JPH08322913A - Tablet inspection system - Google Patents

Abstract

PURPOSE: To correctly check tablets enclosed in a chartula in a tablet inspection system which is connected to a tablet wrapping device for enclosing plural tablets in a light transmitting chartula according to prescription information and adapted to inspect the tablets by recognizing the images of the tablets in the chartula discharged from the device. CONSTITUTION: An inspection apparatus 3 comprises a transport mechanism 4 for transporting a chartula 15 discharged from a tablet wrapping device 1 to a designated inspection position, a vibration applying mechanism 5 for applying vibration to a chartula to be fed into the inspection position to eliminate overlapping of tablets, a CCD camera 7 for photographing the chartula in the inspecting position, an image recognition processing circuit 31 for processing images obtained from the CCD camera 7 to recognize the tablets in the chartula, and a determining circuit 32 for comparing the result of image recognition with prescription information concerning each chartula to determine abnormality.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a tablet inspection system for automatically inspecting tablets enclosed in a medicine package.

[0002]

2. Description of the Related Art Conventionally, a tablet packing apparatus (1) as shown in FIG. 1 has been developed in order to automate tablet packing operations in hospitals and pharmacies. The tablet packaging device (1) includes a tablet supply unit (11) in which a plurality of types of tablets are stored according to type,
Various tablets (10) discharged from the tablet supply unit (11)
Automatically into the transparent packing paper (14) via the belt conveyors (12) (13), and the packing mechanism (17) seals the packing paper (14) for each person. Is. In addition, a printing mechanism (not shown) for automatically printing the patient name, usage, etc. on the surface of the packaging paper (14) is provided in the transport path of the packaging paper (14).

The packaging paper (14) thus obtained is cut and
The individual medicine packages (15) can be obtained by separating each person by the aligning device (16). The medicine package (15) has a printing part for the patient name, usage, etc., and is convenient because it contains a plurality of types of tablets (10) to be taken by the patient at one time. There is no risk of mistakes in taking the drug.

[0004]

However, in the tablet packing apparatus, when a predetermined number of tablets (10) are discharged from the tablet supply unit (11) onto the belt conveyor (12), there is a certain probability of an error in the number of tablets. Occurs. Further, there is a possibility that the number of sheets may be wrong when the sheets are put into the packaging paper (14). As a result, the medicine package (15) containing the number of tablets (10) different from the prescription instructions
May be created. With respect to this problem, conventionally, an inspector visually confirms the medicine package (15) discharged from the tablet packaging device (1) and compares it with the description on the prescription to confirm the presence or absence of abnormality. When an abnormality is found, the medicine package (15) is removed. This visual confirmation and elimination work is complicated and has been a cause of reducing the efficiency of the entire packaging work.

By the way, widely in the field of image recognition,
Various methods have been proposed for extracting the contour shape and the characteristic part of each object based on the images of a plurality of target objects (for example, Japanese Patent Publication 1-30183 [G06F15-70], Japanese Patent Publication 2-4948 [G06F15-7].
0], Japanese Patent Publication No. 3-57501 [G06F15-70], etc.). Therefore, by applying such image recognition technology to tablet inspection,
It is possible to automate the inspection.

However, in the inspection of the tablets in the medicine package, the tablets are often overlapped with each other, and the overlapped states are also various. Particularly when the overlap is large, it is difficult to identify an image of two tablets that overlap each other as two tablets by an inspection of a tablet using a conventional image recognition technique, and it is possible to obtain an accurate inspection result. Can not.

It is an object of the present invention to provide a tablet inspection system capable of accurately inspecting a tablet enclosed in a medicine package.

[0008]

A tablet inspection system according to the present invention comprises a conveying means for conveying a medicine package discharged from a tablet packing device to a predetermined inspection position, and a discharging means discharged from the tablet packing device and sent to the inspection position. A vibration adding mechanism for applying vibration to the medicine package to be inspected or the medicine package at the inspection position to remove the overlap of the tablets in the medicine package, an imaging means for photographing the medicine package at the inspection position, and an imaging Image recognition processing means for recognizing the tablets in the medicine package by processing the image obtained from the means, and comparing the result of the image recognition and the prescription information for each medicine package,
And a judging means for judging an abnormality.

In a specific configuration, the vibration adding mechanism applies vertical or horizontal vibration to the tablets in the medicine package, and for example, applies vibration to the conveyor belt on which the medicine package is placed. It has a vibration source or a guide mechanism for periodically bending the movement path of the conveyor belt on which the medicine package is placed.

Further, in a concrete configuration, the tablet inspection system includes a control means for stopping the vibration adding mechanism during the photographing of the medicine package by the image pickup means.

Further, the carrying means has an intermittent feeding function of stopping for a predetermined time when the medicine package reaches the inspection position. Furthermore, a sensor means for detecting the time when the medicine package reaches the detection position is provided, and the vibration adding mechanism and the intermittent feeding mechanism are controlled by the detection signal of the sensor means.

[0012]

In the tablet inspection system of the present invention, the medicine is discharged by the vibration adding mechanism while the medicine package discharged from the tablet packaging device is conveyed to the predetermined inspection position or when the medicine package reaches the inspection position. Vibration is applied to the tablets in the packet.
As a result, the tablets that have overlapped with each other are separated from each other.

After that, an image of the tablets in the medicine package is photographed at the inspection position, and a predetermined image recognition process is performed on the image to recognize the number and types of tablets in the medicine package. Here, the overlap of the tablets in the medicine package is slight, and the tablets slightly overlap each other, and are simply in contact with each other or separated from each other.
The image recognition process is performed accurately.

Then, by comparing the result of the image recognition for each medicine package with the prescription information, the abnormality is determined, and when the abnormality is determined, for example, a warning is issued for the medicine package.

In the concrete structure of the vibration adding mechanism, when vibration is applied to the conveyor belt on which the medicine package is placed in the vertical or horizontal direction by the vibration source, the vibration is transmitted to the tablets via the conveyor belt. Further, by bending the moving path of the conveyor belt carrying the medicine package up and down in the vertical plane or left and right in the horizontal plane, a vibration force is applied to the tablets in the medicine package along with the change of the traveling direction.

Further, by stopping the vibration adding mechanism during the photographing of the medicine package by the image pickup means, the medicine package during the photographing does not vibrate, which enables the normal photographing.

Further, if the conveying means is provided with an intermittent feeding mechanism,
Since the medicine package is stationary when it reaches the inspection position,
Normal shooting is possible. Furthermore, if the position of the medicine package being conveyed is detected by the sensor means, it is possible to control the timing of stopping the vibration adding mechanism and the intermittent feeding mechanism based on the detection signal.

[0018]

In the tablet inspection system according to the present invention, even if there is an overlap between tablets in the medicine package discharged from the tablet packaging apparatus, the overlap is released by the addition of vibration. The abnormality determination based on the image capturing becomes accurate.

[0019]

1 shows a tablet inspection system according to the present invention, in which an inspection device (3) is connected to an outlet of a tablet packaging device (1). The tablet packing device (1) is a control computer.
It is controlled by the control signal C from (2).

The specific structure of the tablet packaging device (1) is similar to that of the conventional device shown in FIG.
Ejects a plurality of tablets (10) onto the belt conveyor (12) according to the prescription information, and these tablets (10) are automatically loaded into the packaging paper (14) via the belt conveyor (13). It After that, the packaging mechanism (17) seals the packaging paper (14) for each person. Further, the packaging paper (14) is cut through the cutting and aligning device (16) into individual medicine packages (15), and these medicine packages (1
5) is placed in an aligned state on a semitransparent conveyor belt (44) extending to the subsequent step.

As shown in FIG. 1, the conveyor belt (44) constituting the transfer mechanism (4) is connected to the outlet of the cutting / aligning device (16), and the medicine package on the conveyor belt (44) is connected. (15)
Is a vibration adding process by the vibration adding mechanism (5), an optical sensor
After passing through the medicine package detecting step (6) and the image recognition processing step using the CCD camera (7), it is finally sent to the sorting device (9).

The transfer mechanism (4) is, for example, a motor as shown in the figure.
The driving roller (42) and the driven roller (43) connected to each other (41) drive the conveyor belt (44) in the horizontal direction. Here, the motor (41) is supplied with a drive current from the drive circuit (40).

The vibration adding mechanism (5) has, for example, a motor (51) as a power source, and vibrates the medicine package (15) on the conveyor belt (44) to cause the tablets in the medicine package (15) to interact with each other. It releases the overlapping of. This is a preliminary process for increasing the detection accuracy in detecting the number of tablets by the image recognition process described later.

As a concrete constitution of the vibration adding mechanism (5),
As shown in FIG. 1 schematically, it is possible to adopt the configuration shown in FIGS. 3 to 6 in addition to the configuration in which the back surface of the conveyor belt (44) is vibrated in the vertical direction. Vibration adding mechanism shown in FIG.
In (5), the conveyor belt (44) is sandwiched between the differently arranged roller rows (52), and the movement path thereof fluctuates in the vertical direction. With the movement of the conveyor belt (44), the conveyor belt (44) )
This is to give vertical vibration to the upper medicine package (15).

Further, the vibration adding mechanism (5) shown in FIG. 4 has a piezoelectric vibrator (53) arranged in contact with the back surface of the conveyor belt (44), and the piezoelectric vibrator (53) is arranged in a drive circuit. Driven by (54), the medicine package (1
It gives vertical vibration to 5).

Further, the vibration applying mechanism (5) shown in FIG. 5 employs a conveyor belt (45) having flexibility in a horizontal plane, and the conveyor belt (45) has bending guides arranged on both sides thereof. Guided by (55) and (56), it moves along a curved path in the horizontal plane. Along with this, the medicine package (1
Horizontal vibration is applied to 5). It should be noted that guide pieces (46) (46) for preventing the medicine package (15) from falling off are provided upright on both sides of the conveyor belt (45).

Further, the vibration adding mechanism (5) shown in FIG. 6 is provided with the guide pieces (46) (46) on both sides of the conveyor belt (45).
The piezoelectric vibrators (57) are arranged in contact with each other, and by driving the piezoelectric vibrators (57), the conveyor belt
(45) It gives a horizontal vibration to the medicine package (15).

In the medicine package detecting step, the optical sensor shown in FIG.
By (6), the detection mark preprinted on the medicine package (15) is optically detected, and the detection signal is supplied to the inspection device (3).

In the image recognition processing step, as shown in FIG. 7, the CCD camera (7) for photographing the medicine package (15) on the conveyor belt (44) is arranged downward. In addition, the conveyor belt (4
The illuminator (71) is located below the CCD camera (7) across the 4).
Are installed facing up, and a semi-transparent conveyor belt (4
Illuminate the medicine package (15) from the back side through 4).

The image signal obtained from the CCD camera (7) is sent to the inspection device (3) composed of a microcomputer and the like, and the number of tablets (10) enclosed in the medicine package (15) is counted. . A display (8) is connected to the inspection device (3), and with respect to a process of image processing described later and a medicine package (15) determined to be abnormal, the medicine package (15) is accompanied by a warning message to that effect. ) Image is displayed.

The inspection device (3) comprises an image recognition processing circuit (31), a judgment circuit (32) and a control circuit (33) as shown in FIG. The image recognition processing circuit (31) receives an image signal from the CCD camera (7) and performs image recognition processing on the image signal. The judgment circuit (32) judges the abnormality by comparing the number of tablets detected by the image recognition processing circuit (31) with the prescription information I obtained from the control computer (2). The abnormality determination signal NG is supplied to the control computer (2) and the control circuit (33). The control circuit (33) receives the determination result from the determination circuit (32), the detection signal from the optical sensor (6), the image recognition processing signal from the image recognition processing circuit (31), etc. 4), vibration adding mechanism (5), CC
It controls the operations of the D camera (7), the display (8), the sorting device (9) and the like.

The sorting device (9) sorts the medicine package (15) whose abnormality has been determined by the inspection device (3) and the medicine package (15) which has not been determined to be abnormal, and which has an abnormality. Eliminate (15),
Only the normal medicine package (15) is discharged to the subsequent stage.

FIG. 8 shows the operation of the above-mentioned inspection device (3). First, in step S1, a control computer
After downloading the prescription information from (2) to the determination circuit (32), in step S2, the transport mechanism (4) and the vibration adding mechanism are added.
Start driving (5). Next, in step S3, it is judged whether or not the detection mark printed in advance on the medicine package (15) is detected by the optical sensor (6).
Further, in step S4, it is determined whether the predetermined time has passed, that is, the medicine package (15) that has passed the position below the optical sensor (6) is CC
It is determined whether or not it has been conveyed to a position below the D camera (7).

If YES is determined in step S4, the transport mechanism (4) and the vibration adding mechanism are determined in step S5.
When (5) is stopped and the medicine package (15) at the lower position of the CCD camera (7), that is, the inspection position is stationary, the image recognition processing by the image recognition processing circuit (31) in step S6 is performed. The specific procedure of the image recognition process will be described later.

After the image recognition processing is completed, in step S7, the judgment circuit (32) compares the downloaded prescription information with the number of tablets obtained by the image recognition processing, and an error ( It is judged whether there is (mismatch) and YE
If S, the process proceeds to step S8. As a result, the abnormality determination signal NG is issued to the control circuit (33) and the control computer (2), and the control circuit (33) accordingly responds to the abnormality determination signal NG.
The display (8) displays a warning message indicating that an abnormality has been determined for the medicine package (15). Further, the control computer (2) issues a repacking instruction to the tablet packaging device (1) regarding the abnormal medicine package (15).

When NO is determined in step S7, it is determined in step S9 whether or not the inspection for all medicine packages is completed. When NO is determined, the process returns to step S2 and the inspection is repeated. .

FIG. 9 shows a series of image recognition processing procedures executed by the image recognition processing circuit (31). First, in step S11, the original image from the CCD camera (7) is captured. FIG. 15 shows an example of the original image captured by the image recognition processing circuit (31). In this example, nine tablets are enclosed in the medicine package, and eight of them are in contact with each other.
In the original image in FIG. 15, some noise is mixed around the image portion of the tablet.

Next, in step S12 of FIG. 9, the original image is binarized at a predetermined threshold level. As a result, each pixel forming the original image is represented by "0" or "1". FIG. 16 shows a binary image in which “0” is represented by white and “1” is represented by black. In the binary image, the noise included in the original image of FIG. 15 remains around the image portion of the tablet.

Next, in step S13 of FIG. 9, noise removal processing consisting of contraction and expansion of the binary image (for example, "basic technique of image processing", published by Technical Review Co., 53-5).
See page 6). The contraction / expansion process is a process of deleting or multiplying all the boundary pixels of the connected components in the image. In this embodiment, the contraction process is performed twice and the expansion process is performed twice. As a result, the printing on the surface of the medicine package (15) is deleted from the image and the isolated points in the image are removed. In the above example, the noise included in the original image of FIG. 16 is removed.

Subsequently, in step S14 of FIG. 9, labeling processing is performed on the binary image that has been subjected to the noise removal processing (for example, "basic technique of image processing" published by Technical Review Co., Ltd., 45-4).
See page 9). The labeling process is a process in which, when adjacent pixels are present in the image data, those image groups are regarded as one group, and a label is assigned. 1
There are 4-connected labeling that is regarded as a group, and 8-connected labeling that is regarded as 1 group when adjacent pixels exist in any of up, down, left, right, and diagonal directions. In tablet inspection, 4-connection labeling, which is excellent in terms of figure separation, is effective.

After that, in step S15, the chain coding process (for example, "Basic technique of image processing", published by Technical Review Co., Ltd., pages 76 to 78) is applied to the connected components in the image to which the label is assigned. The chain coding process is a process of tracing the contour or line segment of the connected component and giving a direction index (chain code) of “0” to “7” in that direction. With respect to each of the images assigned to, the contour (outer contour) formed by the outer shape of the tablet and the contour (inner contour) of a gap formed between a plurality of tablets in contact with each other are tracked. Then, the chain code is recorded for each label.

Next, for each connected component to which a label is assigned, the contact point of the image caused by the overlapping of tablets is detected based on the recorded chain code of the contour line.

FIG. 10 shows the procedure of contact detection. First, in step S21, contact candidate points are selected.
The contact candidate points have chain codes 0 → 6, 1 → 7, 2 → 0, 3 → 1, 4 → 2, 5 → 3, 6 as illustrated in FIG.
→ 4, 7 → 5 It is a point that changes rapidly, and 0 →
In the case of a combination of 6 and changing chain code, 0 → 7 →
1 or 2 between chain codes like 6 or 7 → 7 → 6
It is also a change point when two chain codes exist. Therefore, there are a total of 24 combinations of chain codes that should be changed.

Next, in step S22 of FIG. 10, at the contact point candidate, the angle of change in direction around the candidate point is calculated based on the contour lines of the front and rear 5 pixels. For example, FIG.
If the contact point is a contact point generated by the contact of two tablets as shown in 2 (a), the angle θ of the direction change is an acute angle.
If the contact point candidate is caused by noise generated on the contour line of one tablet as shown in (b), the angle θ is an obtuse angle. Therefore,
In step S23, the angle θ is compared with a predetermined threshold value (for example, 120 °), and if it is smaller than the threshold value, it is determined to be a true contact point in step S24. In S25, it is determined that the noise is on the contour line.

FIGS. 13 (a) and 13 (b) are enlarged views in the vicinity of the direction change points in FIGS. 12 (a) and 12 (b), and circles represent pixels forming a figure. Here, a chain code that changes from 5 to 3 is detected as a contact point candidate point, and the opening angle of the fifth pixel before and after the candidate point as the center is detected as the direction change angle. In this case, Fig. 13 (a)
The contact candidate point of is judged to be a true contact, and the contact candidate point of FIG. 13B is judged to be noise on the contour line and ignored.

Regarding the true contact point due to the overlapping of tablets,
As shown in FIG. 14, a pair of contacts (a,
b) and (c, d) are present. Therefore, step S in FIG.
At 17, for all labeled connected components,
Create such contact pairs. To create a contact pair,
For example, a method of discovering two contact points satisfying the following conditions 1 to 3 is adopted.

(Condition 1) The angle formed by the contact points of the two contact points to be judged as a pair (for example, for the contact points a and b in FIG. 14, the direction change vectors A and B at the contact points a and b). θ ₁ (0 ° ≤ θ ₁ ≤ 180 °) is
It is equal to or more than a predetermined threshold value (for example, 150 °).

(Condition 2) A contact vector of interest and a straight line from that contact to another contact (for example, contact a in FIG. 14,
Regarding b, the angle θ ₂ formed by the straight line E from the contact point a to the contact point b is not less than a predetermined threshold value (for example, 150 °).

(Condition 3) In the contact point pair satisfying the above conditions 1 and 2, the distance between the two points has a predetermined threshold value (in the case of a circular tablet, for example, its diameter is 0.3 to 1.0). It is less than or equal to 2 times) and the minimum.

In the case of FIG. 14, the contact points a and b and c and d are
Although it is determined to be a pair, the contact points a and c or a and d are not determined to be a pair.

After the contact pairs have been created for all the labels, in step S18 of FIG. 9, the two contacts determined to be a pair are connected to each other by a straight line to create a boundary line. The attached graphic element is separated into the connected components of the two tablet graphics at the boundary line. This processing is performed by replacing the pixel "1" on the boundary with "0".

FIG. 17 is a diagram showing an example of the binary image shown in FIG. 16, in which contact pairs are created at the contact portions of two tablets that are in contact with each other, and these contact pairs are interconnected to create a boundary line. FIG. 18 shows a state in which the binary image is divided into images of individual tablets at the boundary line.

Then, in step S19 of FIG. 9, relabeling is performed on the divided binary image of the image of each tablet. Then, the number of tablets is recognized by counting the number of labels in step S20. In the example of FIG. 18, the number of tablets is counted as 9.

According to the image recognition processing method described above, an accurate contact pair is created especially when the overlapping area of tablets is small, and as a result, the number of tablets can be recognized with extremely high accuracy.

FIG. 19 shows a processing method for efficiently separating the images of two tablets which overlap each other in the contraction processing of step S13 of FIG. That is, in addition to the deletion of the boundary pixels described above, when the binary state of 3 × 3 pixels shown in FIG. 19 has the illustrated pattern, the pixels at the positions with double circles are also set to “0” at the same time.

As another processing method, when the pattern of FIG. 19 is displayed, the pixel at the position marked with a square is "1".
It is also possible to set the pixel at the position marked with a double circle to "0" only in the case of. This method is intended to shrink the double circle pixels after determining whether or not the pattern of FIG. 19 is truly a pattern caused by the overlapping of tablets. By adding this determination, it is possible to prevent erroneous determination that the edge generated in the capsule lid portion of the capsule is the tablet overlap.

Further, in the contraction process, it is possible to add a process for removing isolated points. For example, if all 8 pixels in the vicinity of the pixel of interest are "1", the pixel of interest is set to "0". In addition to the above method, as shown in FIGS.
A method can be adopted in which when two pixels or eight neighboring pixels are all "1", a pixel surrounded by these pixels is "1", and when all are "0". The latter method basically aims to remove the isolated points of two or less connected pixels, but the isolated points as shown in FIG. 20C are judged to be meaningful pixels and are not removed.

The image recognition processing method shown in FIG. 9 described above is particularly effective when the overlapping area of the tablets is small.
Reference numeral 1 represents the procedure of the image recognition processing method which is effective when the tablet overlapping area is large. Therefore, when the detection of the contact point pair fails by the method of FIG. 9, if the images are separated by the method of FIG. 21, the reliable quantity recognition becomes possible.

First, in step S31, a CCD camera
A grayscale image of 512 pixels × 512 lines is taken in from (7). Also in this case, the illumination by the illuminator (71) is performed from below the medicine package (15) as shown in FIG. This is because when the illumination from above the medicine package (15) is adopted, the image information of the tablet is lost due to the reflection on the surface of the medicine package, or in the case of tablets with knurled surfaces or capsule tablets This is because the step portion of the capsule is detected as an edge.

Next, contrast conversion is performed in step S32 without binarizing the grayscale image. The density conversion formula used in this case is well known (see, for example, the magazine "Electronic Life", February 1989, pages 145-153). This facilitates extraction of the overlapping portion of the tablet.

Subsequently, in step S33, edge detection filter processing is performed. In the embodiment, as a result of examining various filter processing methods, as the best method, the first derivative so
A process using a bel filter (see, for example, the magazine "Electronic Life", December 1988, pages 148 to 155) is adopted. As a result, the outline of the tablet and the overlapping area between the tablets are emphasized. Further, in step S34, the contour-enhanced image is binarized at a predetermined threshold level to extract only the contour (edge).

Then, in step S35, thinning is performed. The thinning is a process of extracting a core line passing through the center of the widthwise edge image in the width direction.
Adopt ch algorithm (for example, Tamura “Various considerations on thinning method” IEICE Technical Report, 1975 PR
See L75-9966). As a result, the boundary line of the overlapping portion with the contour lines of the plurality of tablets is obtained as a line image.

Then, in step S36, the mustache and noise are removed from the line image. The conditions for removal are as follows. (1) 1-pixel isolated point (2) Lines that do not include branch points or intersections and have 20 or fewer connected pixels (3) The number of connected pixels between an end point and a branch point or intersection is 1
Lines less than 0

Further, in step S37, an effective thin line necessary for the difference extraction step (S38) with the next binary image is selected. This is a procedure for removing long whiskers and loop-like pseudo-rays that remain without being removed even in step S36 described above, and FIG. 22 shows a specific procedure thereof.

First, in step S41, an inner contour and an outer contour are extracted based on the line image. Then step S4
At 2, inflate these contour lines inward and outward,
Create a search area. For example, in the case of a line image in which two circular tablets are overlapped as shown in FIG. 23, a double line M, N is created surrounding a thin line L, and the area between them is used as a search area. The line E outside the search area is excluded from the target and is removed.

Then, in step S43 of FIG. 22, a branch point or an intersection existing in the fine line in the search area is detected. In the example of FIG. 23, a branch point exists at a position surrounded by a circle in the search area.

After that, in step S44 of FIG. 22, a line protruding from the search area at a distance of 20 pixels or more from the branch point or the intersection is determined to be an effective thin line.
In the example of FIG. 23, A and D in the figure are effective thin lines,
The other lines B and C are removed as invalid thin lines. The line D is a thin line that should be invalidated originally, but it does not pose a problem in recognizing the number of tablets as described later.

After selecting the effective thin line, step S in FIG.
At 38, the difference between the separately created binary original image and the image of the effective thin line is extracted, thereby separating the images of the two tablets that are connected to each other.

FIG. 24 shows the result of the difference extraction in the example of FIG. As shown, the images of the two tablets are separated by an effective thin line. The effect of the line D remaining without being removed appears in the binary image in FIG. 24, but this does not cause any problem in recognizing the quantity.
Then, labeling is performed in step S39 of FIG. 21, and the number of labels is counted in step S40.
It recognizes the quantity of tablets.

According to the above image recognition processing method, even when two tablets are largely overlapped, the illumination light leaks from the vicinity of the overlapped portion and the boundary line is surely detected by the edge detection. Separation is ensured and accurate quantity recognition is realized. Further, by combining the method of FIG. 19 and the method of FIG. 21, it becomes possible to inspect various medicine packages in the overlapping state of tablets.

In addition to the image recognition processing method of FIG. 19 or FIG. 21, the peripheral length and area of the binary image of the divided tablet, and other shape characteristic data are calculated to calculate the registered tablet in advance. By comparing with the data for each type, it is possible to further improve the accuracy of quantity recognition or determine the type of tablet.

According to the above-mentioned tablet inspection system, a plurality of tablets (10) in the medicine package (15) discharged from the tablet packing device (1).
Even when the tablets overlap with each other, by vibrating these tablets (10) by the vibration adding mechanism (5),
Overlapping tablets are pulled apart. As a result, the quantity recognition by the image recognition processing becomes more accurate.

Further, according to the above-mentioned tablet inspection system, the medicine package (15) whose number does not match the prescription information is automatically detected, and the medicine package (15) is eliminated by the sorting device (9). Therefore, conventional visual inspection and manual exclusion are not required. Of course, for the medicine package (15), a tablet packing device
Since a repacking order is issued to (1), regardless of whether there is an abnormality, all medicine packages (1
5) will be obtained.

The above description of the embodiments is for explaining the present invention, and should not be construed as limiting the invention described in the claims or reducing the scope. The configuration of each part of the present invention is not limited to the above-mentioned embodiment, and it goes without saying that various modifications can be made within the technical scope described in the claims.

For example, with respect to the medicine package in which two CCD cameras are installed with the vibration adding step interposed therebetween and the abnormality is judged by the image recognition processing using the first CCD camera, the vibration is applied in the next vibration adding step. It is possible to adopt a double-check method in which the image recognition processing using the second CCD camera is performed after the application, and the abnormality is checked again. In this case, the medicine package that has not been determined to be abnormal by the first image recognition process can be sent to the sorting device without being vibrated in the next vibration adding step. This double check can be performed by one camera if the image recognition processing step and the vibration adding step are configured to be performed at the same position.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a tablet inspection system according to the present invention.

FIG. 2 is a partially cutaway perspective view showing the configuration of the tablet packaging device.

FIG. 3 is a side view showing a configuration example of a vibration adding mechanism.

FIG. 4 is a side view showing another configuration example of the vibration adding mechanism.

FIG. 5 is a plan view showing still another configuration example of the vibration adding mechanism.

FIG. 6 is a plan view showing still another configuration example of the vibration adding mechanism.

FIG. 7 is a perspective view showing a device arrangement in an image recognition processing step.

FIG. 8 is a flowchart showing an inspection procedure of a medicine package.

FIG. 9 is a flowchart showing a procedure of image recognition processing.

FIG. 10 is a flowchart showing a procedure of contact detection.

FIG. 11 is a diagram illustrating a method of selecting contact point candidate points.

FIG. 12 is a diagram illustrating a method of discriminating between a true contact point and noise.

FIG. 13 is a diagram illustrating a method of calculating a direction change angle around a contact point.

FIG. 14 is a diagram illustrating a method of discriminating contact pairs.

FIG. 15 is a diagram showing original images of a plurality of tablets.

16 is a diagram showing a binary image in the example of FIG.

FIG. 17 is a diagram showing a state in which a contact pair is created in the above example.

FIG. 18 is a diagram showing a state in which images of a plurality of tablets are divided in the above example.

FIG. 19 is a diagram illustrating a method of increasing the efficiency of figure separation in the contraction processing.

FIG. 20 is a diagram illustrating a method for improving the efficiency of isolated point removal in the contraction processing.

FIG. 21 is a flowchart showing the procedure of another image recognition process.

FIG. 22 is a flowchart showing a procedure for selecting an effective thin line.

FIG. 23 is a diagram illustrating a method of removing a beard or a pseudo line generated at a branch point or an intersection.

FIG. 24 is a diagram showing a binary image obtained by extracting a difference from an effective thin line.

[Explanation of symbols] (1) Tablet packing device (2) Control computer (3) Inspection device (4) Conveying mechanism (5) Vibration adding mechanism (6) Optical sensor (7) CCD camera (8) Display (9) Sorting equipment (10) Tablets (15) Medicine packages

Claims (8)

[Claims] 1. A tablet packaging device for enclosing a plurality of tablets according to prescription information in a light-transmitting medicine package, and recognizing an image of the tablet in the medicine package discharged from the device. A tablet inspection system for inspecting tablets by means of: a conveying means for conveying a medicine package discharged from a tablet packaging device to a predetermined inspection position; and a medicine package to be discharged from the tablet packaging device and sent to the inspection position. Alternatively, a vibration adding mechanism for applying vibration to the medicine package at the inspection position to remove the overlap of the tablets in the medicine package, an imaging unit for photographing the medicine package at the inspection position, and an imaging unit An image recognition processing means for recognizing a tablet in a medicine package by processing an image and a judging means for judging an abnormality by comparing the result of image recognition for each medicine package with prescription information are characterized. And tablet inspection system. 2. The tablet inspection system according to claim 1, wherein the vibration adding mechanism applies vertical vibration to the tablet in the medicine package. 3. The tablet inspection system according to claim 1, wherein the vibration adding mechanism applies horizontal vibration to the tablets in the medicine package. 4. The vibration adding mechanism comprises a vibration generating source for vibrating a conveyor belt carrying a medicine package.
Alternatively, the tablet inspection system according to claim 3. 5. The tablet inspection system according to claim 2, wherein the vibration adding mechanism includes a guide mechanism that periodically bends the movement path of the conveyor belt on which the medicine package is placed. 6. The tablet inspection system according to claim 1, further comprising control means for stopping the vibration adding mechanism during imaging of the medicine package by the imaging means. 7. The tablet inspection system according to claim 1, wherein the conveying means has an intermittent feeding function of stopping for a certain period of time when the medicine package reaches the inspection position. 8. A sensor means for detecting the time when the medicine package reaches the detection position.
Tablet inspection system described in.