CN103063137B - A kind of medicine bottle measuring system based on machine vision and measuring method thereof - Google Patents

Abstract

The invention discloses a medicine bottle measurement system based on machine vision and a measurement method thereof, wherein the medicine bottle measurement system based on machine vision includes: an image acquisition device, a control device, a loading turntable, a laser scanner and an LED backlight source. Wherein, the image acquisition device is fixed directly above the loading turntable, the control device is connected to the image acquisition device, the LED backlight is arranged outside the loading turntable, and the laser scanner is fixed beside the LED backlight. The measurement method can accurately measure the characteristic parameters of the medicine bottle, so that the manipulator can accurately grasp, open the medicine bottle and prepare the medicine liquid, thereby reducing the workload of the dispensing personnel, and has good economic and social benefits. The outlook is better.

Description

A medicine bottle measuring system and its measuring method based on machine vision

technical field

The invention relates to the technical field of medicine measurement, in particular to a medicine bottle measurement system and a measurement method based on machine vision.

Background technique

Dimension measurement is an important application of digital image measurement. In industrial production, the image-based non-contact measurement method is used to replace the traditional contact measurement method, which not only greatly improves the work efficiency, but also makes it possible to unify the processing, detection and control processes. For image-based size measurement, its measurement accuracy is affected by the following three aspects: first, the object plane resolution of the camera system; second, the calibration and error correction accuracy of the camera system; third, the positioning accuracy of the edge of the measured object in the image.

At present, almost all the specifications, sizes and appearance defects of medicine bottles in China have been manually inspected, which is labor-intensive and the inspection effect is unstable. The glass product industry in some developed countries abroad has begun to use computer vision testing equipment. However, a complete set of non-contact detection equipment abroad is extremely expensive.

In view of this, the prior art needs to be improved and improved.

Contents of the invention

In view of the deficiencies in the prior art, the purpose of the present invention is to provide a medicine bottle measuring system and its measuring method based on machine vision. The invention aims to solve the problems of low efficiency, poor detection effect and the like caused by manually detecting the size of the traditional Chinese medicine bottle in the prior art.

Technical scheme of the present invention is as follows:

A medicine bottle measurement system based on machine vision, including:

The image acquisition device is used to collect the image information of the medicine bottle, and send the image information to the following control device;

The control device is used to process the image information, determine the shape characteristic parameters of the medicine bottle, and then send relevant instructions to the dispensing robot to perform corresponding actions;

Carousel for conveying medicine bottles;

A laser scanner for scanning the drug barcode of the medicine bottle to determine the type and position of the medicine bottle;

LED backlight is used to enhance the contrast of the background of the medicine bottle and reduce the influence of external light;

The image acquisition device is fixed directly above the loading turntable, the control device is connected to the image acquisition device, the LED backlight is arranged outside the loading turntable, and the laser scanner is fixed beside the LED backlight.

Preferably, in the medicine bottle measurement system based on machine vision, the image acquisition device is an industrial camera; the control device is an industrial control computer.

Preferably, in the medicine bottle measurement system based on machine vision, the processing of the image information by the control device specifically includes: image binarization, filtering, edge detection and size measurement.

Preferably, the medicine bottle measurement system based on machine vision, wherein the LED backlight is a red LED backlight for emitting parallel light, which is connected to a light source controller for adjusting the light intensity of the LED backlight; The light source controller is fixed on the base.

Preferably, the machine vision-based medicine bottle measurement system, wherein, the control device is sequentially provided with:

An image processing module, configured to process the image information, including: binarization processing, median filtering, edge detection and image marking;

The size measurement module is used for determining the size of the medicine bottle according to the mark of the image processing module.

A kind of measuring method of above-mentioned medicine bottle measuring system based on machine vision, wherein, comprise the following steps:

S1. The loading carousel starts to rotate under the drive of the motor, and transfers the medicine bottle to the scanning station;

S2. The laser scanner scans the drug barcode on the drug bottle at the scanning station, determines the type of the drug bottle, and accurately positions it, and sends the corresponding data to the control device;

S3. The loading carousel transmits the medicine bottle to the collection station, and the image acquisition device collects the image information of the medicine bottle, and sends the image information to the following control device;

S4. The control device processes the image information to determine the shape characteristic parameters of the medicine bottle, and then sends relevant instructions to the dispensing robot to perform corresponding actions.

Preferably, in the measurement method of the medicine bottle measurement system based on machine vision, the step S4 further includes: collecting standard piece data information by an image acquisition device, and calibrating the medicine bottle measurement system.

Preferably, in the measurement method of the medicine bottle measurement system based on machine vision, the step S3 further includes: the LED backlight emits parallel light to enhance the contrast of the background of the medicine bottle on the collecting station.

Preferably, the measurement method of the medicine bottle measurement system based on machine vision, wherein, in the step S4, the control device processes the image information, and determines that the shape characteristic parameters of the medicine bottle specifically include:

S41. Perform binarization processing on the image information;

S42. Perform filtering processing on the binarized image to remove interference on the image;

S43. After the filtering is completed, edge detection is performed on the image to obtain the edge profile of the medicine bottle;

S44. Mark the left and right edge lines of the medicine bottle, calculate the number of pixels between the left and right edge lines, and then calculate the size of the medicine bottle in the left and right direction;

S45. Mark the upper and lower edge lines of the medicine bottle, calculate the number of pixels between the upper and lower edge lines, and then calculate the height of the medicine bottle.

Preferably, in the measurement method of the medicine bottle measurement system based on machine vision, the step S2 further includes: when the item scanned by the laser scanner is not a medicine bottle, stop the measurement.

Beneficial effect:

The medicine bottle measurement system and its measurement method based on machine vision of the present application can accurately measure the characteristic parameters of the medicine bottle, so that the manipulator can accurately grasp, open the medicine bottle and prepare the medicine liquid, thereby reducing the workload of the dispensing personnel. It has very good economic and social benefits, and has a good market promotion prospect.

Description of drawings

FIG. 1 is a schematic diagram of an embodiment of a machine vision-based medicine bottle measuring system of the present application.

Fig. 2 is a flow chart of the measurement method of the medicine bottle measurement system based on machine vision of the present application.

detailed description

The present invention provides a medicine bottle measuring system and its measuring method based on machine vision. In order to make the purpose, technical solution and effect of the present invention clearer and clearer, the present invention will be further described in detail below. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

Please refer to FIG. 1 , which is a schematic diagram of an embodiment of a machine vision-based medicine bottle measuring system of the present application. As shown in the figure, the medicine bottle measurement system based on machine vision includes: an image acquisition device 100 , a control device 200 , a loading carousel 300 , a laser scanner (not shown in the figure) and an LED backlight 500 . Wherein, the image acquisition device 100 is fixed directly above the loading carousel 300, the control device 200 is connected to the image acquisition device 100 (the connection is indicated by a schematic line in the figure, and it can also be connected wirelessly), and the LED backlight 500 is set on On the outside of the loading carousel 300 , the laser scanner 400 is fixed beside the LED backlight 500 .

Specifically, the image acquisition device 100 is used to collect image information of the medicine bottle 10, and send the image information to the control device 200. In this embodiment, the image acquisition device 100 is an industrial camera; The control device 200 is an industrial control computer, and the industrial camera is connected to the industrial control computer through a USB cable. The control device 200 is used to process the image information, determine the shape characteristic parameters of the medicine bottle, and then send relevant instructions to the dispensing robot to perform corresponding actions. The specific image processing process will be described in detail later. The loading carousel 300 is used to convey the medicine bottle (rotated by the motor 310), and it also includes a medicine clamp, the medicine clamp is mainly used to clamp the placed medicine, and the medicine clamp unit has an adaptive adjustment function, The size can be automatically adjusted to hold medicine bottles of different sizes, and the loading carousel 300 is mainly used to transport medicines or medicine bottles to a designated position for the next step of operation. The laser scanner 400 is used to scan the drug barcode of the medicine bottle to determine the type and position of the medicine bottle. Specifically, the laser scanner is fixed next to the LED backlight 500 and is mainly used to scan the loading carousel 300 to transmit The drug barcode of the medicine bottle, in order to determine whether the item is a medicine or a medicine bottle, if it is a medicine bottle, it can also be distinguished whether it is an ampoule or a vial, and then its location can be precise to provide data for subsequent operations. The LED backlight 500 is used to enhance the contrast of the background of the medicine bottle and reduce the influence of external light. It is arranged on the fixed bracket 600. In this embodiment, the LED backlight 500 uses a red LED backlight (Of course, other colors are also available), its penetrating ability is very strong, and the light it emits is parallel light, the imaging effect is very good, and the obtained image is clearer. In this embodiment, the LED backlight 500 is connected to a light source controller, and the light source controller is fixed on the base, and is mainly used to adjust the light intensity of the LED backlight so that it emits different lights in different occasions and environments. Brightness of parallel light.

Among them, the processing of the image information by the control device 200 is the key of the present invention. In this embodiment, the control device is sequentially provided with: an image processing module and a size measurement module, and the image processing module is used to process the image information, which includes: binarization processing, median filtering, Edge detection and image marking; the size measuring module is used to determine the size of the medicine bottle according to the marking of the image processing module.

In this embodiment, when the medicine bottle scanned by the laser scanner is an ampoule, it is placed at a specific position between the industrial camera and the LED backlight through the loading carousel, and then the industrial camera is triggered to capture the image, and further Image matching, to confirm whether it is an ampoule, and then perform image binarization, filtering, edge extraction and size measurement, and then determine its diameter, bottleneck position, etc., where the left edge is first marked with a yellow line when measuring the size Mark it, then mark its right edge with a green line, and then calculate the diameter of the medicine bottle. For the measurement of the bottleneck position, the bottleneck position measured by the left edge is first obtained through a series of algorithms, and then the bottleneck position measured by the right edge is obtained, and finally the final size of the medicine bottle is determined according to the measurement results of the left and right bottlenecks. Bottleneck location.

When the scanned medicine bottle is a vial, it is placed at a specific position between the industrial camera and the LED backlight through the loading carousel, and then the industrial camera is triggered to capture an image, and further image matching is performed to confirm whether it is a vial , and then perform image binarization, filtering, edge extraction and size measurement, and then determine its bottleneck position, where the left edge is first marked with a yellow line when the size is measured, and then its right edge is marked with a green line Mark it out, and then calculate the diameter of the medicine bottle. For the measurement of the bottleneck position, the bottleneck position measured by the left edge is first obtained through a series of algorithms, and then the bottleneck position measured by the right edge is obtained, and finally the final size of the medicine bottle is determined according to the measurement results of the left and right bottlenecks. Bottleneck location.

The present invention also provides a measurement method of the above-mentioned medicine bottle measurement system based on machine vision, as shown in Figure 2, comprising the following steps:

S1. The loading carousel starts to rotate under the drive of the motor, and transfers the medicine bottle to the scanning station;

S2. The laser scanner scans the drug barcode on the drug bottle at the scanning station, determines the type of the drug bottle, and accurately positions it, and sends the corresponding data to the control device;

S3. The loading carousel transmits the medicine bottle to the collection station, and the image acquisition device collects the image information of the medicine bottle, and sends the image information to the following control device;

S4. The control device processes the image information to determine the shape characteristic parameters of the medicine bottle, and then sends relevant instructions to the dispensing robot to perform corresponding actions.

Further, the step S3 also includes: the LED backlight emits parallel light to enhance the contrast of the background of the medicine bottle on the collecting station. In addition, the step S4 also includes: collecting standard part data information by an image collection device, and calibrating the medicine bottle measurement system.

In this embodiment, the calibration process is as follows: take a standard calibration medicine bottle, such as height H=30 mm, diameter W=30 mm, and place it on the loading turntable; the turntable rotates once, and the laser scanner scans the medicine bottle on the turntable Scanning the barcode, the computer recognizes the scanned data, confirms it as a standard medicine bottle, and starts the system calibration program, wherein the calibration program is consistent with the image information processing program in step S4, which will be described in detail below.

Furthermore, in the step S4, the control device processes the image information, and determines that the shape characteristic parameters of the medicine bottle specifically include:

S41. Perform binarization processing on the image information;

S42. Perform filtering processing on the binarized image to remove interference on the image;

S43. After the filtering is completed, edge detection is performed on the image to obtain the edge profile of the medicine bottle;

S44. Mark the left and right edge lines of the medicine bottle, calculate the number of pixels between the left and right edge lines, and then calculate the size of the medicine bottle in the left and right directions;

S45. Mark the upper and lower edge lines of the medicine bottle, calculate the number of pixels between the upper and lower edge lines, and then calculate the height of the medicine bottle.

The above steps are described in detail below:

First of all, the computer performs binarization processing on the captured image. There are many binarization methods. In this system, the gray threshold segmentation method is adopted. The algorithm is simple, the operation speed is fast, and the efficiency is high; the expression is: ;

Among them, T is the threshold value, g(i,j)=1 represents the image element of the target object, and g(i,j)=0 represents the image element of the background.

After binarization, filter the image; remove the interference that exists in the process of forming, collecting and receiving the medicine bottle image, such as complex background, light source illumination and signal transmission, etc.; this system uses median filtering to process the image , after many experiments, the processing effect is good; after the filtering is completed, edge detection is performed on the image to obtain the edge profile of the medicine bottle.

After obtaining the edge contour of the medicine bottle, mark the left and right edge lines of the medicine bottle, and calculate the number of pixels between the left and right edge lines, such as N=30; then it can be known that in the left and right direction, the pixel The relationship between N and the actual size of the vial is as follows:

W=30/30*N; where, W is the diameter of the medicine bottle/mm, and N is the pixel/unit.

In the same way, the relationship between the height of the medicine bottle and the pixel can be calculated in the up and down directions:

H=30/30*N; Among them, H is the height of the medicine bottle/mm, and N is the pixel/unit.

Further, the step S2 also includes: when the item scanned by the laser scanner is not a medicine bottle (for example: medicine or other products), stop the measurement.

In summary, the machine vision-based medicine bottle measurement system and its measurement method of the present invention, wherein, the machine vision-based medicine bottle measurement system includes: an image acquisition device, a control device, a loading carousel, a laser scanner and LED backlight. Wherein, the image acquisition device is fixed directly above the loading turntable, the control device is connected to the image acquisition device, the LED backlight is arranged outside the loading turntable, and the laser scanner is fixed beside the LED backlight. The measurement method can accurately measure the characteristic parameters of the medicine bottle, so that the manipulator can accurately grasp, open the medicine bottle and prepare the medicine liquid, thereby reducing the workload of the dispensing personnel, and has good economic and social benefits. The outlook is better.

It should be understood that the application of the present invention is not limited to the above examples, and those skilled in the art can make improvements or transformations according to the above descriptions, and all these improvements and transformations should belong to the protection scope of the appended claims of the present invention.

Claims (2)

1., based on a medicine bottle measuring system for machine vision, it is characterized in that, comprising:

Image collecting device, for gathering the image information of medicine bottle, and is sent in following control device by described image information;

Control device, for processing described image information, judges the parameters for shape characteristic of medicine bottle, then sends relevant instruction and carries out corresponding action to the robot that makes up a prescription;

Loading rotating disk, for transmitting medicine bottle;

Laser scanner, for scanning the medicine bar code of medicine bottle, determines type and the position of described medicine bottle;

LED backlight, for strengthening the contrast of the background of medicine bottle, reduces the impact of ambient light;

Described image collecting device is fixed on directly over loading rotating disk, and control device connects image collecting device, and LED backlight is arranged on the outside of loading rotating disk, and laser scanner is fixed on the side of LED backlight;

Described loading rotating disk also comprises a medicine fixture, and described medicine fixture is used for clamping the medicine placed, and this medicine fixture has Automatic adjusument function;

Described control device is disposed with: image processing module and dimensional measurement module, described image processing module is used for processing described image information, and it comprises: binary conversion treatment, medium filtering, rim detection and image tagged; Described dimensional measurement module is used for the mark according to image processing module, determines the size of medicine bottle;

First with yellow line, medicine bottle left hand edge is marked when dimensional measurement, then with green line, medicine bottle right hand edge is marked, then by calculating the diameter of medicine bottle;

For the measurement of bottleneck position, first drawn the bottleneck position surveyed by left hand edge by series of algorithms, then draw the bottleneck position surveyed by right hand edge, finally again according to the measurement result of left and right bottleneck, determine the final bottleneck position of this medicine bottle;

Described image binaryzation process adopts gray level threshold segmentation method, and described gray level threshold segmentation method expression is:

; Wherein, T is threshold value, and g (i, j)=1 represents the pictorial element of target object, and g (i, j)=0 represents the pictorial element of background;

Described LED backlight is the red LED backlight for sending directional light, and it connects one for regulating the light source controller of LED backlight light intensity; Described light source controller is fixed on base;

Described image collector is set to industrial camera; Described control device is industrial control computer;

When laser scanner scans to medicine bottle be ampoule bottle, then place it in the ad-hoc location between industrial camera and LED backlight by loading rotating disk, then trigger industrial camera to carry out catching image, carry out images match further, confirm whether it is ampoule bottle, and then carry out image binaryzation, filtering, edge extracting and dimensional measurement, then determine its diameter, bottleneck position;

When the medicine bottle scanned is cillin bottle, then place it in the ad-hoc location between industrial camera and LED backlight by loading rotating disk, then trigger industrial camera to carry out catching image, carry out images match further, confirm whether it is cillin bottle, and then carry out image binaryzation, filtering, edge extracting and dimensional measurement, then determine its bottleneck position.

2. a measuring method for the medicine bottle measuring system based on machine vision according to claim 1, is characterized in that, comprise the following steps:

S1, loading rotating disk start to rotate under the driving of motor, medicine bottle are sent to scanning station;

S2, laser scanner scan the medicine bar code on the medicine bottle of scanning station, determine bottle-type, and accurate its position, corresponding data is sent to control device;

Medicine bottle is sent to collection station by S3, loading rotating disk, the image information of image acquisition device medicine bottle, and described image information is sent to control device;

S4, control device process described image information, judge the parameters for shape characteristic of medicine bottle, then send relevant instruction and carry out corresponding action to the robot that makes up a prescription;

In described step S4, control device processes described image information, judges that the parameters for shape characteristic of medicine bottle specifically comprises:

S41, binary conversion treatment is carried out to described image information;

S42, filtering process is carried out to the image after binary conversion treatment, remove the interference on image;

After S43, filtering complete, rim detection is carried out to image, obtain the edge contour of medicine bottle;

S44, the left and right edge line of medicine bottle to be marked, calculate the number of pixels between left and right edge line, then calculate the size on medicine bottle left and right directions;

S45, the upper and lower edge line of medicine bottle to be marked, calculate the number of pixels between upper and lower edge line, then calculate the height of medicine bottle;

Described image binaryzation process adopts gray level threshold segmentation method, and described gray level threshold segmentation method expression is:

; Wherein, T is threshold value, and g (i, j)=1 represents the pictorial element of target object, and g (i, j)=0 represents the pictorial element of background;

Also comprise in described step S3: LED backlight sends directional light, strengthen medicine bottle in the contrast gathering background on station;

Also comprise in described step S4: by image acquisition device standard component data message, described medicine bottle measuring system is demarcated;

Also comprise in described step S2: when the article that laser scanner scans goes out are not medicine bottles, stop measuring;

Described image collector is set to industrial camera; Described control device is industrial control computer;

When laser scanner scans to medicine bottle be ampoule bottle, then place it in the ad-hoc location between industrial camera and LED backlight by loading rotating disk, then trigger industrial camera to carry out catching image, carry out images match further, confirm whether it is ampoule bottle, and then carry out image binaryzation, filtering, edge extracting and dimensional measurement, then determine its diameter, bottleneck position;

When the medicine bottle scanned is cillin bottle, then place it in the ad-hoc location between industrial camera and LED backlight by loading rotating disk, then trigger industrial camera to carry out catching image, carry out images match further, confirm whether it is cillin bottle, and then carry out image binaryzation, filtering, edge extracting and dimensional measurement, then determine its bottleneck position.