the object 6 .

x14:

the container overhang shadow projected onto the base 14 .

x _basis :

the distance between camera 5 and transmitter 2 ,

The shadow length x12 is independent of the position of the object 6 in X direction.

In the embodiments according to 3 to 13 for the detection of objects 6 in the container 7 the arrangement according to 2 is preferred with lighting equipment 2a to 2d worked, whose transmitted light rays 3 create a homogeneous lighting.

This is for a reliable detection of the objects 6 at least as long as the container 7 not complete with objects 6 is filled.

Is the container 7 brimming with objects 6 filled, it may be required by the lighting equipment 2a to 2d emitted transmitted light rays 3 to structure, that is spatially varying form. Examples are in the 14 and 15 described. The show 14a . 15a in each case the optoelectronic device 1 according to 2 in which the illumination device is activated and spatially varying transmitted light beams 3 emitted. The remaining lighting devices emit corresponding transmitted light beams 3 ,

The shadow length x12 is independent from the position of the object 6 in X direction.

14a shows an embodiment of the optoelectronic device 1 with lighting equipment 2a to 2d , whose transmitted light rays 3 a barcode pattern on the base of the container 7 project.

14b shows the pattern that the camera 5 sees if no container 7 is available. Through the container 7 it changes through the camera 5 taken picture like in 14c shown.

Through the object 6 in the container 7 results in the image pattern accordingly 14d , where the area MB6 of the pattern on the object 6 falls, in the taken picture by the amount dx6 shifted appears. From the displacement dx6, the object height y6 can be calculated. From the degree of compression in the x-direction compared to the pattern after 14c can directly specify the y-coordinate as the height of the object 6 be determined over the base area.

15a shows an embodiment of the optoelectronic device 1 , in which the illumination is designed so that the light intensity in the x-direction varies sinusoidally. Through the shadow 12 an object 6 results in the displacement, or gap dx6, from which the object height y6 can be calculated. In case the object 6 close to the wall of the container 7 rests and no shadow 12 the phase shift dxp of the light intensity can be used to determine the object height y6.

15b shows the over x varying gray value as seen through the camera 5 if there is no container 7 is detected.

15c shows the collapse of the gray value by the shadow 12 with the shadow width dx6.

To determine the gray value phase shift in the x direction, as in 15d shown, the gray scale of 15b moved so far, until he in the area MB6 with the recorded course of 15c covers. The displacement dxp then corresponds to the value dx6, from which the object height can be calculated.

1

Optoelectronic contraption

2a

lighting device

2 B

lighting device

2c

lighting device

2d

lighting device

3

Transmitted light beams

4

Receiving light rays

5

camera

5a

field of view

6

object

6a

object

6b

object

6c

object

6 '

Area

7

container

8th

picking robot

9

interface

10

object scene

11

evaluation

12

object shadow

13

shadow gap

14

Container supernatant shadow

Claims (18)

Optoelectronic device ( 1 ) with a camera ( 5 ), a plurality of lighting devices associated therewith in different positions ( 2a to 2d ) and with an evaluation unit ( 11 ), whereby for the detection of objects ( 6 ) in the evaluation unit ( 11 ) at least one difference image is formed from two images, which upon activation of different illumination devices ( 2a to 2d ) in the camera ( 5 ), and wherein object features and / or object coordinates from the shadows of objects determined in the difference image ( 6 ) be derived. Optoelectronic device according to claim 1, characterized in that the camera ( 5 ) has a matrix-like arrangement of receiving elements. Optoelectronic device according to claim 2, characterized in that the camera ( 5 ) is a CMOS or CCD sensor. Optoelectronic device according to one of claims 2 or 3, characterized in that for generating a differential image for the two images, the differences of the present as gray values received signals for the individual receiving elements of the camera ( 5 ) are formed. Optoelectronic device according to one of claims 1 to 4, characterized in that in the evaluation unit ( 11 ) the sum of several difference images is formed, which are generated by means of different pairs of illumination devices ( 2a to 2d ) are generated. Optoelectronic device according to one of claims 1 to 5, characterized in that the lighting devices ( 2a to 2d ) are each formed by a point light source. Optoelectronic device according to claim 6, characterized in that the Punktlichtquel are each formed by a light emitting diode or a laser diode. Optoelectronic device according to one of claims 1 to 5, characterized in that the lighting devices ( 2a to 2d ) are each formed by a row-shaped arrangement of light sources. Optoelectronic device according to one of claims 1 to 8, characterized in that it above a rectangular container ( 7 ) for the detection of objects arranged therein ( 6 ) is installed. Optoelectronic device according to claim 9, characterized in that above each side wall of the container ( 7 ) a lighting device ( 2a to 2d ) is installed. Optoelectronic device according to claim 10, characterized in that a difference image is generated from two images which, when activated, are assigned to illumination devices (2) associated with opposite side walls. 2a to 2d ). Optoelectronic device according to Claim 9, characterized in that two lighting devices designed as point light sources are provided which are located above diagonally opposite corners of the container ( 7 ) are arranged. Optoelectronic device according to one of claims 1 to 12, characterized in that the height of an object ( 6 ) from the width of the shadow of this object ( 6 ) is determinable. Optoelectronic device according to one of claims 1 to 13, characterized in that from the outlines of shadows the orientations of objects ( 6 ) are determinable. Optoelectronic device according to one of claims 13 or 14, characterized in that from the height of an object ( 6 ) and the outline, the object coordinates can be determined in a plane. Optoelectronic device according to one of claims 12 to 15, characterized in that the layers of objects ( 6 ) within the container ( 7 ) are determinable. Optoelectronic device according to claim 16, characterized in that columns between objects ( 6 ) are determinable. Optoelectronic device according to one of the claims 16 or 17, characterized in that as parameter the shadow of the container ( 7 ) can be taught in a teach-in process.