JPH03235011A - Method of selecting container of different kind - Google Patents

Abstract

PURPOSE:To check the configuration of a container in real time and select the container of a different kind by converting image signals obtained by image sensing to binarized image data, integrating the number of white or black pixels due to a binarized level and comparing image areas with each other. CONSTITUTION:Area measuring ranges A and B stored in an input buffer 45 are assigned by a frame setting volume group 49 and binarized data in the area measuring ranges A and B are fed to a measuring unit 42. A main control unit 43 calculates an image area from the number of black pixels measured by the measuring unit 42 followed by a second image area calculation performed as delayed a little. Calculating a difference between the areas in the first and second calculation, whether the area increases is judged and the stop of an increase is confirmed on the area with a large increase rate. The image area of a bin 1 is detected in the same timing as a confirmation timing. On the other hand, the reference values of the image areas of the bin of the same kind as that of the bin 1 in the measuring ranges A and B are stored in the input buffer 45 by a reference input button 50. The bin of a different kind is selected by comparing the reference values with the detected area values.

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial Application Field) The present invention relates to a method for sorting out different types of containers mixed into a production line in real time in a production process that handles containers such as beer bottles.

(Conventional technology) Traditionally in the beer industry, collected bottles have been mixed with other companies' bottles, and even if the mixed bottles have another company's name on them, they are likely to be sold with a new company label attached. There seemed to be no problems. However, in today's highly competitive society, beer industry seems to be a problem.

Therefore, as a method for selecting other companies' bottles, there has been a method in which a photosensor is placed in a characteristic part of the bottle as shown in FIG. 6, and other companies' bottles are detected by the ON/OFF state of the photosensor.

(Problem to be Solved by the Invention) The above method has poor sorting accuracy, and if the accuracy is to be improved, it is necessary to increase the number of photosensors, and even if the number is increased, reliability regarding accuracy does not improve easily. Further, there is a problem in that the adjustment of the photosensor is often difficult and the reliability of the device 5 as a whole is low.

The present invention aims to solve the above-mentioned problems and to perform the sorting of other companies' bins with high precision and in real time.

(Means for Solving the Problems) The present invention captures an image of a transported container using a two-dimensional imaging device, specifies a measurement range in the image, and converts the imaged image signal into 21fi image data. By integrating the number of white or black pixels depending on the binarization level,
The method is characterized in that the image area of the container within the imaged measurement range is calculated, and when the image area reaches a certain value, it is compared with the area of containers of the same type stored and selected.

(Function) Converts the captured video signal to binary image data,
Various methods can be considered to calculate the image area of the imaged container by integrating the number of white or black pixels depending on the value level, but in the present invention, the method disclosed in Japanese Patent Application Laid-Open No. 63-206156 by the applicant The area is calculated by the planimetry method disclosed in .

As disclosed in the same publication, the same area measurement method is based on the fact that the object whose area is to be measured has a contrast, is entirely brighter or entirely darker than the background within the measurement range, and the time during which the binarization level is changed (actually (between 0.17 seconds and 8.53 seconds) is within the measurement range, the processing time is short and economical, and the distortion caused by fluctuations in surrounding brightness is corrected and the area is accurately measured. can be measured.

Also, the timing when the image area reaches a certain value is the timing when the increase in image area is compared, and if it is increasing, the area is continued to be monitored and the increase has stopped, as shown in the flowchart in Figure 4. The time point is defined as the timing when the image area reaches a certain value, and is defined as the detection point.

Therefore, it is possible to compare and select the image area of the transported container with the area of containers of the same type in real time and with high precision.

(Example) Embodiments of the present invention will be specifically described below with reference to the drawings.

FIG. 1 is a schematic perspective view showing the implementation state of the method of the present invention.

The beer bottles 1.1... that have been collected are placed on a conveyor container 2 and supplied to a recycling process 1it (not shown). 3 captures an image of the beer bottle being conveyed with a TV camera, and inputs the image signal to the dissimilar container sorting device 4. Reference numeral 5 denotes a monitor television, which is used by the worker to specify the measurement range, etc.

The conveyor container 2 for conveying the beer bottle 1 may be any type of conveyor as long as it conveys the beer bottle 1 in an upright manner,
It is necessary that the lateral and height positional dimensions of the object relative to the stationary system be guaranteed. Here, a belt conveyor runs on a fixed guide plate 6, and guides 7.7 are provided on both sides of the upper part of the belt conveyor.

Next, the configuration of the dissimilar container sorting device 4 will be explained with reference to FIG. FIG. 2 is a block diagram showing the overall configuration of the dissimilar container sorting device 4. As shown in FIG.

The video signal of the beer bottle l captured by the TV right camera is input to the binarization and image control unit 41, where it is amplified and noise-cut to convert the video signal into a continuous analog signal, with horizontal resolution depending on the sampling frequency. , the vertical resolution is converted into a discrete digital signal in units of scanning lines, and converted into a 21-digit signal based on a separately set binarization level. In addition, the area measurement ranges A and B (fifth
) is specified, and the binarized data of the measurement ranges A and B are sent to the measurement section 42.

The area measurement ranges A and B specified here include ranges A and B that include the features of the shape of the bins on the monitor screen 54 shown in FIG. Although two measurement ranges were specified,
It may be one place instead of one place, or it may be more than one place.

The measurement unit 42 calculates the image area of bin l within the measurement ranges A and B by counting the number of different pixels (or the number of white pixels) from the received binary data of the measurement ranges A and B. In order to do this, data is transmitted to the main control unit 43.

As shown in the timing flowchart of FIG. 4, the main control unit 43 performs calculation 1 of the image area (S31) based on the number of black pixels measured by the measurement unit 42, and then, after a slight time shift, performs calculation 1 (S332) of the image area. 534)
, for those with a large increase rate (335), confirm that the increase has stopped (536), use this confirmed timing as the detection timing for the image area of bin l, and detect the area. By pressing the button 50, the base 1! of the image area in the measurement range A5B of the same type of bottle as the bottle 1 is selected! The value is stored, and by comparing the same base value with the detected area value, if it is within the tolerance range of the base $ value, it is sorted into the same kind of bin, and if it is outside the tolerance range, it is divided into different kinds of bins. Sorting out different types of bins is performed.

The binarization level determination unit 44 determines the optimum binarization level with high precision without being affected by the surrounding brightness, and the details are described in the above-mentioned Japanese Patent Laid-Open No. 63-206156. The change in the number of black pixels measured by the measurement unit 42 is expressed as 2f+
The optimum binarization level is determined with high precision by calculating the black pixel difference when the i conversion level is sequentially changed from dark to bright, and further calculating the difference between the black pixel differences (two-time difference).

The manual buffer 45 includes an allowable width setting volume 48, a frame setting volume group 49, a reference input button 50, and a start button 5.
1, etc. are connected, each storing a set value.

The allowable width setting volume 48 sets the allowable width of the reference value of the image area of bins of the same type as bin l. Although the allowable width can be given arbitrarily, it is set based on actual values measured by sampling bins of the same type as bin 1.

The frame setting volume group 49 is for setting area measurement ranges A and B, and is set at a location where the shape of the bin is characteristic.

The reference input button 50 sets a reference value for the image area of a bin of the same type as the bin 1, so it is set based on an actual value measured by sampling.

The start button 51 is a start button for the sorting device.

A 7-report buzzer 52 and a contact output relay 53 are connected to the output buffer 46, and output the results of sorting different types of bins. 47 is a power supply section.

Next, the details of the beer bottle sorting process will be explained according to the floater yard of BOTTLE in FIG. However, FIG. 3 is a flowchart of the selection process, and FIG. 4 is a flowchart for determining the detection timing as described above.

When the built-in control program for sorting different types of bins starts, the main control unit 43 first sets the stack pointer (Sl), clears the RAM (32), and
10 initialization is performed (S3).

Next, prepare to store the digital switch (S4),
Set the 1-shift level (S5) and the area measurement ranges A and B.S6) Select the monitor display screen (S7).

Depending on the operator's needs, the monitor display can be performed by displaying raw video signals, binarized video signals only within the area measurement ranges A and B, undisplayed video signals outside the frames, or binarized video signals inside the frames. It is possible to select a raw video signal, a mixed video signal, etc.

Next, check whether the reference value of the image area of the beer bottle of the same kind has been set (SS). If the base r$ value has been set, proceed to (Sll) and check the start button 51. If the reference value has not been set, proceed to (S9) and check the set state of the reference value input button 50, and if it is set, set the reference value input flag to ON.
Sll) Confirm the start button 51.

In (Sll), if the reference value input button is not set (S4), the process returns to preparing to store the digital switch and starts over.

Also, if you can confirm the start button 51 with (Sll), the next (
The process moves to S12).

In (S12), the allowable width of the area reference value is set, the binarization level is set again (513), and it is confirmed whether there is any change in the digital switch data (S14).

If there is a change in (S14), reset the reference lamp flag (322), or return to (S4) and try again, but if there is no change, check the detection timing (S1'5).
.

As described above, the detection timing is checked according to the timing flowchart shown in FIG. 4. If the detection timing is detected, the start flag is set (S37), so the process returns to the BOTTLI flowchart.

The detection timing is confirmed in (S16), and if the start flag is not set (S37), the process returns to (SL2) and starts over. If the start flag is set (S37), the process moves to (S17), and the measurement data of area measurement range A is input and compared with the 5 area reference value (318).

If the image area is not within the allowable range of the base fI value (S
23), outputs ON to the buzzer and relay contacts. If it is within the allowable range of the standard value, move to (S l 9) and manually calculate the measurement data of area measurement range B.
9), and compare with the area reference value (S20). Similar to menjo,
If it is not within the allowable range, output is turned ON (S23),
If it is within the allowable range, output QFF (S21),
Judging to be bottles of the same type.

When the output is turned on in (S23), it is determined that the bin is of a different type, and necessary measures are taken when the output is turned on.In (S24), a timer is set for the necessary measures, and after that, Go to (S21), turn off the output, and (S21) turn off the output.
Return to step 12) and proceed to the next sorting process.

At position C, where the TV camera 3 images the beer bottle L, it is necessary to take care to ensure that the contrast between the background and the beer bottle is clear. Sometimes it is necessary to make it interesting.

In addition, the sorting process is not limited to beer bottles; all containers, whether transparent, opaque, circular, or rectangular, can be sorted as long as they can contrast with the background and can be imaged.

(Effects) According to the method of the present invention having the above configuration, it is possible to check the shape of containers being transported in real time and to sort out different types of containers, and the sorting process is highly accurate and reliable. can.

[Brief explanation of drawings]

Fig. 1 is a schematic perspective view showing the implementation state of the method of the present invention;
The figure is a block diagram showing the overall configuration of the dissimilar container sorting device 4,
FIG. 3 is a flowchart of the sorting process, and FIG. 4 is a flowchart of determining the detection timing. FIG. 5 is an explanatory diagram showing a monitor screen, and FIG. 6 is a schematic explanatory diagram of a sorting method using a photosensor. l... Beer bottle 2... Conveyor = 4...
Dissimilar container sorting device

Claims (1)

[Claims] An image of the container being transported is captured by a two-dimensional imaging device, a measurement range is specified in the image, the captured image signal is converted to binary image data, and white or black data is generated depending on the binarization level. A method for sorting containers of different types, in which the image area of containers within the imaged measurement range is calculated by integrating the number of pixels, and when the image area reaches a certain value, it is compared with the area of containers of the same type stored in memory. .