CN108875845B - Medicine sorting device - Google Patents
Medicine sorting device Download PDFInfo
- Publication number
- CN108875845B CN108875845B CN201810834149.8A CN201810834149A CN108875845B CN 108875845 B CN108875845 B CN 108875845B CN 201810834149 A CN201810834149 A CN 201810834149A CN 108875845 B CN108875845 B CN 108875845B
- Authority
- CN
- China
- Prior art keywords
- medicine
- classification
- unit
- medicines
- conveying belt
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 239000003814 drug Substances 0.000 title claims abstract description 138
- 229940079593 drug Drugs 0.000 title claims abstract description 46
- 238000012545 processing Methods 0.000 claims abstract description 41
- 238000003860 storage Methods 0.000 claims abstract description 38
- 230000005540 biological transmission Effects 0.000 claims abstract description 23
- 238000000034 method Methods 0.000 claims description 15
- 238000013527 convolutional neural network Methods 0.000 claims description 10
- 238000003708 edge detection Methods 0.000 claims description 8
- 238000004519 manufacturing process Methods 0.000 claims description 4
- 239000007921 spray Substances 0.000 claims description 3
- 238000013145 classification model Methods 0.000 claims description 2
- 238000001514 detection method Methods 0.000 abstract description 9
- 241000282414 Homo sapiens Species 0.000 abstract description 6
- 238000007689 inspection Methods 0.000 description 7
- 238000011161 development Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 5
- 238000012549 training Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000012546 transfer Methods 0.000 description 3
- 230000002950 deficient Effects 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 239000000047 product Substances 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000002347 injection Methods 0.000 description 1
- 239000007924 injection Substances 0.000 description 1
- 210000001503 joint Anatomy 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000004630 mental health Effects 0.000 description 1
- 239000012466 permeate Substances 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000003908 quality control method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 239000013598 vector Substances 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Medical Treatment And Welfare Office Work (AREA)
- Sorting Of Articles (AREA)
Abstract
The invention discloses a medicine classification device, which comprises a transmission unit, a camera, a processing unit, a classification execution unit and a storage unit, wherein the processing unit is respectively connected with the camera and the classification execution unit in a signal manner; the transmission unit is used for arranging medicines in the pipeline for transmission, the camera acquires medicine images, the processing unit performs identification and classification according to the medicine images, and the classification execution unit is controlled to push the medicines to the storage unit correspondingly. According to the medicine classifying device, the transmission unit is arranged to enable medicines to be transmitted on the pipeline of the transmission unit, the camera is used for acquiring full-feature images of each medicine in real time, the processing unit is used for judging which classification the medicines belong to through identifying and analyzing the images, and controlling the classifying execution unit to push the medicines to the storage units corresponding to the classifications, so that full automation of medicine detection classification is realized, human errors caused by manual detection classification are avoided, labor cost is saved, and medicine classifying efficiency is improved.
Description
Technical Field
The invention relates to the field of image recognition processing, in particular to a medicine classification device.
Background
Along with the development of the national innovation, market economy is rapidly developed, meanwhile, the phenomenon of counterfeiting inferior products is frequent, and the inferior products gradually permeate into the aspects of human life along with the development of the market. It destroys the normal market order and investment environment of society, infringes the legal rights and interests of legal operators and consumers, and brings great harm to the social development. The medicine is an indispensable part of human life, and the qualified medicine plays an important role in guaranteeing the healthy development of human beings. The quality control of medicine detection classification is a bridge connecting medicine development, production and market sales and use. It can be seen that the quality of the medicine has a direct influence on the physical and mental health and life safety of common people.
Currently, in a conventional medicine sorting method, in order to sort medicines by checking whether the medicines are broken or incomplete, it is necessary to perform sorting by a worker who designates a detection mechanism. The classification inspection technology is behind, almost manual operation is carried out, the classification period is longer, and the input labor cost is high. In addition, due to uneven quality of inspection classification staff, the working efficiency of medicine inspection classification is restricted to a great extent.
With the development and progress of image recognition processing technology, a medicine inspection and classification method based on image recognition is already an advanced and convenient medicine detection technology. The method adopts a convolutional neural network algorithm, and realizes higher recognition accuracy by establishing a medicine recognition model based on deep learning, extracting important features in medicine images in the deep learning model through supervised training of a convolutional layer and a downsampling layer of extracted features and extracting important features from the medicine images through a classifier. Based on this, how to use image recognition processing technology to realize automatic inspection classification of medicines based on pipelines is a problem that needs to be solved at present.
Disclosure of Invention
In order to overcome the defects in the prior art, the invention aims to provide a medicine classification device which can automatically identify and classify medicines.
The invention adopts the following technical scheme:
the medicine classifying device comprises a transmission unit, a camera, a processing unit, a classifying executing unit and a storage unit, wherein the processing unit is respectively connected with the camera and the classifying executing unit in a signal mode; the transmission unit is used for placing the medicines on a production line for transmission, the camera acquires medicine images, the processing unit performs identification and classification according to the medicine images, and controls the classification execution unit to correspondingly push the medicines to the storage unit;
the processing unit performs the following method: the method comprises the steps of obtaining a shot medicine full-feature image, obtaining an image gray level image, positioning the outer frame position of the medicine through edge detection and outer frame processing, obtaining medicine features through a convolutional neural network algorithm, and comparing according to a pre-established medicine classification model to determine the classification of the medicine.
Further, the transmission unit comprises a bed body and a conveying belt, wherein the conveying belt is arranged at the top of the bed body, and the conveying belt drives the conveying belt to rotate through a rotating shaft.
Further, the transmission unit further comprises an input port, and the input port is arranged at one end of the conveying belt.
Further, the conveying unit further comprises a baffle plate, the baffle plate is arranged at one end of the top of the conveying belt, which is abutted to the input port, and the baffle plate is used for adjusting the conveying route of the medicines.
Further, the classification execution unit is a plurality of air guns, the air guns are arranged at one side of the conveying line of the conveying belt at intervals, and gun openings of the air guns face the conveying belt.
Further, the air gun is also provided with an air gun fixing seat.
Further, the storage unit comprises a plurality of first storage boxes, the first storage boxes are arranged on the other side of the conveying line of the conveying belt at intervals, and the first storage boxes are arranged corresponding to the air guns.
Further, the storage unit further comprises a second storage box, and the second storage box is correspondingly arranged at the tail end of the conveying line of the conveying belt.
Compared with the prior art, the invention has the beneficial effects that:
according to the medicine classifying device, the transmission unit is arranged to enable medicines to be transmitted on the pipeline of the transmission unit, the camera is used for acquiring full-feature images of each medicine in real time, the processing unit is used for judging which classification the medicines belong to through identifying and analyzing the images, and controlling the classifying execution unit to push the medicines to the storage units corresponding to the classifications, so that full automation of medicine detection classification is realized, human errors caused by manual inspection classification are avoided, labor cost is saved, and medicine classifying efficiency is improved.
Drawings
Fig. 1 is a schematic diagram of a medicine sorting device.
In the figure: 11. a bed body; 12. a conveyor belt; 13. a rotating shaft; 14. an input port; 15. a baffle; 2. a camera; 3. a processing unit; 41. an air gun; 42. an air gun base; 51. a first storage case; 52. and a second storage box.
Detailed Description
The present invention will be further described with reference to the accompanying drawings and detailed description, wherein it is to be understood that, on the premise of no conflict, the following embodiments or technical features may be arbitrarily combined to form new embodiments.
The medicine classifying device shown in fig. 1 comprises a transmission unit, a camera 2, a processing unit 3, a classifying executing unit and a storage unit, wherein the processing unit 3 is respectively connected with the camera 2 and the classifying executing unit in a signal manner; the transmission unit is used for placing medicines on a production line for transmission, the camera 2 acquires medicine images, the processing unit 3 performs identification and classification according to the medicine images, and the classification execution unit is controlled to push the medicines to the storage unit correspondingly. It should be noted that, in the classification process of each medicine, the classification execution unit should execute the classification after the camera 2 shoots and the processing unit 3 analyzes and processes the medicine.
Specifically, the conveying unit comprises a bed body 11 and a conveying belt 12, wherein the conveying belt 12 is arranged at the top of the bed body 11, and the conveying belt 12 is driven to rotate through a rotating shaft 13. The bed body 11 is mainly used for supporting the rotating shaft 13 and the conveying belt 12, and provides a stable workbench for drug conveying. The conveyer belt 12 provides the conveying line of medicine, and the axis of rotation 13 circular telegram rotates the back, drives conveyer belt 12 and rotates, and the medicine is put into from the one end of conveyer belt 12, moves along the conveying line of conveyer belt 12, and the full characteristic image of medicine is shot through camera 2 earlier in the removal in-process, and processing unit 3 analysis processing back controls categorised execution unit and carries out corresponding categorised operation. The transfer unit further includes an input port 14, the input port 14 being disposed at one end of the conveyor belt 12. The input port 14 is parallel to and above the conveyor belt 12. It is mainly used for delivering medicines to be sorted onto the conveyor belt 12. The camera 2 can be specifically arranged at one end, close to the input port 14, of the edge of the bed 11, and is fixed through screws, and a shooting area of the camera is locked in an area where the input port 14 is in butt joint with the conveying belt 12, so that the processing unit 3 can conveniently acquire a medicine characteristic image at the first time. It should be noted that the shape design of the input port 14 is required to be such that the drug rolls during delivery for the camera 2 to capture the full extrinsic feature. In addition, in order that the transfer unit further includes a baffle 15, the baffle 15 is disposed at one end of the top of the conveyor belt 12 where the input port 14 is abutted, and the baffle 15 is used for adjusting the transfer route of the medicine. The baffle 15 has flat face, and baffle 15 is vertical to be put, and the transmission route that the face formed is certain angle with the transmission route of output port medicine for the medicine is along the face transmission when passing through baffle 15. Due to the adjusting action of the baffle 15, the medicines can be transported in a straight line after passing through the baffle 15. And errors of shifting the conveying route of the medicines when the follow-up classifying and executing unit pushes the medicines are avoided.
The sorting execution unit is a plurality of air guns 41, the air guns 41 are arranged at intervals on one side of the conveying line of the conveying belt 12, and the muzzle of the air guns 41 faces the conveying belt 12. The air gun 41 is also provided with an air gun mount 42 to fix the ejection direction of the air gun 41. The storage unit includes a plurality of first storage boxes 51, the first storage boxes 51 are disposed at intervals on the other side of the conveying line of the conveying belt 12, and the first storage boxes 51 are disposed corresponding to the air gun 41. The storage unit further includes a second storage box 52, where the second storage box 52 is correspondingly disposed at the end of the conveying line of the conveying belt 12. It should be noted that, when a medicine category needs to be bound to an air gun 41, the air gun 41 sprays the medicine to blow the medicine to the corresponding first storage box 51 when the medicine category passes through the air gun 41. Based on the horizontal distance between the medicine and the position of the air gun 41, the processing unit 3 needs to perform initialization setting for binding the medicine type and the air gun 41. The processing unit 3 binds the first type of medicine to the distance AB based on the horizontal vertical distance AB, with the current position of the medicine falling from the input port 14 to the conveyor belt 12 as the start point a, the medicine being conveyed along the conveyor belt 12 from one end of the conveyor line to the other end, and the current position of the medicine overlapping the position of the first air gun 41 as the end point B. Then, the current position of the medicine overlapping with the position of the second air gun 41 is taken as an end point C; based on the horizontal-vertical distance AC, the processing unit 3 binds a second category of drugs to the distance AC. And so on until all drug classes have been bound to the air gun 41. Then, based on the determination of the type of the medicine and the binding relation between the type of the medicine and the air gun 41, the processing unit 3 starts from the current position of the medicine falling from the input port 14 to the conveyor belt 12, and when the horizontal and vertical distance between the medicine and the position of one air gun 41 accords with the binding relation, the processing unit 3 energizes the air gun 41 through the electric wire to spray the medicine, and pushes the medicine into the first storage box 51 of the type. The number of air guns 41 is the same as that of the first receiving boxes 51, and the air guns 41 are correspondingly arranged on two sides of the conveying belt 12. After receiving the current, the air gun 41 energizes its own solenoid valve to complete the injection and push the medicine into the corresponding type of storage box. In view of the existence of medicines that do not belong to all of the groups to be classified, a medicine classifying device of the present embodiment is further provided with a second housing box 52 that is mainly used for storing non-classified medicines or defective medicines that are broken and incomplete, which are provided at the end of the conveying line. In addition, it should be noted that, when the processing unit 3 in this embodiment performs recognition processing on the acquired drug feature image, a convolutional neural network algorithm is adopted, an image gray level map is first obtained, and after edge detection and outer frame processing, a recognition result is calculated and output through the convolutional neural network algorithm, and a comparison is performed according to a pre-established drug classification database to determine the classification to which the drug belongs. The specific method comprises the following steps:
1. obtaining a gray scale map
In the preprocessing stage, since the color image contains a large amount of information, which seriously affects the image processing speed, we first perform the graying process on the image. The color image may also be called an RGB image, in which each pixel has 3 components R, G, B, and in this embodiment, a weighted average method is used, i.e., 3 components of each pixel are assigned different weights, and then their weighted average is calculated, where the formula is as follows
Wherein: the weight values of WR, WG and WB are R, G, B respectively, and through multiple tests and verification, in the picture library of the existing medicine package in the hospital pharmacy, we find that in the complex image background, when wr=0.291, wg=0.592 and wb=0.117, the most reasonable gray level map is obtained.
2. Edge detection and frame processing
Because the duty ratio and the position of the medicine in the image are not fixed in the conveying process, in order to further improve the subsequent processing speed and enable the image recognition to be more accurate, the outer frame position of the medicine needs to be positioned first, the subsequent processing can be concentrated on the needed part, a large amount of background image interference is eliminated, the recognition operation burden is further reduced, and the detection speed is improved. Firstly, after the graying treatment, we carry out edge detection treatment on the obtained graph, so that the character area is more highlighted. This embodiment uses laplacian-gaussian (LOG) based edge detection, which uses the principle of maximum first derivative of amplitude when the second derivative of the signal is equal to zero, to determine the edges of the image by finding the zero of the second derivative of the image. When edge detection is carried out by utilizing the Laplace Gaussian operator, firstly, convolution operation is carried out on the image and the Gaussian function, and a smooth image processed by the Gaussian function is obtained.
h(x，y)＝f(x，y)×G(x，y)
Sigma is a gaussian function variance proportional to smoothness; and then carrying out Laplace transformation on the smoothed image h (x, y) to obtain a Laplace Gaussian operator edge detection image.
3. Building a model
Setting initial parameters of each network layer and classifier of CNN (convolutional neural network), and inputting the preprocessed image into CNN to obtain required medicine image characteristics. The model is composed of seven layers, layer0 is a data layer, and is image data preprocessed by us; layer1 is a new convolution layer conv1 defined, and the input is the output of layer 0; layer2 is a new downsampling layer pool1 defined; layer3 is a new convolution layer conv2 defined, and the input is the output of layer 2; layer4 is a new downsampling layer pool2 defined; layer5 is the full-connection layer, the output of layer4 is connected, the feature images of the same image in the upper layer through different convolution kernels are combined into one-dimensional vectors, layer6 is the classification layer, finally, the error value is calculated by adopting a loss function, and the training error is output. The weight is corrected by adopting a backward conduction mode through the error value, so that the aim of training the network is fulfilled. And updating the model parameters according to the corrected weight, and repeating training for a plurality of times on training sample images of the existing hospital medicines so as to determine the currently learned model parameters including the parameters of CNN and the parameters of a classifier. The classifier is formed by combining 2 layers of full-connection layers in series with SOFTMAX. After the preparation is completed, the target image after preprocessing can be identified, and the classifier outputs an identification result according to a pre-established identification model (medicine classification database).
The medicine identification method based on the convolutional neural network algorithm is the prior art, and will not be repeated here. Note that, in the medicine inspection sorting process, the processing unit 3 does not control the air gun 41 to perform the sorting operation for the defective medicines that are broken and incomplete, and the broken medicines are directly pushed into the second storage box 52.
According to the medicine classification device, medicines are transmitted on the pipeline of the transmission unit by the aid of the transmission unit, all-feature image acquisition is carried out on each medicine in real time by the camera 2, the processing unit 3 judges which classification the medicines belong to through identifying and analyzing the images, and the classification execution unit is controlled to push the medicines to the storage units corresponding to the classifications, so that full automation of medicine detection classification is realized, human errors caused by manual detection classification are avoided, labor cost is saved, and medicine classification efficiency is improved.
The above embodiments are only preferred embodiments of the present invention, and the scope of the present invention is not limited thereto, but any insubstantial changes and substitutions made by those skilled in the art on the basis of the present invention are intended to be within the scope of the present invention as claimed.
Claims (4)
1. A drug sorting device, characterized in that: the device comprises a transmission unit, a camera, a processing unit, a classification execution unit and a storage unit, wherein the processing unit is respectively connected with the camera and the classification execution unit in a signal manner; the transmission unit is used for placing the medicines on a production line for transmission, the camera acquires medicine images, the processing unit performs identification and classification according to the medicine images, and controls the classification execution unit to correspondingly push the medicines to the storage unit;
the processing unit performs the following method: acquiring a shot medicine full-feature image, solving an image gray level image, positioning the outer frame position of the medicine through edge detection and outer frame processing, acquiring medicine features through a convolutional neural network algorithm, and comparing according to a pre-established medicine classification model to determine the classification of the medicine;
the conveying unit comprises an input port, a bed body and a conveying belt, wherein the conveying belt is arranged at the top of the bed body and is driven to rotate by a rotating shaft; the input port is arranged at one end of the conveying belt;
the classifying and executing units are a plurality of air guns, the air guns are arranged at one side of a conveying line of the conveying belt at intervals, and gun openings of the air guns face the conveying belt;
the storage units comprise a plurality of first storage boxes which are arranged on the other side of the conveying line of the conveying belt at intervals, and the first storage boxes are arranged corresponding to the air guns;
the processing unit is further configured to:
initializing the binding of the medicine category and the corresponding air gun based on the horizontal and vertical distance between the current position of each medicine falling to the conveyor belt and the position of the corresponding air gun until all medicine categories and all air guns are bound; binding an air gun for each medicine category;
based on the determination of the category to which the current medicine belongs and the binding relation between the category of the current medicine and the air gun, taking the current position of the current medicine falling from the input port to the conveying belt as a starting point, when the horizontal and vertical distance between the current medicine and the current position of the corresponding air gun accords with the binding relation, controlling the corresponding air gun to spray so as to blow the current medicine to the corresponding first storage box.
2. A drug sorting apparatus as in claim 1 wherein: the conveying unit further comprises a baffle plate, the baffle plate is arranged at one end of the top of the conveying belt, which is abutted to the input port, and the baffle plate is used for adjusting the conveying route of medicines.
3. A drug sorting apparatus as in claim 1 wherein: the air gun is also provided with an air gun fixing seat.
4. A drug sorting apparatus as in claim 1 wherein: the storage unit further comprises a second storage box, and the second storage box is correspondingly arranged at the tail end of the conveying line of the conveying belt.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810834149.8A CN108875845B (en) | 2018-07-26 | 2018-07-26 | Medicine sorting device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810834149.8A CN108875845B (en) | 2018-07-26 | 2018-07-26 | Medicine sorting device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108875845A CN108875845A (en) | 2018-11-23 |
| CN108875845B true CN108875845B (en) | 2024-02-20 |
Family
ID=64305405
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810834149.8A Active CN108875845B (en) | 2018-07-26 | 2018-07-26 | Medicine sorting device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108875845B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110665826A (en) * | 2019-08-30 | 2020-01-10 | 杭州金凯包装材料有限公司 | Medical product sorting system and sorting method thereof |
| CN110781298B (en) * | 2019-09-18 | 2023-06-20 | 平安科技（深圳）有限公司 | Medicine classification method, apparatus, computer device and storage medium |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010151444A (en) * | 2008-12-23 | 2010-07-08 | Mitsubishi Electric Corp | Method and apparatus for sorting resin |
| JP2010182287A (en) * | 2008-07-17 | 2010-08-19 | Steven C Kays | Intelligent adaptive design |
| CN103551742A (en) * | 2013-10-30 | 2014-02-05 | 马志敏 | High-speed controlled-release tablet perforating machine |
| CN204134916U (en) * | 2014-08-27 | 2015-02-04 | 台虹科技股份有限公司 | sorting device |
| CN104634790A (en) * | 2015-02-09 | 2015-05-20 | 杭州乔戈里科技有限公司 | Capsule detection method and high-speed fully-automatic detection device |
| WO2015170761A1 (en) * | 2014-05-09 | 2015-11-12 | 株式会社湯山製作所 | Drug assortment device and drug assortment method |
| CN205087576U (en) * | 2014-01-07 | 2016-03-16 | 陆洪瑞 | Transmission carrier and pneumatic pipeline transmission system |
| CN105973892A (en) * | 2016-05-04 | 2016-09-28 | 成都贝森伟任科技有限责任公司 | Automatic medicine tablet packaging quality monitoring system |
| CN206184783U (en) * | 2016-09-05 | 2017-05-24 | 贵州鼎圣药业有限公司 | Can error correcting letter sorting system |
| CN107451592A (en) * | 2017-06-30 | 2017-12-08 | 广东数相智能科技有限公司 | A kind of ethical goods checking method and device |
| CN107545150A (en) * | 2017-10-13 | 2018-01-05 | 张晨 | Medicine identifying system and its recognition methods based on deep learning |
| CN108311414A (en) * | 2018-01-10 | 2018-07-24 | 陈德言 | A kind of tablet press machine tablet automatic detection device |
| CN208460038U (en) * | 2018-07-26 | 2019-02-01 | 广东数相智能科技有限公司 | A kind of classifying drugs device |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI831775B (en) * | 2018-06-19 | 2024-02-11 | 日商湯山製作所股份有限公司 | Drug sorting device |
-
2018
- 2018-07-26 CN CN201810834149.8A patent/CN108875845B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010182287A (en) * | 2008-07-17 | 2010-08-19 | Steven C Kays | Intelligent adaptive design |
| JP2010151444A (en) * | 2008-12-23 | 2010-07-08 | Mitsubishi Electric Corp | Method and apparatus for sorting resin |
| CN103551742A (en) * | 2013-10-30 | 2014-02-05 | 马志敏 | High-speed controlled-release tablet perforating machine |
| CN205087576U (en) * | 2014-01-07 | 2016-03-16 | 陆洪瑞 | Transmission carrier and pneumatic pipeline transmission system |
| WO2015170761A1 (en) * | 2014-05-09 | 2015-11-12 | 株式会社湯山製作所 | Drug assortment device and drug assortment method |
| CN204134916U (en) * | 2014-08-27 | 2015-02-04 | 台虹科技股份有限公司 | sorting device |
| CN104634790A (en) * | 2015-02-09 | 2015-05-20 | 杭州乔戈里科技有限公司 | Capsule detection method and high-speed fully-automatic detection device |
| CN105973892A (en) * | 2016-05-04 | 2016-09-28 | 成都贝森伟任科技有限责任公司 | Automatic medicine tablet packaging quality monitoring system |
| CN206184783U (en) * | 2016-09-05 | 2017-05-24 | 贵州鼎圣药业有限公司 | Can error correcting letter sorting system |
| CN107451592A (en) * | 2017-06-30 | 2017-12-08 | 广东数相智能科技有限公司 | A kind of ethical goods checking method and device |
| CN107545150A (en) * | 2017-10-13 | 2018-01-05 | 张晨 | Medicine identifying system and its recognition methods based on deep learning |
| CN108311414A (en) * | 2018-01-10 | 2018-07-24 | 陈德言 | A kind of tablet press machine tablet automatic detection device |
| CN208460038U (en) * | 2018-07-26 | 2019-02-01 | 广东数相智能科技有限公司 | A kind of classifying drugs device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108875845A (en) | 2018-11-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107486415A (en) | Thin bamboo strip defect on-line detecting system and detection method based on machine vision | |
| CN110314854A (en) | A kind of device and method of the workpiece sensing sorting of view-based access control model robot | |
| CN108875845B (en) | Medicine sorting device | |
| CN109794437A (en) | Intelligent sorting system based on computer vision | |
| CN212301356U (en) | A visual inspection device for wheel hub welds | |
| CN113077450B (en) | Cherry grading detection method and system based on deep convolutional neural network | |
| CN110252674A (en) | A kind of method and system of the packing box quality testing based on machine vision | |
| CN117649404A (en) | Medicine packaging box quality detection method and system based on image data analysis | |
| CN109693140A (en) | A kind of intelligent flexible production line and its working method | |
| CN116893175A (en) | Multi-station multi-defect unified connector measurement system based on deep learning | |
| Zou et al. | Improved ResNet-50 model for identifying defects on wood surfaces | |
| Ramirez et al. | Agroindustrial plant for the classification of hass avocados in real-time with ResNet-18 architecture | |
| CN111178405A (en) | A Similar Object Recognition Method Integrating Multiple Neural Networks | |
| CN108465644B (en) | Artificial intelligent wheat quality inspection robot and quality inspection method | |
| CN117289626B (en) | Virtual simulation method and system for industrialization | |
| CN208125628U (en) | A kind of engine commutator quality detecting system based on machine vision | |
| CN108176603A (en) | The more code name recognition methods of engine link and its automatic sorting apparatus | |
| CN209550027U (en) | Intelligent sorting system for disposable diapers based on computer vision | |
| CN111368833A (en) | A fully automatic steel loading method for slab number detection based on machine vision | |
| CN116152228A (en) | Tire defect detection method and system based on machine vision and machine learning | |
| CN116342502A (en) | A method of industrial vision detection based on deep learning | |
| CN113128502A (en) | Single piece separation system, method and device based on deep learning and machine vision | |
| TWI815318B (en) | Warehousing automatic sorting system | |
| CN112240888A (en) | Visual detection method and visual detection system | |
| CN208460038U (en) | A kind of classifying drugs device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |