KR100205172B1 - Apparatus for monitoring the number of sheets transfered - Google Patents

Abstract

The present invention relates to a sheet conveying quantity monitoring apparatus for monitoring the conveying quantity of the sheet, and in detail, the laser beam transmitter and the monitoring camera are inclined at the upper and lower portions of the sheet slightly spaced apart from the side of the conveying sheet to provide high brightness to the side of the sheet. The reflected light spot is formed, and the surveillance camera acquires the reference thickness of the sheet by computer learning through several times of self-learning, and stores it in the computer memory. In comparison, the alarm signal and the control signal are output when the reference thickness is out of range.

Description

Sheet conveying quantity monitoring device

1 is a block diagram of the present invention.

2 is an image of a state in which the laser beam and the filter is not used on the side of the plate in the present invention.

3 is an image of a state in which the laser beam is irradiated on the side of the plate in the present invention.

4 is a video screen in which noise is removed by using a laser filter in FIG.

5 is a binary image screened in the present invention.

6 is a video screen when the plate material is overlapped in two sheets in the fourth invention of the present invention.

7 is a video screen in the case where the plate is overlapped in two sheets in the fifth invention of the present invention.

8 is a graph obtained by scanning the reflected light point of a sheet of the sheet in the present invention.

9 is a graph obtained by scanning the reflection light points of two sheets in the present invention.

10 is a self-learning flow chart of the present invention.

11 is a monitoring flowchart of the present invention.

* Explanation of symbols for main parts of the drawings

2: surveillance camera 4: plate

6: laser beam 8: side of plate

10: reflected light point 12: laser filter

14 laser beam projector 16 bottom surface of the plate

18 computer 20 memory (memory means)

22: controller 24: processing equipment

26: transport robot 28: background image

30: horizontal scanning line 32: zoom lens

θ1, θ2: slope

The present invention relates to a sheet conveying quantity monitoring apparatus for monitoring the conveying quantity of the sheet, and in detail, the laser beam transmitter and the monitoring camera are inclined at the upper and lower portions of the sheet slightly spaced apart from the side of the conveying sheet to provide high brightness to the side of the sheet. The reflected light spot is formed, and the surveillance camera acquires the reference thickness of the sheet by computer learning through several times of self-learning, and stores it in the computer memory. In comparison, the alarm signal and the control signal are output when the reference thickness is out of range.

Generally, the plate material (metal plate, synthetic resin plate, plate glass) that is conveyed by processing equipment (press, cutter) with the help of the transfer robot is transferred to two or more sheets by oil, moisture, or static electricity. There is a problem that the processing apparatus is broken or the plate can not be processed.

Therefore, in the related art, a plurality of optical sensors are installed in a sheet conveying section to monitor the number of sheets of sheet conveying, and when two or more sheets are conveyed, there is a monitoring device that generates an alarm by blocking the sensor light. Since the number of sheets is monitored, it may malfunction even in the case of weak vibration or noise, so the detection reliability is low and the response speed is very slow (about 1/5 second), and even if the thickness or shape of the plate is changed, the position of the optical sensor is readjusted every time. There is a problem such as a narrow detection range.

Therefore, the present invention irradiates a laser beam having a beam width of a certain width to the side of the transfer plate to generate a high light spot reflected light point, then captured by a surveillance camera, processed into a binary image signal and then scanned (scan) Calculate the number of pixels, but use several times of repetitive learning (self-learning) to obtain the average number of pixels per sheet thickness and store them in memory to use as reference pixel values for comparison. The reflected light point of the transfer plate inputted to the camera is subjected to binary image processing as described above, and scanned to calculate the number of pixels, and then compared with the reference pixel value, the input pixel number is equal to the reference pixel number by self-learning or the tolerance range If the board is judged to be transferred to the desired number of sheets, and the input pixel is out of the tolerance range of the standard pixel by self-learning Is the control signal and the alarm output signal is determined as being more than the number of transmission failure or transfer sheet causes the transfer or processing of the sheet material break.

On the other hand, when acquiring a thickness image signal of a plate with a camera under general illumination, the amount of light (or brightness or luminance) is so weak that the thickness cannot be detected or the error of detecting the plate thickness is large. By irradiating a laser beam with a high width to the side of the plate with a predetermined width to be incident to the surveillance camera with a light point of sufficient light amount to maximize the detection capability.

In addition, by using a laser filter in the surveillance camera to filter out various noises other than the laser reflected light (images other than the noise, the bright portion reflected by the laser beam) as shown in Figure 4 to remove the unwanted noise, the laser beam A specific wavelength, such as a red laser, which is clearly visible from the visible light or the illumination of the laser beam, is used, and the laser filter also uses a filter that passes only the specific laser wavelength to prevent malfunction by preventing ambient noise.

In addition, the monitoring camera and the laser beam projector are inclined at the upper and lower portions of the point slightly spaced apart from the side of the plate, so that even if the reflected light point due to the laser beam irradiation is formed on the side and the bottom of the plate, the surveillance camera installed at the top is inclined to the side of the plate By only receiving the reflected light point to be formed to prevent the error due to the expansion of the reflected light point even if the plate flows.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

In the present invention, the number of plates may be detected using only a surveillance camera (CCD camera) 2 as shown in FIG. 2, but under normal lighting, the amount of the sheet 4 is sufficiently low because the amount of light is weak or not greatly distinguished from the background image. Since it cannot be detected, the laser beam 6 having a sufficient amount of light and a wide width by the straightness or the local illumination is irradiated to the side surface 8 of the plate 4 so as to obtain the reflected light point 10 of sufficient brightness. 10) obtains the average thickness of the plate 4 by receiving it with the surveillance camera 2, converting it into a binary image signal, and then scanning it to calculate the number of pixels. When the number of pixels is compared and judged to be within the tolerance range, it is determined that the plate 4 is the desired number of sheets. do.

FIG. 1 is a configuration diagram of the present invention, wherein the surveillance camera 2 having the laser filter 12 and the laser beam projector 14 having the beam width are inclined at the upper and lower portions of the sheet 4 slightly spaced apart from the side of the plate 4. The laser beam 6 of a predetermined width is irradiated upward by facing the side surface 8 of the plate 8 so that the reflected light spot 10 is formed on the side surface 8 of the plate 4, and the reflected light spot 10 is a surveillance camera. Video input to (2).

In the above, the laser beam projector 14 and the surveillance camera 2 may be installed in front of the side surface of the plate 8 to obtain the reflected light point 10, but the laser beam ( 6) This is irradiated not only to the side surface of the plate 4, but also to the top or bottom surface, so that the reflected light point 10 is greatly expanded and a detection error may occur accordingly, so that the present invention is slightly spaced apart from the side of the plate 4. The surveillance camera 2 and the laser beam projector 14 are inclined at the upper and lower portions of the spot to face the side surface 8 of the plate 4 so that the reflected light point 10 and the side surface 8 of the plate 4 are inclined. The irradiated to the bottom surface 16, so that the monitoring camera 2 located at the upper side is only the reflection light point 10 formed on the side of the plate 4 is input, so that the detection error is prevented even if the plate 4 flows.

Although the inclinations θ1 and θ2 of the surveillance camera 2 and the laser beam projector 14 may be approximately 25 degrees, the inclination θ2 of the laser beam projector 14 may be determined by the plate material 4 due to external pressure. It is sufficient that the laser beam 6 does not irradiate the upper surface of the plate 4 even if it flows, and the inclination θ1 of the monitoring camera 2 is also used even if the plate 4 flows by external pressure. It is enough that the bottom face 16 is not caught.

On the other hand, the reflected light point 10 image captured by the surveillance camera 2 as shown in FIG. 4 is binary imaged by the computer 18 as shown in FIG. 5 to remove unnecessary noise, and then scanned and scanned. By calculating the thickness value of the plate (4) by obtaining the average number of pixels by the self-learning several times the above process, and then stores the basic thickness value of the plate 4 in the memory 20 of the computer 18 By calculating the number of pixels by binarizing and scanning the reflected light points 10 of the plate 4 which is subsequently input to the surveillance camera 2 in the state where the basic thickness value is obtained, the number of pixels of the monitor plate 4 is calculated. The controller 22 connected to the computer 18 determines that the plate 4 is conveyed in an undesired number of sheets when the thickness value is obtained and is out of the reference thickness value or the reference thickness tolerance range by comparing with the reference thickness value. Process by outputting control signal and alarm signal Allow the device 24 or the transfer robot 26 to stop.

In the present invention, since the side surface thickness of the plate 4 is monitored by the highly luminous reflected light point 10 by the laser beam 6, regardless of the surrounding conditions (weather, ambient light, spatter, oxidation degree of the metal plate, reflection coefficient) Accurate detection is possible.

In addition, the surveillance camera 2 is attached with a laser filter 12 through which only the laser beam is transmitted so as to remove unnecessary noise other than the laser beam 6. That is, when the laser filter 12 is not used as shown in FIG. 2, the background image 28 around the plate 4 is simultaneously input in addition to the side image signal of the plate 4 to become an image noise factor. In the case of irradiating the laser beam 6 on the side surface, as shown in FIG. 3, a distinct reflection light point 10 is formed on the side of the plate 4, and in this state, the laser filter 12 is attached to the front surface of the camera 2. If only the laser beam 6 is inputted as shown in FIG. 4, the unwanted noise is removed, and the thickness of the input plate 4, which is self-learned by binarizing the signal where most of the noise is removed, as shown in FIG. In comparison, the transfer length of the plate 4 is determined.

In the present invention, the thickness of the plate 4 is to detect the number of pixels per unit light point by scanning the light point converted into a binary image as shown in Figure 8 by the scanning line, the pixel distributed on the horizontal scan line 30 is self-learning If the number of pixels is equal to or within the allowable error range, the plate 4 is judged to be one, and if the number of pixels distributed on the horizontal scan line 30 is about two times, the computer 18 determines that two sheets of plate 4 are overlapped. By outputting a control signal to the controller 22, the sheet processing apparatus 24 or the sheet conveying robot 26 are controlled to prevent the sheet 4 from being conveyed to two or more sheets.

In the present invention, a wide angle or plate 4 is used by using a zoom lens 32 in the camera 2 in consideration of the separation distance between the plate 4 and the surveillance camera 2 and the type, size, and thickness of the plate 4. Adjust the separation distance from

On the other hand, Figure 10 is a self-learning flow chart (Self learning rootine), when the operator selects the type or thickness of the plate 4, if there is no self-learning data (reference thickness of the plate) plate to the monitoring camera (2) Acquire the average thickness according to the number of pixels by capturing the reflected light point 10 of (4), and obtain the average thickness by repeated learning (average 5 to 10 times) several times, and then store it in the memory 20 of the computer 18. It is stored and used as a reference thickness for comparison, and if the learned reference thickness already exists in the computer memory 20, the average thickness is obtained by calculating the thickness of the self-learned plate 4 based on the computer memory 20. Self-learning is completed by resave to thickness

In addition, the supervisory rootine is shown in FIG.

When the operator selects the type or thickness of the plate 4, the type or thickness signal of the plate 4 is inputted. If there is no learning data in the computer 18, the self-learning routine is executed as in the above-described self-learning flowchart. The average thickness of the plate 4 is obtained to store the reference thickness. If the self-learning data is present, the light spot 10 of the plate 4 is captured by the surveillance camera 2, and then binary imaged and scanned. The thickness of the plate 4 is obtained by calculating the number of pixels, and the reflection light point 10 of the plate 4 to be monitored is obtained, and then one sheet (or the set number or number of sheets) is compared with the reference thickness by self-learning. In order to judge that there is no abnormality in the number of conveyances on the back side, each time the plate 4 is monitored, the thickness of the self-learned plate 4 and the input plate 4 are averaged and stored in the memory 20. The reference thickness of (4) was made to learn in an optimal state. .

Reference numeral 34 is a plate adsorption port.

According to the present invention constructed as described above, the laser beam projector 14 and the surveillance camera 2 are installed obliquely using the transfer robot 26, and then the plate 4 is transferred to be monitored while the plate 4 is transferred. When the laser beam 6 is irradiated to the side surface of the plate 4, the high light intensity reflected light point 10 is generated on the side surface of the plate 4, and the light point 10 is incident to the surveillance camera 2 inclined downward and is binary imaged. The number of pixels distributed on the horizontal scan line 30 is matched with the number of pixels according to the reference thickness stored in the memory 20 or compared with the reference thickness processed by the signal and then input to the computer 18. If the transfer plate 4 is determined to be the normal number of sheets, the number of pixels distributed on the horizontal scan line 30 as shown in the sixth, seventh and eleven degrees is stored in the memory 20. If the number of reference pixels is two or more times, the plate 4 To be judged.

As described above, the present invention can accurately monitor the longevity of the plate 4 by the reflected light point 10, and even if the type of the plate 4 is changed, an auxiliary jig for correcting the position of the optical sensor is unnecessary as in the prior art. Productivity can be improved by fully automated and unmanned, and the input quantity, production quantity, error quantity, etc. of plate 4 can be built as a database and used as production information, and can be extended to other parts inspection. The response speed is 1/30 seconds, which is very useful invention with very fast and reliable effect.

Claims (2)

The surveillance camera 2 having the laser beam projector 14, the laser filter, and the zoom lens at a point slightly spaced from the side of the plate 4 is tilted upward and downward to have an inclination θ1 and θ2 with respect to the plate 4. Installed so as to face the side of the plate 4, and having a predetermined width of the reflected light point 10 formed on the side of the plate 4 by the laser beam projector 14 to be input to the monitoring camera 2, monitoring The light spot 10 input to the camera 2 is scanned to obtain the number of pixels, and the above process is repeated to obtain the reference pixel number according to the thickness of one sheet of paper and to store it in the memory 20. The reflected light point 10 of the monitoring target plate inputted in the process is scanned, takes the number of input pixels, and compares it with the reference pixel number to determine that the plate 4 is an undesired number of times when it is out of the tolerance range. Sheet conveying characterized in that signals are output It can monitor. The method according to claim 1, wherein the reference thickness value is corrected each time by averaging the thickness of the self-learned plate and the input monitoring plate in the memory 20 each time the plate 4 is monitored. Plate transfer quantity monitoring device.