CN114264668A - Method and system for detecting three-stage flaws of medical packaging box by image processing - Google Patents
Method and system for detecting three-stage flaws of medical packaging box by image processing Download PDFInfo
- Publication number
- CN114264668A CN114264668A CN202210204376.9A CN202210204376A CN114264668A CN 114264668 A CN114264668 A CN 114264668A CN 202210204376 A CN202210204376 A CN 202210204376A CN 114264668 A CN114264668 A CN 114264668A
- Authority
- CN
- China
- Prior art keywords
- training
- roi
- defect
- model
- packaging box
- 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.)
- Pending
Links
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention provides a method and a system for detecting three-stage flaws of a medical packaging box by using image processing, wherein the system comprises the following steps: the system comprises a camera, a control device, a processor and a machine vision processing tool, wherein the processor is connected with the machine vision processing tool, and the machine vision processing tool is used for obtaining an ROI needing deep learning model detection; the device comprises a detection model, a target detector, a writing module and a classification module; and the training model is obtained by establishing a training deep neural network for iterative computation according to the defect state of the historical sample. In order to realize the detection of the defects of the aluminum-plastic blister packaged medicines and achieve the best detection effect, a large number of defect samples are collected, accurate labeled data are trained through a deep neural network, a deep learning model is further obtained, and the effect of high-accuracy detection exceeding that of a human expert is achieved after multiple iterations.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a method and a system for detecting three-stage flaws of a medical packaging box by using image processing.
Background
With the continuous upgrade of the medicine production quality management practice (GMP) in recent years, the requirement of the medicine industry on the medicine production quality is higher and higher, and pharmaceutical manufacturers have urgent needs for improving the production efficiency on the premise of ensuring the quality, so that a three-stage detection system with high production efficiency and strong product compatibility of the medical packaging box packing machine is imperative.
In the prior art, manual visual inspection is mostly adopted or a visual inspection station is set for a certain product. Manually and visually inspecting the three-stage characters by a special light source, manually adjusting the angle to enable the characters to be highlighted above the background of the medicine box, observing that defective products are manually removed, and failing to realize high-efficiency full-automatic production; the mode that sets up the visual inspection station alone to different products, according to the mode highlighting character of product character characteristic through mechanical adjustment camera and light source angle, reuse dedicated traditional algorithm to solve the third phase flaw detection of certain product, can't accomplish product compatibility and change the type fast.
Disclosure of Invention
In view of the above, the present invention provides a method and a system for detecting a three-stage defect of a medical packaging box by image processing.
The technical scheme adopted by the invention is as follows:
a method for detecting three-stage flaws of a medical packaging box by using image processing comprises the following steps:
arranging a photoelectric sensor on a conveyor belt, when a medicine packaging box passes through the photoelectric sensor on the conveyor belt, the photoelectric sensor starts to acquire an encoder pulse value to calculate the running distance of a product, a rising edge signal is given when the medicine packaging box reaches a triggered photographing position to trigger a light source to strobe a synchronous camera to photograph, and a first image with a relatively fixed position is acquired;
extracting the position of an integral character from the first image according to relative coordinates through a machine vision positioning algorithm, performing segmentation processing according to the specification of the character, obtaining an ROI (region of interest) needing deep learning model detection by using a machine vision processing tool, inputting the ROI into a detection model, and obtaining a corresponding detection result from a preset scheme obtained through training of the detection model;
if the ROI does not obtain a corresponding detection result in the preset scheme, extracting the ROI, sending the ROI to a target detector to obtain an ROI attribute, writing identification features into the ROI attribute by using a writing module arranged in the target detector, and classifying the ROI into a preset sub-library with the same identification features in a training model according to the identification features;
according to the deep neural network branches used for training corresponding to the preset sub-library, the training model starts an avoidance mechanism, training resources of the training model are distributed to the deep neural network branches for performing iterative training on the ROI, and after training is completed, the training resources are distributed and stored to a subset corresponding to a preset scheme of the learning model according to the recognition features.
Further, the training model is obtained according to the following method:
collecting a plurality of groups of defect samples, inputting the defect samples into a processing module to mark defect positions, recording defect states of the defect positions in the defect samples, simultaneously sending the defect samples to a target detector to obtain attributes of the defect samples, writing identification features in the attributes of the defect samples by using a writing module arranged in the target detector, and classifying the defect samples into preset sub-libraries with the same identification features according to the identification features;
correspondingly establishing a training deep neural network according to a preset sub-library, distributing training resources to the training deep neural network by a training model according to the size of a memory of the preset sub-library, performing iterative training on the defect samples in the corresponding preset sub-library, and distributing and storing the training deep neural network into a subset corresponding to a preset scheme of a learning model according to recognition characteristics after training.
Further, a corresponding detection result is obtained in a preset scheme obtained through the training of the detection model, the detection result comprises good products or defective products in different states, an ok or ng signal is sent to a rear-end removing device according to the detection result, and the result is displayed on a UI interactive interface.
The invention also provides a system for detecting the third-stage flaws of the medical packaging box by using image processing, which comprises a conveying belt, wherein a photoelectric sensor is arranged on the conveying belt;
the camera is used for reaching the shooting triggering position, giving a rising edge signal to trigger the light source to strobe the synchronous camera to shoot, and acquiring a first image with a relatively fixed position;
the control device is used for controlling the trigger rising edge signal to be sent to the camera;
a processor for performing segmentation processing on the first image and the second image;
the processor is connected with a machine vision processing tool, and the machine vision processing tool is used for obtaining an ROI needing deep learning model detection;
the detection model receives the ROI and acquires a corresponding detection result from a preset scheme obtained by training the detection model;
the target detector is used for extracting the ROI and acquiring the attribute of the ROI when the ROI does not acquire a corresponding detection result in a preset scheme;
a writing module, disposed in the target detector, for writing the identifying feature in the ROI attribute;
the classification module classifies the ROI into a preset sub-library with the same identification characteristics according to the identification characteristics;
and the training model is obtained by establishing a training deep neural network for iterative computation according to the defect state of the historical sample.
Further, the photoelectric sensor is used for obtaining the encoder pulse value after the medicine packing box passes through the photoelectric sensor on the conveying belt, so that the product running distance is calculated, the rising edge signal is given when the medicine packing box reaches the triggering photographing position, the light source stroboscopic synchronous camera is triggered to photograph, and the first image with the relatively fixed position is obtained.
Furthermore, a processing module, a recording module, a classification module, a plurality of preset sub-libraries and a plurality of training deep neural networks are arranged in the training model;
the processing module is used for marking the defect positions of a plurality of groups of collected defect samples;
the recording module is used for recording the defect state of the defect position in the defect sample;
the classification module is used for classifying the defect samples into preset sub-libraries with the same identification characteristics according to the identification characteristics written in the attributes of the defect samples;
the preset sub-library is used for storing corresponding defect samples according to the identification characteristics;
correspondingly establishing a training deep neural network according to a preset sub-library, distributing training resources to the training deep neural network by a training model according to the size of a memory of the preset sub-library, performing iterative training on the defect samples in the corresponding preset sub-library, and distributing and storing the training deep neural network into a subset corresponding to a preset scheme of a learning model according to recognition characteristics after training.
Furthermore, an avoidance mechanism is arranged in the training model and used for selectively distributing the training resources of the training model to one of the deep neural network branches.
Furthermore, an avoidance mechanism is arranged in the training model and used for selectively distributing the training resources of the training model to one of the training deep neural networks.
And further, the system also comprises an interactive display control screen used for displaying the result on the UI interactive interface.
This application adopts conveyer belt material loading mode, and the medicine packing carton walks over photoelectric sensor, begins to acquire encoder pulse value, calculates product travel distance, and the arrival triggers the position of shooing and gives rise to the synchronous camera of edge signal trigger light source stroboscopic and shoot, acquires the relatively fixed image in position. The multi-angle light source is adopted at the photographing position, so that the device is compatible with products of different specifications, and clear character imaging is ensured.
After the image is obtained, the position of the whole character is extracted according to the relative coordinates through a traditional machine vision positioning algorithm, the position of each line of characters is segmented according to the character specification to detect the line spacing too wide, too narrow and poor deviation, and then each line of characters is segmented into single numbers.
Drawings
The invention is illustrated and described only by way of example and not by way of limitation in the scope of the invention as set forth in the following drawings, in which:
FIG. 1 is a flow chart of a method of the present invention;
FIG. 2 is a schematic diagram of the framework of the system of the present invention;
FIG. 3 is a schematic diagram of a framework of a training model according to the present invention;
FIG. 4 is a flow chart of the detection in the present invention;
FIG. 5 is a flow chart of training model building in the present invention.
Detailed Description
In order to make the objects, technical solutions, design methods, and advantages of the present invention more apparent, the present invention will be further described in detail by specific embodiments with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Explanation: the machine vision processing tool is HALCON of MVTec.
ROI region of interest, a region having defects in the present invention.
Referring to fig. 1 to 5, the present invention provides a system for detecting a three-stage defect of a medical packaging box by using image processing, comprising:
the conveying belt is provided with a photoelectric sensor;
the camera is used for reaching the shooting triggering position, giving a rising edge signal to trigger the light source to strobe the synchronous camera to shoot, and acquiring a first image with a relatively fixed position; when the medicine packaging box passes through the photoelectric sensor, the photoelectric sensor starts to acquire an encoder pulse value to calculate the running distance of a product, a rising edge signal is given when the product reaches a triggered photographing position to trigger a light source strobe synchronous camera to photograph, and an image with a relatively fixed position is acquired.
The control device is used for controlling the trigger rising edge signal to be sent to the camera;
the processor is used for segmenting the first image, specifically, extracting the position of the whole character according to relative coordinates by a machine vision positioning algorithm after the first image is obtained, segmenting the position of each line of characters according to the character specification to detect too wide and too narrow line spacing and poor deviation, and segmenting each line of characters into single numbers;
the processor is connected with a machine vision processing tool, and the machine vision processing tool is used for obtaining an ROI needing deep learning model detection;
the detection model receives the ROI and acquires a corresponding detection result from a preset scheme obtained by training the detection model;
the target detector is used for extracting the ROI and acquiring the attribute of the ROI when the ROI does not acquire a corresponding detection result in a preset scheme;
a writing module, disposed in the target detector, for writing the identifying feature in the ROI attribute;
the classification module classifies the ROI into a preset sub-library with the same identification characteristics according to the identification characteristics;
and the training model is obtained by establishing a training deep neural network for iterative computation according to the defect state of the historical sample.
Further, the photoelectric sensor is used for obtaining the encoder pulse value after the medicine packing box passes through the photoelectric sensor on the conveying belt, so that the product running distance is calculated, the rising edge signal is given when the medicine packing box reaches the triggering photographing position, the light source stroboscopic synchronous camera is triggered to photograph, and the first image with the relatively fixed position is obtained.
Preferably, a processing module, a recording module, a classification module, a plurality of preset sub-libraries and a plurality of training deep neural networks are arranged in the training model;
the processing module is used for marking the defect positions of a plurality of groups of collected defect samples;
the recording module is used for recording the defect state of the defect position in the defect sample;
the classification module is used for classifying the defect samples into preset sub-libraries with the same identification characteristics according to the identification characteristics written in the attributes of the defect samples;
the preset sub-library is used for storing corresponding defect samples according to the identification characteristics;
correspondingly establishing a training deep neural network according to a preset sub-library, distributing training resources to the training deep neural network by a training model according to the size of a memory of the preset sub-library, performing iterative training on the defect samples in the corresponding preset sub-library, and distributing and storing the training deep neural network into a subset corresponding to a preset scheme of a learning model according to recognition characteristics after training.
Preferably, an avoidance mechanism is further arranged in the training model, and the avoidance mechanism is used for selectively allocating the training resources of the training model to one of the deep neural network branches.
Preferably, an avoidance mechanism is further arranged in the training model, and the avoidance mechanism is used for selectively allocating the training resources of the training model to one of the training deep neural networks.
Preferably, the system further comprises an interactive display control screen for displaying the result through the interactive interface.
This application adopts the conveyer belt material loading mode, walks over photoelectric sensor when medicine packing carton, and photoelectric sensor begins to acquire the encoder pulse value, calculates product running distance, and the arrival triggers the position of shooing and gives rise to the synchronous camera of edge signal trigger light source stroboscopic and shoot, acquires the first image of position relatively fixed. The multi-angle light source is adopted at the photographing position, so that the device is compatible with products of different specifications, and clear character imaging is ensured.
After a first image is obtained, extracting the position of an integral character according to relative coordinates through a traditional machine vision positioning algorithm, segmenting the position of each line of characters according to the specification of the character, segmenting an ROI (region of interest) needing deep learning model detection through image preprocessing, transmitting the ROI into a model for analysis, predicting whether the result is a good product or various defective products, sending an ok or ng signal to a removing device according to the detection result, and displaying the result on an interactive interface.
In order to realize the detection of the defects of the medicine packaging box and achieve the best detection effect, a large number of defect samples such as too wide line spacing, too narrow line spacing and poor deviation are collected, then each line of characters are divided into single numbers, accurate-labeled data are trained through a deep neural network, a deep learning model is further obtained, the effect of high-accuracy detection exceeding that of human experts is achieved after multiple iterations, and finally intelligent detection of the aluminum plastic blister packaging medicine is achieved.
During model training in the early stage, pictures with different numbers are manually marked and divided into good products and various flaws, and deep learning model training is performed to obtain a model. In the actual production process, the trained model is read, deep learning model prediction is carried out on each digit, whether the digit is a defective product or not is judged, and if the digit is the defective product, a rejecting signal is sent out. The rejecting device rejects the defective products and displays the prediction result on the UI.
Referring to fig. 3, in addition, the invention also provides a method for detecting the third-stage flaw of the medical packaging box by using image processing, which comprises the following steps:
arranging a photoelectric sensor on a conveyor belt, when a medicine packaging box passes through the photoelectric sensor on the conveyor belt, the photoelectric sensor starts to acquire an encoder pulse value to calculate the running distance of a product, a rising edge signal is given when the medicine packaging box reaches a triggered photographing position to trigger a light source to strobe a synchronous camera to photograph, and a first image with a relatively fixed position is acquired;
extracting the position of an integral character from the first image according to relative coordinates through a machine vision positioning algorithm, performing segmentation processing according to the specification of the character, obtaining an ROI (region of interest) needing deep learning model detection by using a machine vision processing tool, inputting the ROI into a detection model, and obtaining a corresponding detection result from a preset scheme obtained through training of the detection model;
if the ROI does not obtain a corresponding detection result in the preset scheme, extracting the ROI, sending the ROI to a target detector to obtain an ROI attribute, writing identification features into the ROI attribute by using a writing module arranged in the target detector, and classifying the ROI into a preset sub-library with the same identification features in a training model according to the identification features;
according to the deep neural network branches used for training corresponding to the preset sub-library, the training model starts an avoidance mechanism, training resources of the training model are distributed to the deep neural network branches for performing iterative training on the ROI, and after training is completed, the training resources are distributed and stored to a subset corresponding to a preset scheme of the learning model according to the recognition features.
Referring to fig. 4, in the above, the training model is obtained according to the following method:
collecting a plurality of groups of defect samples, inputting the defect samples into a processing module to mark defect positions, recording defect states of the defect positions in the defect samples, simultaneously sending the defect samples to a target detector to obtain attributes of the defect samples, writing identification features in the attributes of the defect samples by using a writing module arranged in the target detector, and classifying the defect samples into preset sub-libraries with the same identification features according to the identification features;
correspondingly establishing a training deep neural network according to a preset sub-library, distributing training resources to the training deep neural network by a training model according to the size of a memory of the preset sub-library, performing iterative training on the defect samples in the corresponding preset sub-library, and distributing and storing the training deep neural network into a subset corresponding to a preset scheme of a learning model according to recognition characteristics after training.
When the medicine packaging box passes through the photoelectric sensor, the photoelectric sensor starts to acquire an encoder pulse value to calculate the running distance of a product, a rising edge signal is given when the medicine packaging box reaches a triggered photographing position to trigger a light source to flash and synchronize a camera to photograph, and a first image with a relatively fixed position is acquired; dividing an ROI (region of interest) needing deep learning model detection through image preprocessing, transmitting the ROI into a model for analysis, predicting whether a result is a good product or various defective products, sending an ok or ng signal to a removing device according to the detection result, and displaying the result on an interactive interface.
If the flaw does not obtain a corresponding detection result in the preset scheme, extracting the flaw, sending the flaw to a target detector to obtain a flaw attribute, writing identification features in the flaw attribute by using a writing module arranged in the target detector, and classifying the flaw into a preset sub-library with the same identification features in a training model according to the identification features;
according to the deep neural network branches used for training corresponding to the preset sub-library, the training model starts an avoidance mechanism, training resources of the training model are distributed to the deep neural network branches for iterative training of the flaws, and after training is completed, the training resources are distributed and stored to the subsets corresponding to the preset scheme of the learning model according to the recognition features.
Having described embodiments of the present invention, the foregoing description is intended to be exemplary, not exhaustive, and not limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen in order to best explain the principles of the embodiments, the practical application, or improvements made to the technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.
Claims (9)
1. A method for detecting three-stage flaws of a medical packaging box by using image processing is characterized by comprising the following steps:
arranging a photoelectric sensor on a conveyor belt, when a medicine packaging box passes through the photoelectric sensor on the conveyor belt, the photoelectric sensor starts to acquire an encoder pulse value to calculate the running distance of a product, a rising edge signal is given when the medicine packaging box reaches a triggered photographing position to trigger a light source to strobe a synchronous camera to photograph, and a first image with a relatively fixed position is acquired;
extracting the position of an integral character from the first image according to relative coordinates through a machine vision positioning algorithm, performing segmentation processing according to the specification of the character, obtaining an ROI (region of interest) needing deep learning model detection by using a machine vision processing tool, inputting the ROI into a detection model, and obtaining a corresponding detection result from a preset scheme obtained through training of the detection model;
if the ROI does not obtain a corresponding detection result in the preset scheme, extracting the ROI, sending the ROI to a target detector to obtain an ROI attribute, writing identification features into the ROI attribute by using a writing module arranged in the target detector, and classifying the ROI into a preset sub-library with the same identification features in a training model according to the identification features;
according to the deep neural network branches used for training corresponding to the preset sub-library, the training model starts an avoidance mechanism, training resources of the training model are distributed to the deep neural network branches for performing iterative training on the ROI, and after training is completed, the training resources are distributed and stored to a subset corresponding to a preset scheme of the learning model according to the recognition features.
2. The method for detecting the third-stage flaw of the medical packaging box by using the image processing as claimed in claim 1, wherein the training model is obtained according to the following method:
collecting a plurality of groups of defect samples, inputting the defect samples into a processing module to mark defect positions, recording defect states of the defect positions in the defect samples, simultaneously sending the defect samples to a target detector to obtain attributes of the defect samples, writing identification features in the attributes of the defect samples by using a writing module arranged in the target detector, and classifying the defect samples into preset sub-libraries with the same identification features according to the identification features;
correspondingly establishing a training deep neural network according to a preset sub-library, distributing training resources to the training deep neural network by a training model according to the size of a memory of the preset sub-library, performing iterative training on the defect samples in the corresponding preset sub-library, and distributing and storing the training deep neural network into a subset corresponding to a preset scheme of a learning model according to recognition characteristics after training.
3. The method for detecting the third-stage defects of the medical packaging box by using the image processing as claimed in claim 1, wherein the corresponding detection results are obtained from a preset scheme obtained by training a detection model, the detection results comprise good products or defective products in different states, ok or ng signals are sent to a removing device at the rear end according to the detection results, and the results are displayed on a UI interactive interface.
4. A system for detecting three-stage flaws of a medical packaging box by using image processing is characterized by comprising:
the conveying belt is provided with a photoelectric sensor;
the camera is used for reaching the shooting triggering position, giving a rising edge signal to trigger the light source to strobe the synchronous camera to shoot, and acquiring a first image with a relatively fixed position;
the control device is used for controlling the trigger rising edge signal to be sent to the camera;
a processor for performing segmentation processing on the first image;
the processor is connected with a machine vision processing tool, and the machine vision processing tool is used for obtaining an ROI needing deep learning model detection;
the detection model receives the ROI and acquires a corresponding detection result from a preset scheme obtained by training the detection model;
the target detector is used for extracting the ROI and acquiring the attribute of the ROI when the ROI does not acquire a corresponding detection result in a preset scheme;
a writing module, disposed in the target detector, for writing the identifying feature in the ROI attribute;
the classification module classifies the ROI into a preset sub-library with the same identification characteristics according to the identification characteristics;
and the training model is obtained by establishing a training deep neural network for iterative computation according to the defect state of the historical sample.
5. The system for detecting the third-stage defects of the medical packaging box by utilizing the image processing as claimed in claim 4, wherein the photoelectric sensor is used for acquiring the encoder pulse value after the medical packaging box passes through the photoelectric sensor on the conveyor belt so as to calculate the product running distance, the rising edge signal is given when the product running distance reaches the shooting triggering position so as to trigger the light source to strobe and synchronize the camera to shoot, and the first image with a relatively fixed position is acquired.
6. The system for detecting the third-stage defects of the medical packaging box by using image processing as claimed in claim 4, wherein a processing module, a recording module, a classifying module, a plurality of preset sub-libraries and a plurality of training deep neural networks are arranged in the training model;
the processing module is used for marking the defect positions of a plurality of groups of collected defect samples;
the recording module is used for recording the defect state of the defect position in the defect sample;
the classification module is used for classifying the defect samples into preset sub-libraries with the same identification characteristics according to the identification characteristics written in the attributes of the defect samples;
the preset sub-library is used for storing corresponding defect samples according to the identification characteristics;
correspondingly establishing a training deep neural network according to a preset sub-library, distributing training resources to the training deep neural network by a training model according to the size of a memory of the preset sub-library, performing iterative training on the defect samples in the corresponding preset sub-library, and distributing and storing the training deep neural network into a subset corresponding to a preset scheme of a learning model according to recognition characteristics after training.
7. The system for detecting the third-stage flaw of the medical packaging box by using the image processing as claimed in claim 4 or 6, wherein an evasion mechanism is further arranged in the training model and used for selectively distributing the training resources of the training model to one of the deep neural network branches.
8. The system for detecting the third-stage flaw of the medical packaging box by using the image processing as claimed in claim 4 or 6, wherein an evasion mechanism is further arranged in the training model and used for selectively distributing the training resources of the training model to one of the deep training neural networks.
9. The system for detecting the third-stage defect of the medical packaging box by using image processing as claimed in claim 4, further comprising a control screen for interactive display, wherein the control screen is used for displaying the result through a UI interactive interface.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210204376.9A CN114264668A (en) | 2022-03-03 | 2022-03-03 | Method and system for detecting three-stage flaws of medical packaging box by image processing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210204376.9A CN114264668A (en) | 2022-03-03 | 2022-03-03 | Method and system for detecting three-stage flaws of medical packaging box by image processing |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114264668A true CN114264668A (en) | 2022-04-01 |
Family
ID=80834025
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210204376.9A Pending CN114264668A (en) | 2022-03-03 | 2022-03-03 | Method and system for detecting three-stage flaws of medical packaging box by image processing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114264668A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN203512149U (en) * | 2013-08-23 | 2014-04-02 | 天津森茂科技有限公司 | Medicine package three-phase code image detecting system |
| CN107657603A (en) * | 2017-08-21 | 2018-02-02 | 北京精密机电控制设备研究所 | A kind of industrial appearance detecting method based on intelligent vision |
| CN207440594U (en) * | 2017-12-01 | 2018-06-01 | 辽宁民康制药有限公司 | A kind of device for detecting product appearance quality |
| CN109724990A (en) * | 2019-01-08 | 2019-05-07 | 上海大学 | The quick positioning and detection method in coding region in a kind of packing box label |
| CN110147792A (en) * | 2019-05-22 | 2019-08-20 | 齐鲁工业大学 | The Key works Drug packing character high-speed detection system and method optimized based on memory |
| CN215545857U (en) * | 2021-08-20 | 2022-01-18 | 湖南雪天盐业技术开发有限公司 | Salt packaging code printing monitoring system |
| CN114088730A (en) * | 2022-01-24 | 2022-02-25 | 心鉴智控(深圳)科技有限公司 | Method and system for detecting aluminum-plastic bubble cap defects by using image processing |
-
2022
- 2022-03-03 CN CN202210204376.9A patent/CN114264668A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN203512149U (en) * | 2013-08-23 | 2014-04-02 | 天津森茂科技有限公司 | Medicine package three-phase code image detecting system |
| CN107657603A (en) * | 2017-08-21 | 2018-02-02 | 北京精密机电控制设备研究所 | A kind of industrial appearance detecting method based on intelligent vision |
| CN207440594U (en) * | 2017-12-01 | 2018-06-01 | 辽宁民康制药有限公司 | A kind of device for detecting product appearance quality |
| CN109724990A (en) * | 2019-01-08 | 2019-05-07 | 上海大学 | The quick positioning and detection method in coding region in a kind of packing box label |
| CN110147792A (en) * | 2019-05-22 | 2019-08-20 | 齐鲁工业大学 | The Key works Drug packing character high-speed detection system and method optimized based on memory |
| CN215545857U (en) * | 2021-08-20 | 2022-01-18 | 湖南雪天盐业技术开发有限公司 | Salt packaging code printing monitoring system |
| CN114088730A (en) * | 2022-01-24 | 2022-02-25 | 心鉴智控(深圳)科技有限公司 | Method and system for detecting aluminum-plastic bubble cap defects by using image processing |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106290382A (en) | Bubble-cap tablet package defective vision detection device and method | |
| CN108229561B (en) | Particle product defect detection method based on deep learning | |
| CN105251707B (en) | Bad part eject sorting equipment based on medical large transfusion visible foreign matters detecting system | |
| CN107966454A (en) | A kind of end plug defect detecting device and detection method based on FPGA | |
| JP2002513463A (en) | System and method for detecting stress in a molded container | |
| CN104118609B (en) | Labeling quality determining method and device | |
| CN106952258A (en) | A kind of bottle mouth defect detection method based on gradient orientation histogram | |
| CN104483320A (en) | Digitized defect detection device and detection method of industrial denitration catalyst | |
| CN104634790A (en) | Capsule detection method and high-speed fully-automatic detection device | |
| CN109297984A (en) | A kind of blister package defect inspection method, device and equipment | |
| WO2023134286A1 (en) | Online automatic quality testing and classification method for cathode copper | |
| EP3532391A1 (en) | System and method for automatic inspection and classification of discrete items | |
| CN107328781A (en) | A kind of columnar product detection method of surface flaw and device based on machine vision | |
| CN210071686U (en) | Fruit grading plant based on orthogonal binocular machine vision | |
| CN111185398A (en) | Online omnibearing intelligent detection system and detection method for vacuum blood collection tube | |
| BR112021010404A2 (en) | Systems and methods to facilitate clone selection | |
| CN114813761B (en) | Double-light-stroboscopic-based film pinhole and bright spot defect identification system and method | |
| CN114088730B (en) | Method and system for detecting aluminum-plastic bubble cap defects by using image processing | |
| CN104574398B (en) | A kind of vanning cigarette filter stick number system based on line scan camera | |
| CN114264668A (en) | Method and system for detecting three-stage flaws of medical packaging box by image processing | |
| CN110508510A (en) | A kind of plastic pump defect inspection method, apparatus and system | |
| CN115218790A (en) | Bar detection method, device and system | |
| CN114324398A (en) | Intelligent visual detection system and detection method for medical and edible packaging label | |
| CN113996545A (en) | Quick sorting production line for main roots of panax notoginseng based on computer vision | |
| CN118130005B (en) | Plant fiber packaging product tightness detection method and application thereof |
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 |