JPS6248878B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention involves attaching a code mark indicating the model number, date of manufacture, etc. to the outer surface of a container such as a glass bottle, and detecting the code mark photoelectrically to determine the code. The present invention relates to a container code identification method for performing identification.

As such a container code identification method, the present applicant has already provided the method described in Japanese Patent Application Laid-Open No. 56-57171.

To summarize this with reference to FIGS. 1 and 2, on the bottom surface 2 of the glass bottle 1 whose code is to be identified, elongated protrusions 3 are formed as code marks during molding.

These convex parts 3... are arranged along a circular orbit R centered on the axis of the glass bottle 1, and the whole has a binary logic code system.

That is, the code is a radial equisector P that radiates from the center on the bottom surface 2 with a certain phase difference.
. . . are assumed, and these radial equisectors P .

The glass bottles 1 are sent one by one onto the detection port 5 of the gauging plate 4, where they remain stationary and are optically scanned (in the direction of the arrow in Fig. 2) on the bottom surface 2 by a rotary scanner 6 along a circular orbit R. .

The rotary scanner 6 includes a light projector 7 that illuminates the bottom surface 2 and a light receiver 8 that receives the reflected light.
It is constantly rotated by a motor 9.

Power is supplied to the rotary scanner 6 by a rotary transformer 10.

The light receiver 8 has one photoelectric conversion element, and its output is transmitted to the light emitter 1 that constitutes the photo isolator 11.
2 and photoreceiver 13, and is then input to an external code conversion circuit.

At the same time as the rotary scanner 6, the rotary encoder 14 rotates to output a clock pulse, which is also input to the code conversion circuit.

The code conversion circuit samples the output of the photoreceiver 13 at intervals corresponding to the equal dividing angles of the radiation equisectors P . . . .

With this, the convex portion 3...2 depending on whether it is present or not.
A progressive electrical signal is obtained, which is analyzed by a microcomputer and finally the code is identified.

Although the method previously provided by the present applicant has such a configuration, it has the following drawbacks.

(1) Since the presence or absence of marks (protrusions) is swiveling and scanned by the rotation of the rotary scanner, a complicated power supply means (having a movable part and a fixed part) is required to supply power to the rotating rotary scanner. (rotary transformer) is required.

(2) Such power supply means generally tend to generate voltage noise, which often leads to false detection.

(3) Since the receiver rotates together with the emitter, a special means (photo-isolator) is required to extract the output of the receiver to the outside.

(4) Among various glass bottles, it is not preferable to form a convex part on the bottom surface, and in such cases, a convex part is formed on the outer peripheral surface of the body. If such a glass bottle is to be kept stationary and the outer peripheral surface of its body is scanned, the projector and receiver must be configured to rotate around the glass bottle. A complicated mechanism is required for its turning operation.

Since the method of rotating and scanning the glass bottle while keeping it stationary has such drawbacks, the present applicant has proposed the method of Utility Model Application No. 1983-
The one described in Publication No. 83858 is provided separately.

When this is schematically illustrated in Figures 3 and 4, a long and narrow convex portion is formed on the lower end (so-called hem) of the outer peripheral surface of the body of the glass bottle 1' using the same principle as that shown in Figures 1 and 2. 3'... is formed.

The glass bottle 1' is placed on a turntable 16 rotated by a motor 15 so as not to fall over, and is rotated in the direction of the arrow.

The presence or absence of a protrusion on the bottom of the rotating glass bottle 1' can be determined by the sensor 1 fixed at a fixed position.
Detected by 7.

As shown in FIG. 4, the sensor 17 includes a built-in light projector 18 that illuminates the bottom of the glass bottle 1' and a light receiver 19 that receives the reflected light.

The photodetector 19 has one photoelectric conversion element, and the output of the photoelectric conversion element is sampled based on clock pulses from a rotary encoder 20 that rotates simultaneously with the turntable 16, as shown in FIGS. 1 and 2 below. The code is identified in the same way.

However, in the case of the method shown in Figures 3 and 4, (a) If the glass bottle and the means for rotating it (turntable) slip, the radial equisector P
. . . and the sampling interval based on the clock pulse of the rotary encoder 20, which causes a misdetection, which is a drawback.

Furthermore, as disclosed in Japanese Patent Application Laid-Open No. 53-1047, code marks and timing marks for reading the codes are provided above and below on the outer peripheral surface of the glass bottle, and the timing marks at both ends are The code mark and timing mark, as well as the marks that serve as the start signal and end signal, are detected separately by the corresponding photodiodes, and the timing mark is activated. Some devices sample the signal associated with code mark detection while detecting and obtaining timing pulses.

However, this method has the following drawbacks.

(A) In addition to the photodiodes placed above and below to read the code mark and timing mark, another photodiode is required next to it to detect the start signal or end signal, making the detection structure complicated. .

(B) In addition to the code mark, a timing mark and marks for starting and ending signals are required.

(C) Since the code mark and timing mark must be detected separately, the device tends to become complicated.

The object of the present invention is to be free from the drawbacks shown in (1) to (4) above, as well as (a) to be free from the drawbacks shown in (A) to (C), and to avoid the drawbacks shown in the conventional art described above. An object of the present invention is to provide a container code identification method that can omit the rotary encoder, sampling circuit, etc. that are necessary in the three methods described above.

In order to achieve such an object, the present invention provides m that are encoded on a circular orbit on the outer surface of the container by varying the length in a direction orthogonal to the circular orbit.
marks are provided with a no-mark section interposed therebetween, and the outer surface of the container is irradiated with light while the container is rotated.
By simultaneously detecting the reflected light as n-bit parallel binary electric signals using n photoelectric conversion elements arranged in parallel in the orthogonal directions, digital data corresponding to the length of each mark and digital data corresponding to the no-mark section are generated. data, and determines whether it is a marked section or an unmarked section based on the content of each digital data, and also compares the size of the previous digital data and the current digital data to determine whether the marked section is a marked section or an unmarked section. It is characterized in that it is determined whether or not the marks are alternating, and when they are alternating, only the digital data corresponding to the mark section is taken out, and the identification code is obtained by repeating this process m times.

Embodiments of the present invention will be described in detail below with reference to FIGS. 5 to 14.

As shown in FIG.
The glass bottles B, which are randomly sent in a row, are guided to the carry-in bypass 22, where they are arranged at regular intervals by a feedworm 23, and then sent to the inspection station CS.

The glass bottles sent here are fitted into and captured by the star wheel 24 that rotates intermittently, and are thereby intermittently forwarded to the first inspection position.
The model number was inspected by the method of the present invention at CP ₁ ,
Next, predetermined defect inspections are carried out at the second inspection position CP ₂ and then at the third inspection position CP ₃ .

As a result of such an inspection, glass bottles that are found to be of good quality are introduced into the discharge bypass 25, through which they are loaded onto the conveyor 21 again and transported to another location.

On the other hand, glass bottles determined to be defective are sent to the rejection position DP and rejected.

First inspection position CP ₁ for carrying out the method of the invention
The details are shown in FIG.

The glass bottle B is captured by fitting into the bottle sockets 26a, 26b of the upper and lower discs 24a, 24b of the star wheel 24, and as the star wheel 24 rotates intermittently, the glass bottle B is captured on the gauging plate 27. The aircraft will be held for a certain period of time in the position shown in the figure while on board.

At this position, a bottle drive wheel 28 is mounted on a shaft, and a light emitter 29 and a light receiver 3
0 is installed.

The bottle drive wheel 28 is constantly rotating, and the glass bottle B, which is held for a certain period of time as described above, is rotated at least once by the bottle drive wheel 28.

On the lower end (hem) of the outer circumferential surface of the body of glass bottle B, a code mark consisting of two long and short protrusions is attached in the following code system.

In addition, below, the longer convex part will be referred to as the long mark, the shorter convex part will be referred to as the short mark, and the long mark will be designated by the symbol M ₁
The short mark is represented by the symbol M ₀ . In addition, when these long and short marks are collectively referred to, they are simply referred to as marks.

As shown in detail in Fig. 7, the long mark _M1 and the short mark _M0 have the same width and are both elongated in the axial direction of the glass bottle B, but the long mark _M1 is on the upper side than the short mark _M0 . It is twice as long as the
It is digitized based on this difference in length and shortness.

Incidentally, if the long mark M ₁ is a binary ``1'', the short mark M <sub>0</sub> is a ``0''.

The total number m of marks per glass bottle is determined, and in this example, it is 10 marks.

The 10 marks are arranged in a row (therefore, parallel to each other) at a constant pitch in the circumferential direction (circular orbit) on the outer circumferential surface of the glass bottle, and are 10 bits as a whole.
It has a hexadecimal code system.

Therefore, in the case of the glass bottle shown in Figure 6, if you look at the mark arrangement counterclockwise, it reads "short, short, long, short, short, long, short, short."
This means that the system is "0001100100".

The light projector 29 illuminates the hem of the glass bottle from below the gauging plate 27 through the window hole 31 thereof. Irradiation area LA of this light beam (Figure 7)
is large enough to illuminate the entire length of one long mark _M1 .

The light receiver 30 is connected to the gauging plate 2 in order to receive the reflected light beam reflected from the glass bottle on the front side.
It is located above 7.

A complex element type photo sensor 32, shown in a simplified manner in FIG. 7, is attached to the front surface of the light receiver 30.

The photo sensor 32 includes n photoelectric conversion elements 33, in this case eight photoelectric conversion elements 33, arranged at a constant pitch in the longitudinal direction of the mark, and a circuit (described later) that amplifies and shapes the electric signals from these elements. It has a one-chip structure.

These eight photoelectric conversion elements 33... and the marks mean that all eight photoelectric conversion elements detect the long mark M1, and the four photoelectric conversion elements on the lower half detect the short mark _M1 .
The correspondence relationship is such that M ₀ can be detected.

In other words, the light beam reflected from a part without a mark (a flat part without a convex part) does not enter any photoelectric conversion element, whereas a long mark
The light beam reflected from M ₁ enters all the photoelectric conversion elements from the first to the eighth counting from the bottom, and the light beam reflected from the short mark M ₀ enters the four photoelectric conversion elements from the first to the fourth. The light enters the photoelectric conversion element.

Therefore, as shown in FIG. 9, the output of each photoelectric conversion element 33 is H level when there is a mark, L level when there is no mark, and ₁ when there is a long mark M1.
The output of all photoelectric conversion elements from the 8th to the 8th is H.
In the case of level and short mark M ₀ , the outputs of the four photoelectric conversion elements from the first to the fourth are at H level, and hereinafter, the section without a mark is referred to as a space.

In addition, even if the long mark M is ₁ , from the 1st to the 8th
There are times when the outputs of the first eight photoelectric conversion elements do not all go to H level, and even if the short mark _M0 , the outputs of the first to fourth photoelectric conversion elements all go to H level. Furthermore, the length of the long mark _M1 and the short mark _M0 may be shorter than the specified length due to molding, and the long mark M1 and the short mark M0 may be shorter than the specified length due to molding. The combination of M ₁ and the short mark M ₀ as described below is intended to be carried out with the required tolerance.

The outputs of the eight photoelectric conversion elements 33 are each amplified by the eight amplification and comparison circuits 34 shown in FIG. 8, then compared with a reference voltage, and marked as shown in FIG. A binary electric signal is output from each output terminal of the photo sensor 32 depending on the presence or absence of the signal.

Therefore, the electrical signal output in this way is a serial binary electrical signal depending on the presence or absence of a mark for each photoelectric conversion element, and the 8 photoelectric conversion elements as a whole is an 8-bit parallel binary electrical signal. It becomes a signal.

This 8-bit parallel binary electric signal is input to the microcomputer 35, where it is digitally processed in real time.Next, the digital processing will be explained with reference to the flowchart in FIG. 10 as appropriate. .

Furthermore, when the 8-bit parallel binary electric signal inputted to the microcomputer 35 is expressed in hexadecimal notation, as shown in FIG. 9, the long mark _M1 is "FF", the short mark _M0 is "0F", and the space is is "00"
However, in the microcomputer 35, as will be described later, after a determination is made as to whether or not marks and spaces are alternated, the codes are identified using those corresponding to both the long and short marks M ₁ and M ₀ . For convenience, in the following, of the input 8-bit parallel binary electrical signals, those corresponding to the mark section (long mark _M1 and short mark _M0 ) will be referred to as data X, and those corresponding to the space section will be referred to as data X. It is called data Y. Furthermore, when these X and Y are collectively referred to, they are simply referred to as data.

When the glass bottle B is positioned at the first inspection position _CP1 as described above, an input timing signal is input to the microcomputer 35.

When it is determined that this input timing signal is input (step 100 in FIG. 10), data is input through the data input port 36 (step 101).

This input data (X or Y)
is first stored in the current value register 37 and compared with the contents of the previous value register 38 by the comparator 39 (step 102).

The comparator 39 compares the contents of both registers 37 and 38, that is, determines whether the previous value>the current value or the previous value≦the current value.

Initially, the contents of the previous value register 38 are 0 in decimal (00 in hexadecimal), so whether the data stored in the value register 37 this time is X or Y, it is stored in the previous value register 38. loaded.

When loaded in this manner, new data is input and stored in current value register 37 and similarly compared by comparator 39.

If the previous value<the current value, that is, the content of the previous value register 38 is data Y and the content of the current value register 37 is data X, the same process is performed again. That is, steps 101 and 102 are repeated until the contents of the previous value register 38 become data X.

When the previous value>the current value, the comparator 40 determines whether the newly inputted data (step 103) and stored in the current value register 37 is 0, that is, whether it is data Y or not ( Step 104) is performed.

If it is data Y, the contents of the previous value register 38, that is, data X, are loaded into the data memory 41 (step 105), and the data
That is, it is stored at the storage address specified by the address controller 42.

At the same time, the address controller 42
The designated address is incremented (step 106), and the data counter 43 is also incremented (step 107).

The count-up number of this data counter 43 corresponds to the total number m of marks per glass bottle. In this example, the number is set to "10", and the above series of steps 101 to 107 are repeated until the data counter 43 reaches "10" (step 108).
Repeated.

To summarize this series of steps 101 to 107, first, data X is detected in steps 101 and 102, then data Y is detected in steps 103 and 104, and then data This process is repeated 10 times to sequentially store data X related to the 10 marks in the data memory 41.

In other words, each time data is input, the microcomputer 35 determines whether it is a mark or a space, and determines whether or not they are alternating. and sequentially store them at specified memory addresses in the data memory 41.
By repeating this process only 10 times, 10-bit code data consisting of 10 marks is obtained.

Therefore, if the long mark M ₁ is a binary number "1" and the short mark M ₀ is "0", in the case of glass bottle B shown in FIG. 6, as shown in the bottom row of FIG. Code data of "0001100100" (200 in decimal) is finally obtained. It is clear that this code data can be accurately obtained regardless of the variation in the rotational speed of the glass bottle B due to slipping between the glass bottle B and the bottle drive wheel 28.

Note that the orientation of the glass bottle B in the circumferential direction when positioned at the first inspection position CP ₁ is random, and the data first input to the microcomputer 35 is determined at which position in the circumferential direction of the glass bottle B. It is not determined whether the data
The input order is random, but the bit array (order from MSB to LSB) of the code data finally obtained is arranged to match the mark arrangement order.

The code data (that is, model number) obtained in this way is inputted to the microcomputer 35 by an output timing signal from the outside (step
109) Output when done (step 110)
be done.

This output data is input to, for example, a printer and printed out, or sent to a host computer for comprehensive data processing.

When the input timing signal input while performing the above series of operations disappears, the reset circuit 44 resets the register 3.
7, 38 and data memory 41 are reset,
Returning to the initial state, the same process as above is performed for the next glass bottle.

In the embodiment described above, the marks were digitized (encoded) by one continuous long mark _M1 and one short mark _M0 , but even if the marks are not necessarily continuous, the photoelectric conversion elements 33... If only the portion detected by the mark is substantially a mark, digitization by length or shortness can be performed in the same manner as above. For example, as shown in _FIG .
M', and the two mark segments correspond to the difference in number, that is, the binary number "1",
What corresponds to "0" may be digitized as one mark segment.

Alternatively, as shown in FIGS. 12 and 13, the mark may be composed of a dot pattern M'' and digitized depending on the number of dots.

When the number of dot patterns constituting one mark is small, as shown in FIG. 13, a plurality of photoelectric conversion elements 33...
If even one of them corresponds to the H level, it is determined that it is a dot pattern, and false detection can be reduced.

Further, such a mark does not necessarily have to be convex, and may have any shape as long as it can be optically detected by reflecting light.

Furthermore, in the embodiments described above, marks and spaces are detected by making sure that only the reflected light from the former enters the photoelectric conversion element and the reflected light from the latter does not enter the photoelectric conversion element. Alternatively, the reflectance of the two may be made different so that the light enters the photoelectric conversion element, and the difference in voltage level based on the difference in reflectance may be used to detect the It may be detected.

In addition, as shown in FIG. 14, marks are attached to the bottom surface B' of the glass bottle B in the circumferential direction (on a circular orbit), for example, a long mark _M1 elongated in the radial direction and a short mark _M0 , and both of these marks are attached. may be detected by photoelectric conversion elements 33 arranged in the radial direction.

Furthermore, it is clear from the above that the method of the present invention can be applied to containers other than glass bottles, and also to identification of codes other than model numbers.

As described above, the present invention has the following effects.

Using n photoelectric conversion elements arranged in parallel in a direction perpendicular to the circular orbit of the container, the lengths of m marks are sequentially detected as digital data, and the magnitude of the previous digital data and the current digital data is compared. By this, it is determined whether marked sections and unmarked sections are alternating,
Since the identification code is obtained by extracting only the digital data corresponding to the mark section when alternating and repeating it m times, there is no need for timing pulses or sampling pulses according to the rotation of the container. There is no need for a counter to count marks, and the detection circuit configuration can be simplified.

It is not necessary to arrange the marks at equal intervals, and no special consideration is required for their arrangement.

Since the marks are detected as a pattern, there is no need to count the marks.

Unlike the conventional method of sampling the electrical signal from a single photoelectric conversion element according to the rotational speed of the container and converting it into a binary electrical signal, fluctuations in the rotational speed of the container may lead to false detections. Even if the rotational speed of the container changes, the code of m marks can be accurately detected and identified.

The rotary encoder, sampling circuit, etc. that were required in the past can be omitted, making it economical and easy to design the circuit.

Other marks than the encoded marks, ie timing marks for its reading timing or special forms of marks for the start and end of reading, are not required.

[Brief explanation of the drawing]

Figures 1 and 2 show a conventional example in which a glass bottle is kept stationary and rotated for scanning. Figure 1 is a simplified sectional view, Figure 2 is a bottom view of the glass bottle, and Figures 3 and 4 are views of the glass bottle. Another conventional example is shown in which detection is performed while rotating. Fig. 3 is a partially cutaway front view of the
The figure is an enlarged sectional view showing the relationship between the glass bottle and the sensor, Figures 5 to 10 show an embodiment of the present invention, Figure 5 is a simplified plan view to explain the flow of the glass bottle, and Figure 6 is A perspective view of the main parts, FIG. 7 is an explanatory diagram showing the relationship between the mark and the photoelectric conversion element, FIG. 8 is a block diagram showing an example of the mark detection and identification device used in the method of the present invention, and FIG. A time chart for explaining the signal processing process, FIG. 10 is a flow chart for explaining its operation, FIGS. 11, 12, and 13 are explanatory diagrams showing other examples of marks, and FIG. 14 is a glass bottle. FIG. 7 is a bottom view showing another example in which a mark is attached to the bottom surface of the device. B...Glass bottle (container), M ₁ , M ₀ , M',
M″...Mark, 33...Photoelectric conversion element, 35...
...A microcomputer that digitally processes electrical signals from photoelectric conversion elements.

Claims (1)

[Claims] 1. m marks encoded by varying lengths in a direction orthogonal to the circular orbit on a circular orbit on the outer surface of the container, with unmarked sections interposed; By irradiating light onto the outer surface of the container while rotating the container, and simultaneously detecting the reflected light as an n-bit parallel binary electric signal with n photoelectric conversion elements arranged in parallel in the above-mentioned orthogonal directions, each mark can be detected. Digital data corresponding to the length and shortness and digital data corresponding to the unmarked section are obtained, and based on the content of each digital data, it is determined whether it is a marked section or a no-marked section, and the previous digital data and the current digital data are By comparing the magnitude of , it is determined whether marked sections and non-marked sections are alternating, and when they are alternating, only the digital data corresponding to the marked sections are extracted, and this is repeated m times. 1. A container code identification method characterized by obtaining an identification code by. 2. Claim 1, in which the mark is digitized by the length in the direction perpendicular to the circular orbit.
Container code identification method described in Section. 3. The container code identification method according to claim 1, wherein the mark is formed into a dot pattern and is digitized based on the difference in the number of the dot patterns. 4. The container code identification method according to claim 1, wherein the mark is formed into segment patterns and is digitized based on the difference in the number of segment patterns.