CN112633410A - Chinese herbal medicine frying process monitoring device and Chinese herbal medicine frying judgment method - Google Patents
Chinese herbal medicine frying process monitoring device and Chinese herbal medicine frying judgment method Download PDFInfo
- Publication number
- CN112633410A CN112633410A CN202110000207.9A CN202110000207A CN112633410A CN 112633410 A CN112633410 A CN 112633410A CN 202110000207 A CN202110000207 A CN 202110000207A CN 112633410 A CN112633410 A CN 112633410A
- Authority
- CN
- China
- Prior art keywords
- frying
- camera
- traditional chinese
- chinese herbal
- fried
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
-
- 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
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N5/00—Details of television systems
- H04N5/76—Television signal recording
- H04N5/765—Interface circuits between an apparatus for recording and another apparatus
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Signal Processing (AREA)
- Medical Preparation Storing Or Oral Administration Devices (AREA)
Abstract
A traditional Chinese medicine frying process monitoring device and a traditional Chinese medicine frying judgment method relate to a detection device and a detection method in a traditional Chinese medicine preparation process and aim to solve the problems that judgment is inaccurate when the frying degree is judged manually in the traditional Chinese medicine frying process, and all fried medicinal materials are difficult to judge, and time and labor are wasted. The invention relates to a Chinese medicinal material frying process monitoring device, which comprises a camera, a computer and a bracket, wherein the camera is arranged on the bracket; the camera is connected with a USB interface on the computer host through a data cable, and the camera is arranged on the bracket. According to the invention, the image of the fried traditional Chinese medicinal material is input into the computer through the camera, the training is carried out by utilizing the cascade mask rcnn network architecture, the fried degree of the traditional Chinese medicinal material in the frying process is identified and judged, the sampling detection of workers for multiple rounds is not needed, the frying degree is monitored in real time, the accuracy is more than 95%, the detection is far higher than that of manual detection, and the production benefit is improved.
Description
Technical Field
The invention relates to a visual identification device and a visual identification method, in particular to a traditional Chinese medicine stir-frying process monitoring device and a traditional Chinese medicine stir-frying judgment method.
Background
The existing medicinal material stir-frying condition is judged as manual judgment, experienced workers extract a part of medicinal materials from the medicinal materials stir-fried for a certain time in a sampling detection mode, and whether the medicinal materials are stir-fried and the stir-frying degree are judged according to experience and standard comparison by observing the condition change of the medicinal materials. The manual judgment accuracy is not enough, the sampling detection is easy to approximate, the judgment of all stir-fried medicinal materials is not easy to realize, time and labor are wasted, and the speed is not fast enough.
Disclosure of Invention
The invention provides a traditional Chinese medicine frying process monitoring device and a traditional Chinese medicine frying judging method, aiming at solving the problems that the judgment is inaccurate when the frying degree is judged manually in the traditional Chinese medicine frying process, the judgment of all fried medicinal materials is difficult, and time and labor are wasted.
The technical scheme adopted by the invention for solving the technical problems is as follows:
the first scheme is as follows: a Chinese medicinal material frying process monitoring device comprises a camera, a computer and a support;
the camera is connected with a USB interface on the computer host through a data cable, and the camera is arranged on the bracket.
Further, the support is provided with the guide rail horizontally, the guide rail is provided with a sliding block in a sliding mode, the sliding block is provided with a connecting portion in a rotating mode, the connecting portion rotate in the vertical direction, and the connecting portion is provided with a camera.
Furthermore, the device also comprises a spotlight, and the camera is fixed on one side of the spotlight.
Further, the shot-light swing sets up on connecting portion, and the shot-light swings along the horizontal direction.
Further, still include weighing platform, weighing sensor and display screen, weighing sensor is fixed to be established on the support, is equipped with weighing platform on the weighing sensor, and the data line on the weighing sensor passes through serial ports switching TTL interface and the USB interface connection on the host computer, and the display screen passes through the data line and is connected with the VGA port on the host computer.
Scheme II: a Chinese medicinal material stir-frying judgment method comprises the following steps:
s1, acquiring a plurality of images of the fried traditional Chinese medicine through a camera;
s2, labeling the image by using labelme open source software;
s3, inputting the marked image into a cascade mask rcnn network architecture for training to generate a regression model;
s4, inputting a plurality of images to the regression model in the frying process, setting the frying numerical value of the Chinese medicinal materials, and judging the frying degree; if the fried quantity of the Chinese medicinal materials is less than the set value, the frying is not finished and the frying is continued; if the stir-frying quantity of the Chinese medicinal materials is more than or equal to the set value, reminding to stop stir-frying.
Compared with the prior art, the invention has the following beneficial effects:
according to the invention, images of fried traditional Chinese medicinal materials are input into a computer through a camera, labeled by labelme open source software and trained in a cascade maskrnn network architecture to generate a regression model, the frying degree of the traditional Chinese medicinal materials in the frying process is identified and judged, and when the number of the frying degree is greater than or equal to a set value, the traditional Chinese medicinal materials are fried; on the contrary, the stir-frying is continued without stir-frying. According to the invention, workers do not need to perform sampling detection for multiple times, the stir-frying degree is monitored in real time, the accuracy is over 95 percent and is far higher than that of manual detection, and the production benefit is improved.
Drawings
FIG. 1 is an isometric view of a CNN-based Chinese medicinal material stir-frying process monitoring device of the present invention;
fig. 2 is a front view of the monitoring device for the traditional Chinese medicine stir-frying process based on CNN.
Detailed Description
The invention will be described in further detail below with reference to the accompanying drawings: the present embodiment is implemented on the premise of the technical solution of the present invention, and a detailed implementation is given, but the scope of the present invention is not limited to the following embodiments.
Example 1: the embodiment is described below with reference to fig. 1 and fig. 2, and the embodiment relates to a monitoring device for a traditional Chinese medicine frying process, which comprises a camera 1, a computer and a bracket 2;
the camera 1 is connected with a USB interface on a computer host through a data cable, and the camera 1 is arranged on the bracket 2.
In order to adjust the shooting position of the camera 1, a guide rail 3 is horizontally arranged on the support 2, a sliding block is arranged on the guide rail 3 in a sliding manner, a connecting portion 4 is arranged on the sliding block in a rotating manner, the connecting portion 4 rotates along the vertical direction, and the camera 1 is arranged on the connecting portion 4. The position of the camera 1 is adjusted through the transverse adjustment of the connecting part 4, so that the lens of the camera corresponds to the traditional Chinese medicine herb stir-frying machine.
Optionally, the camera 1 is an industrial camera.
In order to make the camera 1 clearly shoot images, the brightness of light needs to be adjusted, the embodiment further comprises a spotlight 5, and the camera 1 is fixed on one side of the spotlight 5.
In order to adjust the irradiation angle of the camera 1 and the light, the spotlight 5 is swing-mounted on the connection portion 4, and the spotlight 5 swings in the horizontal direction. Optionally, two sides of the spot light 5 are respectively fixedly connected with the connecting portion 4 through screws, the connecting portion 4 is provided with an arc-shaped long hole, a screw fixedly connected with the connecting portion 4 is arranged in the arc-shaped long hole, the screw is loosened to adjust the vertical swing angle of the spot light 5, and the screw is locked and fixed.
For the convenience of weighing the weight around the chinese-medicinal material, it is the same with stir-fry system back before guaranteeing that the chinese-medicinal material is fried to be made, monitor around the chinese-medicinal material is fried to prevent to fry system process problem, this embodiment still includes weighing platform 7, weighing sensor 8 and display screen 9, weighing sensor 8 is fixed to be established on support 2, be equipped with weighing platform 7 on the weighing sensor 8, the data line on the weighing sensor 8 passes through serial ports switching TTL interface and the USB interface connection on the host computer, display screen 9 passes through the data line and is connected with the VGA port on the host computer. The Chinese herbal medicines in the frying process are placed on a weighing platform 7 to be weighed, a weighing sensor generates signals and transmits the signals to a computer, and the computer outputs the signals to a display to enable a person to directly see the weighing result; meanwhile, the camera 1 shoots to judge whether the traditional Chinese medicinal materials are fried thoroughly, and the herbal medicine frying machine is closed when the traditional Chinese medicinal materials are fried thoroughly.
Example 2: the embodiment relates to a Chinese herbal medicine stir-frying judgment method, which comprises the following steps:
s1, acquiring a plurality of images of the fried traditional Chinese medicine through the camera 1: from the beginning of frying the medicinal materials to the end of frying, acquiring images with different frying degrees to train the machine, so that the machine can accurately identify the medicinal materials with different frying degrees; optionally, the number of the shot images of the traditional Chinese medicinal materials is at least 10000;
s2, labeling the image by using labelme open source software;
s3, inputting the marked image into a cascade mask rcnn network architecture for training to generate a regression model; optionally, the annotated images are grouped in 100 sheets;
s4, inputting a plurality of images to the regression model in the frying process, setting the frying numerical value of the Chinese medicinal materials, and judging the frying degree; if the fried quantity of the Chinese medicinal materials is less than the set value, the frying is not finished and the frying is continued; if the stir-frying quantity of the Chinese medicinal materials is more than or equal to the set value, reminding to stop stir-frying.
Preferably, the method for labeling the image by using labelme open source software in step S2 specifically includes:
a1, transmitting the images shot in the frying process into a computer, then opening labelme Open source software, clicking an Open Dir button, selecting a folder where the images to be marked are located, and loading all the images;
a2, clicking a button 'Create Plygons' to label the image along the edge of the medicinal material decoction piece; marking three states of immature, mature and coke according to the judgment condition of the medicinal material decoction pieces by manual judgment, wherein the numerical value interval of an immature degree label is [ a, b ], the numerical value interval of a mature degree label is (b, c), the numerical value interval of a coke label is (c, d), wherein a is more than or equal to 0 and less than b and less than or equal to 1;
a3, after the current Image is labeled, clicking a Next Image button, entering a Next Image interface and generating a corresponding json file, wherein the json file stores the corresponding information of the label until the last Image is labeled, and the labeling of the Image is finished.
Further, the method for inputting the image labeled in step S3 into the cascade mask rcnn network architecture for training includes:
a. reading in an image and a corresponding stir-frying degree label, wherein the value of the stir-frying degree label is [0-1], and the sampling interval is 0.1;
b. normalizing the image, i.e. dividing the image pixel values by 255;
c. randomly extracting 20 regions with the size of 512 multiplied by 512 from each image to form a training data set, and randomly scrambling the data set;
d. and inputting the training data set into a resnet50 network architecture for regression training learning to generate a regression model.
Further, the resnet50 network structure is:
where convb denotes the convolutional layer, stride denotes the step size, max pool denotes the maximum pooling, average pool denotes the average pooling, fc denotes the fully-connected layer, and 1-d denotes the final output 1 dimension.
While the invention has been described in detail and with reference to specific examples thereof, it will be understood by those skilled in the art that the foregoing examples are for the purpose of illustration only and are not intended to limit the scope of the invention. It will be appreciated by those skilled in the art that modifications may be made to the above embodiments without departing from the scope and spirit of the invention. The scope of the invention is defined by the appended claims.
Claims (6)
1. Traditional chinese medicinal material system process monitoring devices is fried, its characterized in that: comprises a camera (1), a computer and a bracket (2);
the camera (1) is connected with a USB interface on the computer host through a data cable, and the camera (1) is arranged on the bracket (2).
2. The device for monitoring the frying process of Chinese herbal medicines as claimed in claim 1, wherein: the camera is characterized in that a guide rail (3) is horizontally arranged on the support (2), a sliding block is arranged on the guide rail (3) in a sliding mode, a connecting portion (4) is arranged on the sliding block in a rotating mode, the connecting portion (4) rotates in the vertical direction, and a camera (1) is arranged on the connecting portion (4).
3. The device for monitoring the frying process of Chinese herbal medicines as claimed in claim 2, wherein: the device also comprises a spotlight (5), and the camera (1) is fixed on one side of the spotlight (5).
4. The device for monitoring the frying process of Chinese herbal medicines as claimed in claim 2, wherein: the spotlight (5) is arranged on the connecting part (4) in a swinging mode, and the spotlight (5) swings in the horizontal direction.
5. The device for monitoring the frying process of Chinese herbal medicines as claimed in claim 1, wherein: still include weighing platform (7), weighing sensor (8) and display screen (9), weighing sensor (8) are fixed to be established on support (2), are equipped with weighing platform (7) on weighing sensor (8), and the data line on weighing sensor (8) passes through serial ports switching TTL interface and the USB interface connection on the computer host computer, and display screen (9) pass through the VGA port connection on data line and the computer host computer.
6. A Chinese herbal medicine frying judgment method by using the Chinese herbal medicine frying process monitoring device of any one of claims 1 to 5, characterized by comprising the following steps:
s1, acquiring a plurality of images of the fried traditional Chinese medicinal materials by the camera (1);
s2, labeling the image with labelme open source software in the computer:
s3, inputting the marked image into a cascade mask rcnn network architecture for training to generate a regression model;
s4, inputting a plurality of images to the regression model in the frying process, setting the frying numerical value of the Chinese medicinal materials, and judging the frying degree; if the fried quantity of the Chinese medicinal materials is less than the set value, the frying is not finished and the frying is continued; if the stir-frying quantity of the Chinese medicinal materials is more than or equal to the set value, reminding to stop stir-frying.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110000207.9A CN112633410A (en) | 2021-01-02 | 2021-01-02 | Chinese herbal medicine frying process monitoring device and Chinese herbal medicine frying judgment method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110000207.9A CN112633410A (en) | 2021-01-02 | 2021-01-02 | Chinese herbal medicine frying process monitoring device and Chinese herbal medicine frying judgment method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112633410A true CN112633410A (en) | 2021-04-09 |
Family
ID=75291454
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110000207.9A Pending CN112633410A (en) | 2021-01-02 | 2021-01-02 | Chinese herbal medicine frying process monitoring device and Chinese herbal medicine frying judgment method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112633410A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118090634A (en) * | 2024-04-17 | 2024-05-28 | 北京崇光药业有限公司 | Drug stir-frying method capable of automatically identifying maturity and drug stir-frying machine |
-
2021
- 2021-01-02 CN CN202110000207.9A patent/CN112633410A/en active Pending
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118090634A (en) * | 2024-04-17 | 2024-05-28 | 北京崇光药业有限公司 | Drug stir-frying method capable of automatically identifying maturity and drug stir-frying machine |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109239102A (en) | A kind of flexible circuit board open defect detection method based on CNN | |
CN108596046A (en) | A kind of cell detection method of counting and system based on deep learning | |
Hu et al. | Deep neural network-based speaker-aware information logging for augmentative and alternative communication | |
Wang et al. | Investigation into recognition algorithm of helmet violation based on YOLOv5-CBAM-DCN | |
CN108686978A (en) | The method for sorting and system of fruit classification and color and luster based on ARM | |
CN107092906A (en) | A kind of Chinese traditional medicinal materials recognition device based on deep learning | |
AU2012328140A1 (en) | System and method for identifying eye conditions | |
CN109638959B (en) | Power equipment remote signaling function debugging method and system based on AR and deep learning | |
CN110533654A (en) | The method for detecting abnormality and device of components | |
WO2024051597A1 (en) | Standard pull-up counting method, and system and storage medium therefor | |
CN108182746A (en) | Control system, method and apparatus | |
CN112633410A (en) | Chinese herbal medicine frying process monitoring device and Chinese herbal medicine frying judgment method | |
CN108107226A (en) | AOI multistation system and methods based on PPC softwares | |
CN109997199A (en) | Tuberculosis inspection method based on deep learning | |
CN115661580A (en) | Convolutional neural network-based traditional Chinese medicine decoction piece image identification method and system | |
CN113239805A (en) | Mask wearing identification method based on MTCNN | |
CN102879404A (en) | System for automatically detecting medical capsule defects in industrial structure scene | |
CN209625346U (en) | A kind of Chinese medicine pharmacy intelligently makes up the prescription check system | |
CN106203525A (en) | Electronic equipment and the image processing method of application thereof and system | |
CN104091557B (en) | A kind of display screen based on image recognition technology test device and method of testing thereof | |
CN103777754B (en) | Hand motion tracking device and method based on binocular infrared vision | |
Huang et al. | Image-recognition-based system for precise hand function evaluation | |
CN109215060B (en) | Weight identification method and system for strength type fitness equipment | |
CN115860591A (en) | AI examination scoring method and system for experimental operation | |
CN109242778A (en) | A kind of picture mark control method, device and its storage medium |
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 |