JPS62185487A - Picture processor - Google Patents

Abstract

PURPOSE:To extract a detected object stably without receiving the influence from the change in an external illumination or the like by detecting the detected object from the difference in lightness and the difference in chromaticity, synthesizing two pictures of the result thereof, and obtaining the final picture of the detected object. CONSTITUTION:The first comparison circuit 16 compares the lightness data signal (D1) of the respective picture elements of a lightness memory 14 with the first set value R1. To the output terminal of the comparison circuit 16, a picture binarization circuit 18 for binarizing the picture from '0', '1' signals is connected and the binarization picture signal BS1 obtained in this circuit 18 is stored in a binarization picture memory 19. The second comparison circuit 20 compares the chromaticity parameter signal D2 of the respective picture elements from a lightness memory 15 with the second set value 2. To the output terminal of the comparison circuit 20, a picture binarization circuit 22 for binarizing the picture from '0', '1' signals is connected and the binarization picture signal BS2 obtained in this circuit 22 is stored in a binarization picture memory 23. To the output terminals of the two binarization picture memories 19, 23, a picture synthesizing circuit 24 is connected and this circuit synthesizes the two binarization pictures to obtain the synthesized picture for a gasoline filling port 3.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to an image processing device, and more particularly to an image processing device used when detecting the position of an object on the body of a vehicle.

(Prior Art) As the above-mentioned type of image processing device, for example, Japanese Patent Laid-Open No. 6
As disclosed in Japanese Patent No. 0-91204, image processing is known in which an image brightness signal from an imaging device is compared with a set value and converted into a binary image that distinguishes a detected object from a background based on the difference in brightness. ing.

(Problems to be Solved by the Invention) However, in the case of a workpiece extraction method using feature extraction parameters simply based on the brightness level difference, as in the above-mentioned conventional 'A', when the difference in brightness level between the workpiece and the background becomes small, Feature extraction becomes difficult and workpiece extraction becomes impossible. Furthermore, the difference in brightness level also changes due to changes in external lighting, making extraction of the workpiece unstable and causing erroneous recognition.

Furthermore, if the work and the background differ only in chromaticity and the like but have the same lightness level, it becomes completely impossible to extract the work.

Therefore, the present invention provides an image that can stably detect objects without being affected by changes in external illumination, etc., and can also detect objects that have the same brightness level as the background but different chromaticity. The purpose is to provide a processing device.

(Means for Solving the Problems) An image processing device according to the present invention compares an image brightness signal from an imaging device that images the body of a vehicle with a first setting value, and detects detected objects on the body based on the difference in brightness. a first image processing means for converting the detected object into a first binary image that distinguishes the detected object from the background; The present invention is characterized by comprising a second image processing means for converting the image into a second binary image, and an image synthesizing means for synthesizing the first and second binary images to obtain a final image.

(Operations and Effects of the Invention) As described above, the image processing device of the present invention detects an object based on both the difference in brightness and the difference in chromaticity, and also combines the two resulting images to detect the object. Since the final image of the object is obtained, it is possible to stably extract the detected object without being affected by changes in external lighting, etc., and it is also possible to extract detected objects with the same brightness level but different chromaticity. It is.

Note that the method for identifying the color of a detected object based on the color components of the reflected light reception signal of the detected object is disclosed in Japanese Patent Application Laid-Open No. 56-43.
The one disclosed in Japanese Patent No. 521 is known.

(Embodiment) An image processing apparatus according to a preferred embodiment of the present invention will be described below with reference to the accompanying drawings.

In the following embodiment, as shown in FIGS. 1 and 2, in the gasoline injection process of the automobile assembly line 1, the position of the gasoline injection port 3 of the automobile 2 is detected and the robot
The case where automatic gasoline injection is performed will be explained below. In order to detect the position of the gasoline injection port, a color imaging device (hereinafter simply referred to as a camera) 5 is fixedly provided. This camera 5 includes an image processing device lO
is connected, and this image processing device 10 receives an image signal Is from the camera 5, processes this signal Is to obtain a final image signal LS for determining the position of the gasoline injection port 3, This is what is output.

A robot controller 6 is connected to the image processing device 10, and this robot controller 6 receives the final image signal LS from the image processing device 10 and positions the gasoline injection port 3 based on this signal LS. is determined, and the gasoline injection nozzle 7 attached to the robot 4
is inserted into the gasoline injection port 3 by operating the robot 4, and then a certain amount of gasoline is injected.

Next, the details of the image processing device 10 will be explained with reference to FIG. 3. The image processing device 10 is: connected to the camera 5, decomposes the image signal Is from the camera 5 into each color, Red image signal ISR.

It is provided with an image memory 11 that obtains a green image signal ISO and a blue image signal ISB and temporarily stores these signals ISG and ISB. This image memory 11
has two output terminals, one output terminal is connected to the brightness calculation circuit 12, and the other output terminal is connected to the chromaticity calculation circuit 13. Above brightness calculation circuit 1
2 converts a color three-dimensional element into a brightness data signal D1 of a brightness-dimensional element by a brightness calculation (ICI=R2+G''+82), and this brightness data signal D
1 is stored in the brightness memory 14. The above chromaticity calculation circuit 1
3 is chromaticity calculation (R/R+G+B.

G/R+G+B, B/R+G+B) to convert the color three-dimensional data to the chromaticity parameter (f (R', G'.

B)) This chromaticity parameter signal D2 is stored in the chromaticity memory 15.

The output terminal of the brightness memory 14 is connected to one input terminal of a first comparison circuit 16.
A first set value setting circuit 17 for setting a first set value R1, which is a set value of brightness, is connected to the other input terminal of the display unit 6. The first comparison circuit 16 compares the brightness data signal D1 for each pixel from the brightness memory 14 with the first set value R1, and outputs a signal "1" when Dl is larger than R1, and when DI is R1. When it is smaller, a signal "0" is output. The output terminal of the comparison circuit 16 has the above “0” and “1”.
An image binarization circuit 18 that binarizes an image from a signal is connected, and a binarized image signal BSI obtained by this image binarization circuit 18 is stored in a binarized image memory 19.

FIG. 4 shows a binarized image I-1 based on the binarized image signal BSI. It is assumed that this binarized image I-1 has a defect F-2 as shown in this figure.

On the other hand, the output terminal of the brightness memory 15 is connected to one input terminal of a second comparison circuit 20, and the other input terminal of the second comparison circuit 20 is connected to a second input terminal which is a set value of chromaticity.
A second set value setting circuit 21 for setting the set value R2 is connected. This second setting value R2 is calculated and set each time in the setting circuit 21 based on the chromaticity parameter signal D2 from the memory 15. The second comparison circuit 20 compares the chromaticity parameter signal D2 for each pixel from the brightness memory 15 with the second set value R2, and outputs a signal "1" when D2 is larger than R2, and when D2 is When it is smaller than R2, a signal "0" is output.

The output terminal of the comparison circuit 20 has the above “0” and “1”.
An image binarization circuit 22 that binarizes an image from a signal is connected, and a binarized image signal BS2 obtained by this image binarization circuit 22 is stored in a binarized image memory 23.

FIG. 5 shows a binarized image 1-2 based on the binarized image signal BS2. It is assumed that this binarized image 1-2 has a defect F-2 as shown in this figure.

An image synthesizing circuit 24 is connected to the output terminals of the two binarized image memos IJ 19 and 23, and this image synthesizing circuit 24 synthesizes the two binarized images I-L I-2. As a result, a composite image I-3 of the defect-free gasoline injection port 3 as shown in FIG. 6, in which the respective defects F-1 and F-2 are eliminated, is obtained.

The signal of the composite image I-3 is further inputted to a final image processing circuit 25 for reducing noise, etc., and subjected to noise processing, etc., and then manually inputted to the robot controller 6. This robot controller 6 uses the processed composite image I.
- Determine the position of the gasoline injection port 3 based on the signal 3,
This causes the robot 4 to operate as described above to automatically inject gasoline.

As described above, even if the brightness of the object to be detected and the background change due to changes in illumination, body color, etc., a binarized image can be obtained, and image processing over a wider range is possible.

[Brief explanation of drawings]

FIG. 1 is a schematic plan view of an automatic gasoline injection device incorporating an image processing device according to an embodiment of the present invention, FIG. 2 is a schematic front view of the automatic injection device shown in FIG. 1, and FIG. , a block diagram showing the configuration of an image processing device according to an embodiment of the present invention, FIG. 4 is a diagram showing an example of a binarized image based on brightness, and FIG. 5 is an example of a binarized image based on chromaticity. FIG. 6 is a diagram showing a composite image obtained by combining the two binarized images shown in FIGS. 4 and 5 based on the present invention. 2...Car 3...Gasoline inlet 5...Camera 11...Image memory 12...Brightness calculation circuit 13...・・・
Chromaticity calculation circuit 16.20... Comparison circuit 18... Image binarization circuit (brightness) 22...
...Image binarization circuit (chromaticity) 24...Image synthesis circuit Fig. 2 Fig. 4 Fig. 5 Fig. 6-J

Claims (1)

[Claims] First image processing that compares an image brightness signal from an imaging device that images the body of the vehicle with a first set value and converts it into a first binary image that distinguishes a detected object on the body from a background based on the difference in brightness. means, a second image processing means that compares the image chromaticity signal from the imaging device with a second set value and converts it into a second binary image that distinguishes the detected object from the background based on the chromaticity difference; An image processing device comprising an image synthesis means for synthesizing first and second binary images to obtain a final image.