JPH08105713A - Method and apparatus for processing inspecting image and inspecting apparatus by image processing - Google Patents

Abstract

PURPOSE: To improve the operability and the efficiency of the inspecting operation by automatically setting set information of the same conditions which is necessary to process the image produced by other imaging means merely by calculating the set information by one imaging means by utilizing the calibration data of each imaging means. CONSTITUTION: A window is opened on a screen on which an image produced by a first camera 1A is displayed, calibration data is extracted, and both are calibrated to calculate the position of the window and the deciding conditions in an actual coordinate system. Then, the calibration data of second camera 1B-N-th camera 1N are extracted, the position and the deciding conditions of the window are calculated by the extracted values in an actual coordinate system, and the position and the deciding conditions of the window are calculated in the camera coordinate system of the camera 1B-1N. The window of the imaged by the cameras 1B-1N are automatically opened, and the deciding condition values are automatically set. Thus, the setting of the same inspecting conditions can be easily conducted for a plurality of imaging means.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a camera in which, for example, the same industrial products to be mass-produced are placed and conveyed on a plurality of belt conveyors arranged in parallel, and they are installed in each inspection conveyance line. Imaged by multiple imaging means such as
The present invention relates to an inspection image processing method and its apparatus applied when visually inspecting the same content from those captured images, and an inspection apparatus by image processing.

[0002]

2. Description of the Related Art When inspecting the same product for the same content, it is naturally necessary to set all of a plurality of inspection devices including a camera as an image pickup means to the same condition. However, in reality, due to differences in physical position such as how to apply lighting and installation distance of each camera,
Even if the same product is imaged, the size of the captured image and the position in the image are often different for each camera. Therefore, conventionally, a window is individually drawn for each of captured images captured by a plurality of cameras, and determination conditions are individually set.

For example, a case where an image obtained by picking up an image of a product is used will be described as an example. As shown in FIG. 8, the images A1 and A2 of the picked-up images A1 and A2 are picked up by a plurality of cameras. Window W to fit the size
1 and W2 are drawn respectively, and those windows W1 and W
The ratio (here, the area) of the number of pixels in which 2 is drawn to the total number of pixels is calculated, and the upper and lower limits of the allowable range, which is the inspection determination condition, are set for each camera based on the calculated area. Was set to.

[0004]

According to the conventional image processing method for inspection as described above, the operator draws as many windows as the number of installed cameras for each camera and sets the determination condition. Therefore, there is a problem that not only the operability for a predetermined inspection becomes very complicated, but also the overall efficiency of the inspection work is very low.

The present invention has been made in view of the above situation. By effectively utilizing the calibration data of each image pickup means, the setting information necessary for processing the imaged image by one image pickup means is calculated. An inspection image processing method capable of automatically setting the setting information of the same condition necessary for processing in a captured image by another imaging means and significantly improving the operability and efficiency of inspection work. It is an object of the present invention to provide an apparatus and an inspection apparatus based on image processing.

[0006]

In order to achieve the above-mentioned object, an inspection image processing method according to the invention of claim 1 is such that an image pickup means for picking up a two-dimensional image of an inspection target from the image pickup means. Calibration data is extracted, and setting information necessary for processing the captured image by the other same-type imaging means is calculated and stored, and necessary for processing the captured image by the imaging means by the setting information and the calibration data. It is characterized by calculating various setting information.

An inspection image processing apparatus according to a second aspect of the present invention is an apparatus used for carrying out the inspection image processing method according to the first aspect of the present invention, which is a two-dimensional inspection object. An image pickup means for picking up an image, a means for extracting and storing calibration data of the image pickup means from the image picked up by the image pickup means, and an image picked up by another same kind of image pickup means by calculation with the stored calibration data. It is provided with means for obtaining setting information necessary for processing and storing the setting information, and means for calculating setting information necessary for processing an imaged image by the image pickup means based on the setting information and the calibration data. .

Further, the inspection apparatus based on the image processing according to the invention of claim 3 draws a window on the image picked up by the first image pickup means for picking up a two-dimensional image of the inspection object. Means for storing it, means for setting a determination condition for inspection on the image picked up by the first image pickup means and storing it, and calibration set for the image picked up by the first image pickup means Means for extracting the data and storing it, calculating the window and the calibration data and the judgment conditions and the calibration data stored in each of the storage means, respectively, and calculating the position of the window in the actual coordinate system and the actual position of the window. Means for obtaining a judgment condition value in a coordinate system, calibration data for an image picked up by another same kind of image pickup means, and the first image pickup means The position of the window in the actual coordinate system and the determination condition value in the actual coordinate system obtained from the captured image by And means for automatically drawing windows and automatically setting judgment conditions.

Further, in the image processing inspection apparatus having the above-mentioned structure, it is preferable that each of the image pickup means is arranged in a vertical posture at a position above the inspection transfer line of the inspection object.

[0010]

According to the first and second aspects of the present invention, the calibration data of the image pickup means is extracted from the two-dimensional image of the inspection object imaged by the image pickup means, and the image is picked up by another image pickup means of the same kind. The setting information necessary for processing the image is calculated and stored. By calculating the stored setting information and the calibration, it is possible to obtain the setting information necessary for the processing of the captured image by the image capturing unit, and also the stored setting information is captured by another similar image capturing unit. It is possible to automatically obtain the setting information of the same condition necessary for the processing of the captured image by the other image capturing means by reading out at the time of processing the image and calculating the calibration of this and the other image capturing means, It is possible to save the operation labor for individually calculating the setting information for each imaging unit.

According to the third aspect of the invention, a window is drawn for the two-dimensional image of the inspection object imaged by the first imaging means, and the inspection determination condition for the image is set. Then, it is possible to obtain the position of the window in the actual coordinate system and the determination condition value by calculating the window and the determination condition and the calibration data of the first imaging means.
The same window as that of the first image pickup means for the image picked up by the other image pickup means by the calibration of the image picked up by the other image pickup means of the same kind and the calculation of the position of the window in the actual coordinate system and the judgment condition value. It is possible to perform automatic drawing and automatic setting of determination conditions similar to those of the first imaging means. Therefore, even if the imaged image of the same inspection object differs for each imaging device due to the difference in the physical installation position of the imaging device, the difference in the way of illuminating, etc., the imaged image by the first imaging device. By performing a simple operation of setting window drawing and determination conditions for the same, it is possible to set the same window drawing and determination conditions corresponding to the images captured by other image capturing means of the same type, and to perform the same operation. The contents inspection work can be efficiently performed.

[0012]

Embodiments of the present invention will be described below with reference to the drawings. 1 and 2 are a schematic plan view and a schematic side view showing an outline of an inspection apparatus using image processing according to an embodiment of the present invention, in which FIG. 1 and FIG.
N is a camera as an image pickup means, and these cameras 1A, 1B, ..., 1N are conveyer conveyors 2A, which are arranged in parallel at predetermined intervals so as to form a plurality of inspection lines. 2B, ..., 2N are vertically installed at the upper positions of the respective conveyors 2A,
, 2N are fixed so as to capture a two-dimensional image of the same inspection object T placed and conveyed on 2B, ..., 2N from above, and the cameras 1A, 1B ,. Inspection devices 3A, 3B, ..., 3N are provided,
These visual inspection devices 3A, 3B, ..., 3N are connected to a central processing unit (CPU) 5 via a CPU bus 4.

FIG. 3 shows the visual inspection devices 3A, 3B, ...
It is a schematic block diagram which shows the internal structure of 3N. In the figure, the visual inspection device 3A connected to the first camera 1A
Only the internal structure of the camera is shown in detail, but other cameras 1
Visual inspection devices 3B, ..., 3N connected to B, ..., 1N
The internal configuration of is exactly the same.

In FIG. 3, reference numeral 6 denotes an image memory for storing an image picked up by the first camera 1A, to which a display unit 7 such as a CRT for displaying the picked up image is connected. 8 is a window memory that draws a window (see FIG. 8) on the captured image, and stores the number of pixels in which the window is drawn, and 9 sets a determination condition for inspection on the captured image. A determination condition memory 10 for storing it is a calibration memory 10 for extracting and storing calibration data set for the captured image. 1
Reference numeral 1 denotes an arithmetic control unit that calculates the window position in the actual coordinate system by operating the windows and calibration data stored in the memories 8 and 10 and stores them in the memories 9 and 10. The determination condition value in the actual coordinate system is calculated by calculating the determined determination condition and the calibration data. The position of the window in the actual coordinate system and the determination condition value obtained by the arithmetic control unit 11 in the visual inspection device 3A corresponding to the first camera 1A having such a configuration are the CPU bus 4 and C.
, 3N corresponding to the other cameras 1B, ..., 1N are transmitted and copied through the PU5.

Next, the image processing operation of the inspection apparatus having the above-mentioned structure will be described with reference to the flow charts shown in FIGS. In order to facilitate understanding of the operation flow, the window drawing operation (FIG. 4) and the determination condition setting operation (FIG. 5) will be described separately, but in reality, both these operations are executed simultaneously. Of course there is.

First, the window drawing processing operation will be described. A window is drawn for an image picked up by the first camera 1A, and the number of pixels (bit number) in which the window is drawn is stored in the window memory 8. At the same time (step S21), the calibration data set for the captured image is extracted and stored in the calibration memory 10 (step S22). Next, the window and the calibration data are calculated based on the following arithmetic expression (1) (step S23) to obtain the actual position of each pixel on which the window is drawn in the actual coordinate system (step S23). S24).

[0017]

[Equation 1]

Next, the calibration data set for the image taken by the second camera 1B is extracted and stored in the calibration memory 10 in the visual inspection device 3B connected to the second camera 1B. (Step S25) Then, the calibration data and the position in the actual coordinate system obtained in step S24 are calculated based on the following calculation formula (2) (step S26), whereby the second camera 1B is operated. The position of each pixel in which the window is drawn in camera coordinates is obtained (step S27). Then, the window is automatically drawn at the position of each pixel obtained by the above equation (2) for the image captured by the second camera 1B (step S28). The operation from step S25 to step S28 is repeated until the window automatic drawing process for the last captured image of the camera 1N is completed (step S2).
9) By doing so, it is possible to draw a window corresponding to each size for the images picked up by all the cameras 1A, 1B, ..., 1N.

[0019]

[Equation 2]

The terms used in the above equations (1) and (2) are defined as follows. (Xr, Yr): Value in actual coordinate system (Xc, Yc): Value in camera coordinate system (Ox, Oy): Value in actual coordinate system of camera coordinate origin P: Actual size per pixel θ: Tilt between camera coordinate system and actual coordinate system

Next, the determination condition setting processing operation will be described. The determination conditions for inspection (rotation direction determination conditions and area determination conditions) are set for the image captured by the first camera 1A. Is stored in the determination condition memory 9 (step S31), and the calibration data set for the captured image is extracted and stored in the calibration memory 10 (step S32). Next, the value of the determination condition in the actual coordinate system is obtained by calculating the determination condition and the calibration data based on the following arithmetic expression (3) (step S33) (step S34).

[0022]

(Equation 3)

Subsequently, the calibration data set for the image taken by the second camera 1B is extracted and stored in the calibration memory 10 in the visual inspection device 3B connected to the second camera 1B. In step S35, the calibration data and the judgment condition value in the actual coordinate system obtained in step S24 are calculated based on the following calculation formula (4) (step S36), whereby the second camera is obtained. A determination condition value in the 1B camera coordinate system is obtained (step S37).
Then, the determination condition value obtained by the above equation (4) is automatically set for the image captured by the second camera 1B (step S38). This step S35 to step S
The operations up to 38 are repeated until the determination condition automatic setting process for the last captured image of the camera 1N is completed (step S39), so that the first camera with respect to the captured images of the second and subsequent cameras 1B, ..., 1N. The same judgment conditions as in 1A can be automatically set.

[0024]

[Equation 4]

The terms used in the above equations (3) and (4) are defined as follows. (Xr, Yr): Values in the X and Y directions in the actual coordinate system (Xc, Yc): Values in the X and Y directions in the camera coordinate system (Ox, Oy): In the actual coordinate system of the camera coordinate origin Values in the X and Y directions of θ: Inclination between the camera coordinate system and the actual coordinate system θr: Inclination in the actual coordinate system θc: Inclination in the camera coordinate system Mr: Area value in the actual coordinate system Mc: Camera Area value in coordinate system P: Actual size per pixel

In the above embodiment, a plurality of cameras 1 are used.
, 3N are separately connected to A, 1B, ..., 1N, respectively. However, as shown in FIG. 6, a plurality of cameras 1A, 1B ,.
.., 1N may be connected to only one visual inspection device 3. In this case, the window memory 8,
A plurality of memories (8a to 8n, 9a to 9n, 10a to 1n in the drawing) correspond to the number of cameras 1A, 1B, ..., 1N.
0n) is required, but the image memory 6 and the display unit 7 are required.
Also, the arithmetic control unit 11 can be shared, and the configuration can be simplified and the cost can be reduced.

In the above-described embodiment, the cameras 1A, 2A, 2B, ..., 2N for the inspection lines, which are arranged in parallel with each other at a predetermined interval, are located at the upper positions of the conveyors 2A, 2B, ... 1B, ..., 1N are installed in the parallel type inspection apparatus, the explanation is given with reference to FIG.
As shown in, a plurality of 1A, 1B, ..., At predetermined intervals in the transport direction of the transport conveyor 2 for one inspection line.
You may apply to the serial type | mold inspection apparatus which installs 1N.

[0028]

As described above, claim 1 and claim 2 are as follows.
According to the invention, the calibration data of the image pickup means and the setting information necessary for processing the imaged image by another image pickup means of the same type are obtained from the two-dimensional image of the inspection object imaged by the image pickup means, and the setting information is stored. Therefore, it is possible not only to obtain the setting information necessary for the processing of the captured image by the image capturing means by the calculation of the setting information and the calibration, but also at the time of processing the captured image by another similar image capturing apparatus. It is possible to automatically obtain the setting information of the same condition necessary for the processing of the captured image by the other image capturing means by the calibration of the other image capturing means of the same type and the calculation of the stored setting information. Therefore, when the same inspection condition is set for each image in the image processing in which the images captured by the plurality of image capturing units that capture the same inspection object are different, a predetermined image is captured for each image captured by one image capturing unit. Just by setting, individual setting operation for the captured image by the same type of imaging means other than that is not necessary at all, and the setting work of the same inspection condition for a plurality of imaging devices can be performed very efficiently and efficiently. There is an effect that can be done.

According to the third aspect of the invention, a window is drawn on the two-dimensional image of the inspection object imaged by the first imaging means, and the inspection determination condition for the image is set. Then, the position of the window in the actual coordinate system and the value of the determination condition are obtained by calculating these windows and the determination conditions and the calibration data of the first imaging means, and the position of the window in the obtained actual coordinate system and By the calculation of the determination condition value and the calibration of the imaged image by another imager of the same type, for the imaged image by the other imager, automatic drawing of a window similar to that of the first imager and determination similar to that of the first imager Conditions can be set automatically. Therefore, even if the imaged image of the same inspection object differs for each imaging device due to the difference in the physical installation position of the imaging device, the difference in the way of illuminating, etc., the imaged image by the first imaging device. By performing a simple operation of setting window drawing and determination conditions for the same, it is possible to set the same window drawing and determination conditions corresponding to the images captured by other image capturing means of the same type, and to perform the same operation. This has the effect that the content inspection work can be executed very efficiently.

[Brief description of drawings]

FIG. 1 is a schematic plan view showing an outline of an inspection apparatus by image processing according to an embodiment of the present invention.

FIG. 2 is a schematic side view of FIG.

FIG. 3 is a schematic block diagram showing an internal configuration of a visual inspection device.

FIG. 4 is a flowchart showing a window drawing operation in the image processing operation.

FIG. 5 is a flowchart showing a determination condition setting operation of the image processing operations.

FIG. 6 is a schematic block diagram showing an internal configuration of a visual inspection device in an inspection device by image processing according to another embodiment of the present invention.

FIG. 7 is a schematic side view showing another application example.

FIG. 8 is a schematic diagram illustrating a conventional inspection image processing method.

[Explanation of symbols]

 1A, 1B, ..., 1N Camera (imaging means) 5 CPU 8 Window memory 9 Judgment condition memory 10 Calibration memory 11 Arithmetic control unit T Inspection object

─────────────────────────────────────────────────── ───

[Procedure amendment]

[Submission date] May 1, 1995

[Procedure Amendment 1]

[Document name to be amended] Statement

[Name of item to be corrected] 0022

[Correction method] Change

[Correction content]

[0022]

(Equation 3)

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI Technical display location G06F 15/64 400 A

Claims (4)

[Claims] 1. The calibration data of the image pickup means is extracted from the image picked up by the image pickup means for picking up a two-dimensional image of the inspection object, and the setting information necessary for the processing of the picked up image by the other similar image pickup means is calculated. Then, the inspection image processing method is characterized by storing and storing the setting information necessary for the processing of the imaged image by the imaging means by the setting information and the calibration data. 2. An image pickup unit for picking up a two-dimensional image of an inspection object, a unit for extracting and storing calibration data of the image pickup unit from an image picked up by the image pickup unit, and the stored calibration data. Means for obtaining setting information necessary for processing of a picked-up image by another same-type image pickup means and storing it, and setting information necessary for processing the picked-up image by the image pickup means by the setting information and the calibration data. And an image processing device for inspection. 3. A first for capturing a two-dimensional image of an inspection object
An image pickup means, a means for drawing a window for an image picked up by the first image pickup means and storing the window, and a judgment condition for an inspection set for the image picked up by the first image pickup means and storing it. Means, means for extracting the calibration data set for the image picked up by the first image pickup means and storing the same, windows stored in each of the storage means, calibration data, determination conditions, and calibration Means for calculating the position of the window in the actual coordinate system and the judgment condition value in the actual coordinate system by respectively calculating the data and the calibration data for the image captured by the other similar image capturing means and the first image capturing means. Calculate the window position in the actual coordinate system and the judgment condition value in the actual coordinate system obtained from the captured image That Te and means for automatically setting the automatic drawing and judgment conditions similar window comprising a first image pickup means with respect to an image captured by other similar imaging means inspecting apparatus by image processing according to claim. 4. The image processing inspection apparatus according to claim 3, wherein each of the image pickup means is arranged in a vertical posture at a position above the inspection object conveyance line.