JPH0721825B2 - Image capture device - Google Patents

Description

TECHNICAL FIELD The present invention relates to an image capturing device, and can be used, for example, for determining whether a target object is correct or defective by image processing.

(Prior Art) An image capturing device is used to input an image signal of a target object to an image processing device or the like. When the target object moves relative to the image capturing device at a constant speed, a line scanner is used in the image capturing device to capture image data of the relative moving object at a predetermined timing. At this time, if the relative speed between the image capturing device and the object is different or fluctuates from the predetermined speed, the image of the object is captured as a distorted image. The image capturing device is driven in synchronization with the conveyance.

8 and 9 show an example of an image capturing device using a conventional line scanner. In FIGS. 8 and 9, a circular target object 21 such as a can lid is a front and rear pulley 3
It is carried on a belt 22 that is stretched between 1 and 32. An illumination light source 23 for illuminating the object 21 is arranged above the transportation path of the object 21, and a lens 24 and a line scanner 25 are arranged in this order on the optical path reflected from the object 21. The lens 24 forms an image of the object 21 on the line scanner 25. The line scanner 25 is, for example, a line sensor using a CCD or the like, and is driven by the line scanner drive circuit 26 at a constant timing, whereby the image data of the object 21 is captured at a constant timing. The image data of the object 21 captured by the line scanner 25 is converted into a digital signal by the analog / digital converter 27 via the line scanner drive circuit 26 and stored in the memory 28. Memory 28
The operation of the micro processing unit (MPU) 29
Further, the MPU 29 performs a calculation based on the stored image data of the object to determine whether the object 21 is good or bad.

Now, as shown in FIGS. 10 to 13, the transport direction of the object is Y, the direction orthogonal thereto is X, and the radius of the object is R. Assuming that the conveyance speed of the object 21 by the belt 22 is a predetermined speed, the image of the object captured and stored by the image capturing device has the same image as the shape of the object 21 as shown in FIG. However, if the speed of the belt 22 is faster than the predetermined speed, the image of the object becomes elliptical with a long axis in the conveying direction and a short radius R as shown in FIG. 11, and the speed of the belt 22 is slower than the predetermined speed. Then, as shown in FIG. 12, the image of the object becomes an ellipse whose long axis is long in the direction orthogonal to the carrying direction and whose major axis is R. Also, if the transport speed fluctuates midway, the image of the object is distorted into an egg shape or the like as shown in FIG.

In any case, if the conveyance speed of the object by the belt 22 is too fast, slow, or fluctuates, the image obtained by the image capturing device becomes distorted, and correct image processing cannot be performed. In the capture device,
The image capturing timing is synchronized with the conveyance of the object 21 by the belt 22. Specifically, a rotary encoder 33 is attached to the pulley 31 of the belt 22 shown in FIG. 8, and the line scanner drive circuit 26 is operated in synchronization with the signal from the encoder 33 so that images are captured line by line. There is.

(Problems to be Solved by the Invention) According to the above-mentioned conventional image capturing device, in order to capture image data with high accuracy, an expensive encoder with high resolution is required, and the encoder provided in the transport system and the image capturing The configuration was troublesome because it needed to be connected to the device.

The present invention has been made in order to solve the problems of the prior art, and provides highly accurate image data without providing an encoder on the transport device side and independently of the transport system and the image capturing device side. It is an object of the present invention to provide an image capturing device capable of obtaining the above.

(Means for Solving the Problem) According to the present invention, a line scanner that captures image data of an object at a predetermined timing while relatively moving an object of a fixed shape, and image data at a predetermined transport speed of the object as a reference image. Memory means for storing and storing image data captured by a line scanner, comparing the stored image data with the reference image, and regarding the transport direction between the captured image data and the reference image due to a difference in transport speed. And a recognition unit that confirms the difference between the data positions of.

(Function) Even if the relative speed between the object and the line scanner is too fast or slow and the captured image data is distorted,
The difference between the reference image and the reference image can be recognized, and the image data can be corrected based on the recognized difference.

(Embodiment) An embodiment of the image capturing apparatus according to the present invention will be described below with reference to FIGS. 1 to 7.

The embodiment shown in FIG. 1 is an example of a target object 1 having a fixed shape, for example, the upper surface of a can or a lid of a bottle, which is used to inspect the target object 1 for stains, scratches, print misalignment, and the like. . The belt 2, which conveys the object 1, the illumination light source 3, the lens 4, and the line scanner 5 have the same positional relationship as in the above-described conventional example. The illumination light source 3 can be configured by arranging LEDs linearly, for example. The line scanner 5 is composed of a one-dimensional CCD, for example, and can capture an image of one line in the width direction of the belt 2. The line scanner 5 is driven by the line scanner drive circuit 6 at a predetermined timing, and takes in the image data of the object 1 relatively moving on the belt 2 at a predetermined timing. The captured image data is converted into a multi-valued digital signal by the analog / digital converter 7 and stored in the input image memory 8. The image memory 8 stores multi-valued image data, but it may store binarized image data. The data reading / reading operation of the input image memory 8 is controlled by the MPU 9. The MPU 9 controls the shape information memory 10, the memory 11, and the correction data memory 12, respectively. The shape information memory 10 stores image data of various objects at a predetermined transport speed as a reference image. MPU9
Selects a reference image of the object 1 to be actually handled from the reference images and stores it in the memory 11, and compares the selected reference image with the data of the input image stored in the input image memory 8. The MPU 9 constitutes a recognition unit that calculates and recognizes a difference in data position between the input image data and the reference image in the conveyance direction due to a relative difference in conveyance speed between the object 1 and the line scanner 5. The difference between the determined data positions is stored in the correction memory 12 as correction data. Image processing may be performed using this correction data, or an input image may be corrected using the correction data and used for image processing. The memory means in the claims is,
It is composed of an input image memory 8 and a shape information memory 10.

Next, the outline of the correction data creating operation according to the above-described embodiment will be described. If the outer shape of the target object 1 is circular, the correction data is created by focusing on one side of the center line along the carrying direction. The shape information memory 10 stores, as a reference image, data of the outer shape of an image input at an accurate transport speed. The reference image may be input while the object 1 is actually conveyed at a predetermined speed, or the value of Y with respect to the value of X may be stored by calculation, for example, by the equation of a circle. Now, as shown in FIG. 2, the reference image stored in the shape information memory 10 is A, and the input image data when the object 1 is actually conveyed is B. Therefore, the X value of the outer shape data of the object 1 is checked from the image data B, and whether X tends to increase or decrease with respect to the reference image A is checked, and the correct Y value for the X value is determined. The difference between the calculated value and the actual value of the object 1 is stored in the correction data memory 12 as correction data.

Next, the correction data creation will be described more specifically with reference to FIGS. 3 to 7.

When the switch is turned on and started in step 1 shown in FIG. 3, the belt conveyor 2 operates in a predetermined conveying direction, the illumination light source 3 is turned on, and the belt conveyor 2 is illuminated.

This signal is input to the MPU9 when the switch is turned on in step 2, and the MPU9 receives this signal.
Outputs a signal for driving the line scanner drive circuit 6. Then, the line scanner 5 takes in the image data of one line. The captured image data is analog
The digital converter 7 converts the digital data into 8-bit 256-gradation digital data, for example.

The converted image data is stored in the column of Y = 1 of the input image memory 8 in step 3. The image data for each cell of one line of the line scanner 5 is stored for each pixel of the input image data as shown in FIG.

Next, in step 4, it is determined whether or not the target object 1 has arrived on the image input surface of the line scanner 5. Here, the object 1 is detected by the MPU 9 using the image data of step 3.
Specifically, as shown in FIG. 7, it is determined whether or not a value greater than or equal to a threshold value Th is stored in the image data, and if it is determined that a value greater than or equal to the threshold value is stored, the object 1 Since it has been detected, the process proceeds to the next step 5. on the other hand,
If all the image data on one line is smaller than the threshold value, it is determined that the object 1 is not within the area where the image is input by the line scanner 5, and the process returns to step 2. Then, the newly input 1-line image data is processed in step 3
In the column of Y = 1. That is, the image data already stored in the column of Y = 1 disappears and is replaced by the newly input data. When it is judged that the object 1 has come in this way, as shown in FIG.
The leading edge of the image data of the actual object 1 is always stored in the column 1.

In step 5, when the object 1 is detected in step 4, the image data of the next one line is input as in step 1 and converted into a digital signal.

In step 6, the next 1-line image data in step 5 is stored in the Y = 2 column of the input image memory 8 as shown in FIG. Steps 5 and 6 are repeatedly executed, and the input image data for each line is sequentially stored in the image input memory 8 from the column of Y = 1 to the column of Y = N.

In step 7, when the row of the written image data reaches the last row, that is, the row of Y = N, the process proceeds to step 8 as a memory end. If the written image data row is not the last row, the procedure returns to step 5,
Input the image data of the next one line and input image memory 8
Are sequentially stored in the column of Y next to. The final row in the Y direction has a value sufficiently larger than the size of the object 1 in the Y direction.

In step 8, the image data input by the operations of steps 5 to 7 is processed by the MPU 9, and the value of X at the position of the left edge of the object 1 in the figure is detected. Here, first, in order to grasp the correspondence between the X coordinate of the tip of the input image data and the X coordinate of the tip of the reference image in the shape information memory 10, the X coordinate at Y ₁ , that is, the X coordinate of the tip is detected. . Hereinafter, the value of the X coordinate of the left edge when Y ₂ , Y ₃ , ... Is detected. The method of detection is the same as the operation performed in step 4, and as shown in FIG. 6, from the image data in the column Y = H, the X at the position where the value equal to or greater than the predetermined threshold value first appears. Find the value of X ₁ . As a result, the coordinates (X ₁ , H) are obtained, and these coordinates are written in the memory as the position of the left edge of the object 1. The right edge of the object 1 can be similarly obtained.

In step 9, the X value X ₁ obtained in step 8 is compared with the X value X ₀ of the immediately preceding column to determine whether X ₁ is in the upper half or the lower half of the object 1. That is, as shown in FIG. 4, the value of Y corresponding to X ₁ is 2 in the upper half and the lower half of the object 1.
This X ₁ is the shape information memory 10 to be described later.
Which Y value in is to be determined. If X ₁ > X _0, the value of X is increasing, so it is determined that this X ₁ is located in the upper half of the object 1. If X ₁ <X _0, the value of X is decreasing, so it is determined that this X ₁ is located in the lower half of the object 1.

In step 10, the value Y ₁ in the Y ′ direction in the shape information memory 10 corresponding to the value X ₁ of X obtained in step 8 is read.
Now, for example, assuming that the image data shown in FIG. 10 is the data of the reference image, in this reference image data, when Y ₄ , X ₃
Has become. However, if the input image data is the data shown in FIG. 11, the input image data is
It becomes X ₃ when Y ₆ .

It is assumed that the X coordinate has a correspondence relationship between the captured image data and the coordinate of the reference image in the shape information memory 10, and the X coordinate at the beginning is the same.

The following two methods are conceivable as a method of storing the reference image data in the shape information memory 10.

One is a method of storing accurate image data of an object as it is at an accurate transport speed.

The other is a method of obtaining the value of Y using the equation (XA) ² + (YB) ² = r ² of a circle when the shape of the target object is circular. r is the radius of the object, and the maximum value of the number of pixels in the X direction that exceeds the threshold value in the input image memory 8 in step 6 is the diameter of the object, and half the diameter is the radius r. Further, (A, B) is the center coordinate, and the value of A is half the number of pixels in the input image memory 8 in the X direction. The value of B is B
= R. Formula to obtain Y from the above formula To here, Is the lower half, Is the upper half.

In step 11, the Y-direction value of the input image data with respect to the X value X ₁ obtained in step 8 and Y ′ of the shape information memory 10 are set.
The difference from the direction value is stored in the correction data memory 12. As the correction data, the difference between the two data may be used as described above, or the respective values may be stored as X ₁ ··· (Y ′, Y).

In step 12, the operations of steps 8 to 11 are repeated until all the Y'directions in the shape information memory 10 are completed, and the correction data is stored in the memory 12 for each.

In step 13, the signal obtained after step 12 is the MPU9
, And the operation of this flow is ended by this signal.

In another flow, the bottle lid is inspected for stains, scratches, print misalignment, etc. based on the stored correction data. In this case, the data in step 11 is converted as follows and the inspection is performed.

First, the case of correcting the input image of the entire object will be described. The correction data of Step 11 or X ₁ ··· (Y ′,
In the Y) data, the Y'th one line of data including X ₁ is transferred to the Y'th column of the input image data,
This is done for all columns of Y'to correct the input image data. Here, when the values in the X direction are the same, the data may be rewritten on the previous data for processing, or may be averaged with the previous data. In the case where the shape information memory has a shape in which the same value continues for the left edge in the X direction, the value in the Y column corresponding to the X value cannot be determined as a single value. Blank data is left on the memory, the next line is processed, the entire line is processed, and then the blank part is processed by interpolation. By increasing the resolution of the line scanner, it is possible to prevent the above-mentioned values in the X direction from becoming the same.

Next, in the case of modifying only the part necessary for the inspection, for example, the window may be set only for the part necessary for the inspection, and the area of this window may be modified. Also,
When correcting the pattern or pattern of the target object,
You may make it correct the data in the set window.

The correction data memory 12 in the embodiment of FIG. 1 may store the correction data of the entire target image as described above, or may correct the window set for image processing. Good.

The target object to which the image capturing apparatus according to the present invention can be applied may have a predetermined shape, and may be a circular object such as a bottle lid or a diamond object. Further, in the case of a quadrangular object, when the object and the line scanner relatively move, the ridge line of the object diagonally crosses the line scanner.

(Effect of the Invention) According to the present invention, the image data stored in the memory means is compared with the reference image, and the difference in the data position between the captured image data and the reference image in the conveyance direction due to the difference in the conveyance speed is calculated. Since the recognition unit for recognizing is provided, the input image can be corrected by the difference in the data position. Therefore,
Since it is not necessary to synchronize the conveyance speed and the image capture by using a speed sensor such as an encoder, the device can be made compact and the cost can be reduced. Also,
Since the transport system and the image capturing section are independent of each other, it is effective when the two are located apart from each other or when physical connection is difficult.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of an image capturing apparatus according to the present invention together with a mechanical portion, FIG. 2 is a diagram showing the principle of correction data creation according to the above embodiment, and FIG. 3 shows the operation of the above embodiment. FIG. 4 is a flow chart, and FIG. 4 is a conceptual diagram showing how the image data is loaded into the input image memory in the above embodiment, and FIG. 5 is a conceptual diagram showing how data of the tip of the object is loaded into the input image memory. , FIG. 6 is a conceptual diagram showing how the data of the edge of the object is captured, FIG. 7 is a diagram showing the principle of edge detection of the object, and FIG. 8 is a block diagram showing an example of a conventional image capturing device together with the mechanical portion. ,
FIG. 9 is a side view of the same mechanical part, FIG. 10 is a diagram showing an example of input image data at a predetermined conveying speed, and FIG. 11 is a diagram showing an example of input image data at a high conveying speed. FIG. 12 is a diagram showing an example of input image data when the transport speed is slow, and FIG. 13 is a diagram showing an example of input image data when the transport speed changes. 1 ... object, 5 ... line scanner, 8, 9 ... memory means.

Claims (1)

[Claims] Claim: What is claimed is: 1. While moving an object having a fixed shape and a line scanner relative to each other, capturing image data of the object,
In the image capturing device for storing the image data in the memory means, the line scanner captures the image data of the relatively moving object at a predetermined timing, and the memory means operates at a predetermined transport speed of the object. The image data is stored as a reference image and the image data captured by the line scanner is stored. The image data stored in the memory means is compared with the reference image, and the image data is captured by a difference in transport speed. An image capturing device having a recognition unit that recognizes a difference in data position between the image data and the reference image in the transport direction.