JPH06148083A - Method and device for detecting contamination of lens used for detecting defect of paint film - Google Patents

Abstract

PURPOSE:To prevent the occurrence of erroneous detection caused by the lens contamination of a camera. CONSTITUTION:A CCD camera 6 successively fetches the picture of the coated surface 1a of a car body 1 and a picture processor 9 processes the pictures. A lens contamination discriminating means 10B receives the output of a storing means 10A and compares the defective position of the picture fetched in the previous time with that of the picture fetched this time. When defects are found at the same position, the means 10B discriminates that the defects are the lens contamination of the camera 6.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a lens stain detecting method and apparatus for detecting coating film defects.

[0002]

2. Description of the Related Art Conventionally, a laser slit light is projected onto a surface to be inspected, and the reflected light is reflected on the surface to be inspected by image processing for inspecting the surface defect of the object to be inspected such as a painted surface of an automobile body panel. It is known to project on a screen and detect the surface defect of the surface to be inspected from the slit image (see, for example, JP-A-62-233710).

By the way, the detector is equipped with a screen made of a diffuser plate such as frosted glass at its tip and two flat mirrors on its inner surface, a condenser lens and a line sensor behind it. It consists of the camera unit.

[0004]

However, in that case,
Since the camera unit is used to detect the surface defect of the object to be inspected, if the lens has dirt such as rubber or paint, there is a problem that the dirt is erroneously detected as a defect.

The present invention provides a lens stain detecting method and apparatus for detecting coating film defects which can prevent erroneous detection due to lens stains on a camera.

[0006]

According to a first aspect of the present invention, there is provided a method for detecting dirt on a lens of a camera when a defect on a coating surface of an inspection object is detected by image processing. The images on the surface are sequentially captured, the previously captured image is compared with the image captured this time, and if there is a defect at the same position on both images, this is determined to be lens stain.

According to a second aspect of the present invention, there is provided a device for detecting dirt on a lens of a camera when a defect on a coated surface of an inspection object is detected by image processing. Image capturing means for capturing, and storage means for receiving the output of the image capturing means and storing the defect position of the captured image,
Upon receiving the output of the storage means, the image captured last time is compared with the image captured this time, and when there is a defect at the same position, a lens stain determination means for determining this as a lens stain is provided. The configuration is provided.

[0008]

According to the invention of claim 1, the images of the painted surface of the object to be inspected are sequentially captured, the previously captured image and the image captured this time are compared, and defects are found at the same position in both images. In some cases, it is determined that this is lens stain.

According to the second aspect of the present invention, when the defect position of the image previously captured and currently stored and the defect position of the image captured this time are at the same position, the lens stain determination means determines Is determined to be lens dirt.

[0010]

Embodiments of the present invention will now be described in detail with reference to the drawings.

In FIG. 1 showing a coating inspection station, reference numeral 1 is a vehicle body which is an object to be inspected, and a surface state inspection for inspecting a coating surface 1a of the vehicle body 1 in the vicinity thereof to detect the presence or absence of coating film defects. The device 2 is arranged.

In this surface state inspection device 2, a robot means 4 is placed on a pedestal 3, and a light source 5 and a CCD camera 6 are provided at the tip of a robot arm 4a of the robot means 4.
Are attached via the support fitting 7. The surface of the vehicle body 1 that has been brought into the coating station S, that is, the vehicle body 1
The light source 5 and the CCD camera 6 trace the coating surface 1a of the above, and at this time, the light emitted by the light source 5 is reflected by the coating surface 1a of the vehicle body 1 and received by the CCD camera 6. Has become.

In the coating defect inspection by the light source 5 and the CCD camera 6, the host computer 8
Drive the robot device 4 based on a predetermined program,
The light source 5 and the CCD camera 6 are moved so as to trace the coating film surface 1a of the vehicle body 1.

The received light image obtained by the CCD camera 6 is sent to the image processor 9. The image processor 9 amplifies the video signal from the CCD camera 6 and then performs image processing by identifying the level difference in the brightness of the received light image indicated by the video signal, and the brightness data is stored in a computer (not shown). It is configured to be transmitted and analyzed to detect the presence or absence of a coating defect on the coating surface 1a of the vehicle body 1, the coordinates of the defective portion, the shape of the coating defect, and the size thereof.

On the other hand, the light / dark data image-processed by the image processor 9 is also sent to the controller 10,
The defective position of the image is stored in the storage unit 10A. Then, the defect position of the previously captured image and the defect position of the image captured this time, which are stored in the storage means 10A, are compared, and if there is a defect at the same position, this is CC
If the lens of the D camera 6 is dirty, the lens dirt determining means 1
0B makes a decision.

The determination of the lens dirt is performed as shown in FIG.

That is, when starting, first, a light-receiving image is captured by the CCD camera 6 (step S1) and sent to the image processor 9 to detect an object corresponding to dirt as position data (step S2). ). Then, it is determined whether or not the uptake process is completed, that is, whether or not the uptake of one automobile is completed (step S
3). If the capturing step is not completed, the object is detected until it is completed, and if it is completed, the logical product of the position data of the object is calculated (step S4).

Then, it is judged whether or not there is the same position data (step S5). If there is not the same position data, the lens is not contaminated. Therefore, the process is ended as it is, and if there is the same position data, an alarm alarm is output, The area of the same position data is masked (step S6), the detection of this area is prohibited, and the process ends.

In this way, dust, paint, etc. will
False detection due to being attached to the lens of the D camera 6 is prevented. For example, when the captured images U11, U12, ... Have dirt P11, P12, ... As shown in FIG. 3, the logical product of the position data of the object is as shown in FIG. Dirt P1 remains. Therefore, as shown in FIG. 5, the area is set as the masking area V1.

According to the above processing, there is a time margin,
This is effective when the storage capacity is small, but conversely, when the storage capacity is large but there is no time margin, processing can be performed as shown in FIG. 6 below.

That is, when starting, first, a light-receiving image is taken in by the CCD camera (step S11), which is sent to the image processor to binarize an object corresponding to dirt (step S12). The host computer stores it as a binarized image (step S13) and detects an object corresponding to dirt as a binarized image (step S1).
Four).

Then, it is judged whether or not the taking-in process is completed, that is, whether or not the taking-in process is completed for one automobile (step S15). If the capturing step is not completed, the object is detected until it is completed, and if it is completed, the logical product of the stored binarized images is calculated (step S16).

Then, it is judged whether or not there is a residual image (step S17), and if there is no residual image, there is no lens stain, so the processing is terminated. If there is a residual image, an alarm alarm is output and the residual image is detected. The area is masked (step S18), detection of the area is prohibited, and the process ends.

For example, as shown in FIG. 7, each image U21, U22, ...
If there are P22, ..., The logical product of the stored binary images will have stains P2 as shown in FIG. Therefore, as shown in FIG. 9, the area is set as the masking area V2.

[0025]

As described above, according to the first aspect of the present invention, when there is a defect at the same position as the defect position of the previously captured image, it is determined that this is a lens stain. It is possible to prevent erroneous detection due to lens stains.

According to the second aspect of the present invention, when there is a defect at the same position as the defect position of the image previously captured and currently stored, the lens stain determination means determines that this is lens stain. Lens dirt can be easily detected.

[Brief description of drawings]

FIG. 1 is a perspective view of a surface state inspection device.

FIG. 2 is a flowchart of a lens dirt determination.

FIG. 3 is an explanatory diagram of a captured image.

FIG. 4 is an explanatory diagram of a logical product of position data of objects.

FIG. 5 is an explanatory diagram of masking processing.

FIG. 6 is a flow chart similar to FIG. 2 for another embodiment.

FIG. 7 is a view similar to FIG. 3 for another embodiment.

FIG. 8 is a view similar to FIG. 4 for another embodiment.

FIG. 9 is a view similar to FIG. 5 for another embodiment.

[Explanation of symbols]

 6 CCD camera 8 Host computer 8A Storage means 8B Lens dirt determination means 10 Image processor

Claims (2)

[Claims] 1. A method for detecting dirt on a lens of a camera when detecting a defect on a painted surface of an object to be inspected by image processing. Compare the image captured with the image captured this time,
A lens stain detection method in coating film defect detection, characterized in that when there is a defect at the same position in both images, it is determined as lens stain. 2. A device for detecting dirt on a lens of a camera when detecting a defect on a coating surface of an object to be inspected by image processing, comprising image capturing means for sequentially capturing images of the surface to be inspected of the object to be inspected. , A storage unit that receives the output of the image capturing unit and stores the defect position of the captured image, and a storage unit that receives the output of the storage unit and compares the previously captured image with the image captured this time. A lens stain detecting method for detecting a coating film defect, comprising: a lens stain determining unit that determines a defect at the same position as a lens stain.