CN114255334B - Shape feature acquisition device, database and identification system for traditional Chinese medicine - Google Patents
Shape feature acquisition device, database and identification system for traditional Chinese medicine Download PDFInfo
- Publication number
- CN114255334B CN114255334B CN202111521236.6A CN202111521236A CN114255334B CN 114255334 B CN114255334 B CN 114255334B CN 202111521236 A CN202111521236 A CN 202111521236A CN 114255334 B CN114255334 B CN 114255334B
- Authority
- CN
- China
- Prior art keywords
- chinese medicine
- traditional chinese
- shape
- standard sample
- identification
- 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
Images
Classifications
-
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06395—Quality analysis or management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/67—Focus control based on electronic image sensor signals
Abstract
The application belongs to the technical field of traditional Chinese medicine quality identification, and particularly relates to a shape characteristic acquisition device, a database and an identification system for traditional Chinese medicines, wherein the traditional Chinese medicine identification system comprises: a database of shape characteristics; the identification unit is used for comparing and identifying at least one shape feature of the traditional Chinese medicine to be identified with the corresponding shape feature in the shape feature database and outputting a comparison identification result; and the identification result display unit is used for displaying the comparison identification result of the identification unit. The traditional Chinese medicine identification system is based on a machine vision technology and combines an artificial intelligence technology to finally generate a standard drawing base of traditional Chinese medicine varieties, so that the standardization of traditional Chinese medicine morphological feature identification is realized, the system is an innovative application of intellectualization and standardization on traditional Chinese medicine morphological identification, and has important functions and significance for promoting the accuracy of traditional Chinese medicine identification, improving the quality of traditional Chinese medicine decoction pieces and ensuring the clinical curative effect.
Description
Technical Field
The application belongs to the technical field of traditional Chinese medicine quality identification, and particularly relates to a traditional Chinese medicine shape characteristic acquisition device, a traditional Chinese medicine shape characteristic database construction method, a traditional Chinese medicine shape characteristic database and a traditional Chinese medicine identification system.
Background
In recent years, the nation pays more and more attention to the development of traditional Chinese medicine, and comprehensively promotes the high-quality development of traditional Chinese medicine from the aspects of perfecting the management mechanism of traditional Chinese medicine, strengthening the traditional Chinese medicine service system, improving the quality and industrial level of traditional Chinese medicine, strengthening the construction of talent team of traditional Chinese medicine, promoting the inheritance and innovation of traditional Chinese medicine, and the like.
The identification of Chinese medicine (or called Chinese medicine identification) is used as key and basis for promoting and ensuring high-quality development of Chinese medicine industry, and its main methods include character identification method, microscopic identification method, physicochemical identification method and biological identification method. The character identification method is to comprehensively judge the authenticity of the traditional Chinese medicine based on the character characteristics caused by various factors such as the variety, cultivation or wild, growth period, production place processing mode, processing method and the like of the traditional Chinese medicine or decoction pieces.
However, the conventional trait identification methods have at least the following problems:
1) the character identification characteristics of the traditional Chinese medicinal materials or decoction pieces can be roughly divided into shape, color, smell, taste and quality, most records related to the identification characteristics are written in characters, and the judgment scale is larger, so that certain limitations exist;
2) the identification expert mainly evaluates the authenticity of the sample according to personal experience, except that the size of the individual can be measured, other index limits are not clear, the judgment subjectivity is strong, and particularly, the judgment results of varieties with complex sources are easy to be inconsistent;
3) the appraisal experts need to have certain working years and abundant professional reviews after long-term experience accumulation, so the time cost for cultivating a qualified traditional Chinese medicine appraisal expert is higher, and the current traditional Chinese medicine practitioner is seriously short.
Therefore, the intelligent standardization of Chinese medicine identification is particularly urgent for the sustainable development of the Chinese medicine industry.
Disclosure of Invention
In order to solve at least one technical problem in the prior art, the application provides a shape feature acquisition device for traditional Chinese medicines, a traditional Chinese medicine shape feature database construction method, a traditional Chinese medicine shape feature database and a traditional Chinese medicine identification system.
In a first aspect, the application discloses a traditional chinese medicine shape feature collection device, includes:
the top of the objective table is provided with an objective plate used for bearing at least one traditional Chinese medicine standard sample, and the bottom of the objective table comprises a multi-degree-of-freedom moving mechanism used for supporting the objective plate and controllably driving the objective plate to move along multiple directions;
the camera is fixed independently of the carrying plate, is positioned right above the traditional Chinese medicine standard sample and is used for photographing the traditional Chinese medicine standard sample from right above;
the light source is arranged around the traditional Chinese medicine standard sample and used for providing illumination when the camera takes a picture;
the objective table control unit is used for responding to a shooting angle instruction received by the traditional Chinese medicine standard sample on the objective plate during shooting, and automatically controlling the multi-degree-of-freedom moving mechanism to drive the objective plate and the traditional Chinese medicine standard sample on the objective plate to move;
the camera control unit is used for responding to a definition instruction received when the traditional Chinese medicine standard sample on the carrying plate is shot and controlling the camera to carry out automatic focusing so as to realize automatic acquisition of an image of the traditional Chinese medicine standard sample with shape characteristics;
the storage unit is used for storing the traditional Chinese medicine standard sample image which is acquired by the camera and comprises shape characteristics;
and the processing unit is used for generating a corresponding shooting angle instruction and a corresponding definition instruction according to the shape characteristics of the pre-collected traditional Chinese medicine standard sample, sending the instructions to the objective table control unit and the camera control unit, and simultaneously, controlling the camera to transmit the collected traditional Chinese medicine standard sample image including the shape characteristics to the storage unit for storage.
According to at least one embodiment of the application, the light sources comprise independently controlled:
the bottom light source is in a straight plate shape and is laid on the carrying plate in parallel, wherein the traditional Chinese medicine standard sample is correspondingly fixed on the top surface of the bottom light source;
the top light source is arranged on the top of the bottom light source in parallel, and a through hole is formed in the position, corresponding to the traditional Chinese medicine standard sample, of the top light source in a penetrating mode, so that the camera can penetrate through the through hole to photograph the traditional Chinese medicine standard sample from the right top;
the two side surface light sources are arranged between the bottom light source and the top light source and are symmetrically arranged along the left side and the right side of the traditional Chinese medicine standard sample;
correspondingly, the traditional Chinese medicine shape characteristic acquisition device further comprises:
the light source control unit is used for responding to an illumination intensity instruction received by the traditional Chinese medicine standard sample on the carrying plate during shooting and automatically controlling the illumination intensity of the corresponding light source;
correspondingly, the processing unit is further used for generating a corresponding illumination intensity instruction according to the shape characteristics of the pre-collected traditional Chinese medicine standard sample, and sending the instruction to the light source control unit.
According to at least one embodiment of the present application, the multi-degree-of-freedom moving mechanism is a three-degree-of-freedom moving mechanism, and can move along the x axis or the y axis in the same two-dimensional coordinate system to realize translation of the object carrying plate, and can rotate by using the x axis as a rotation axis to realize rotation of the object carrying plate.
According to at least one embodiment of the present application, the shape characteristics of the pre-collected chinese medicine standard sample at least include a plane size, a thickness, a shape, a surface characteristic, a cross-sectional texture, a color, and a luster of the chinese medicine standard sample.
In a second aspect, the application also discloses a method for constructing a database of shape and characteristics of traditional Chinese medicines, which comprises the following steps:
step one, acquiring a traditional Chinese medicine standard sample image of various traditional Chinese medicines collected and stored by the traditional Chinese medicine shape characteristic collecting device in any one of the first aspect;
secondly, positioning, measuring and calibrating the shape characteristics in the traditional Chinese medicine standard sample image of any traditional Chinese medicine by a traditional Chinese medicine shape identification method, defining the successfully calibrated shape characteristics as shape identification characteristic points, and teaching each shape identification characteristic point;
step three, deep learning is carried out on each shape recognition feature point taught in the step two;
and step four, processing the traditional Chinese medicine standard sample images of all the traditional Chinese medicines in the step one by using the method of the step two and the method of the step three to obtain shape recognition characteristic data with various traditional Chinese medicine standard sample images being perfect, so as to obtain a traditional Chinese medicine shape characteristic database.
According to at least one embodiment of the present application, in the first step, at least 100 different images of the standard Chinese medicine sample are collected and stored for any Chinese medicine; and
in the second step, the shape characteristics of at least 100 different Chinese medicine standard sample images of any Chinese medicine need to be correspondingly processed.
According to at least one embodiment of the present application, in the third step, when performing deep learning on each shape recognition feature point of each traditional Chinese medicine standard sample image, the adopted learning sample data is classified at least as follows:
time of harvest, cutting mode, place of production, wild or family species and counterfeits.
In a third aspect, the application also discloses a traditional Chinese medicine shape feature database, which is constructed by the traditional Chinese medicine shape feature database construction method in any one of the second aspect.
In a fourth aspect, the present application further discloses a system for identifying traditional Chinese medicine, comprising:
a shape feature database using the database of the shape features of the chinese medicine according to the third aspect;
the identification unit is used for comparing and identifying at least one shape feature of the traditional Chinese medicine to be identified with the corresponding shape feature of the corresponding traditional Chinese medicine in the shape feature database and outputting a comparison identification result;
and the identification result display unit is used for displaying the comparison identification result of the identification unit.
According to at least one embodiment of the application, when the identification unit performs comparison identification, if a certain shape feature of the traditional Chinese medicine to be identified is the same as the corresponding shape feature comparison identification in the shape feature database, outputting a yes result and displaying the identification similarity percentage; if not, outputting a 'no' result; and
and in the identification result display unit, displaying the traditional Chinese medicine to be identified, the shape characteristics of the traditional Chinese medicine to be identified and the 'yes' or 'no' in the identification result.
The application has at least the following beneficial technical effects:
1) the traditional Chinese medicine shape feature acquisition device can enable shape feature acquisition to be highly automatic and intelligent through controlling multi-degree-of-freedom movement of the objective table and automatic exposure of the camera, is not influenced by the placement position, the posture, the presence or absence of superposition and the like of traditional Chinese medicines, and can acquire accurate and clear shape features;
2) according to the traditional Chinese medicine shape characteristic acquisition device, the light sources are distributed around the traditional Chinese medicine standard sample, and the intensity of the light sources around the traditional Chinese medicine standard sample can be automatically adjusted, so that the automation and the intellectualization of the acquisition device are further improved, and image data with higher quality are acquired;
3) the image data of the traditional Chinese medicine standard sample with accurate and clear shape characteristics is obtained based on the traditional Chinese medicine shape characteristic acquisition device, and after a traditional Chinese medicine shape characteristic database (or called a standard image library) is constructed, personnel who have requirements on traditional Chinese medicine identification are facilitated, and no matter professional personnel or non-professional personnel, the system can be used for comparing and identifying the detected traditional Chinese medicine and the standard image library, so that the authenticity and quality of the detected variety are identified in the aspect of form;
4) the construction method of the traditional Chinese medicine variety standard image library is a mode of truly simulating an expert identifying traditional Chinese medicine in the traditional Chinese medicine identification aspect, and is actually a process of carrying out identification on the identified traditional Chinese medicine by an old expert with abundant experience, and the characteristics of the identified traditional Chinese medicine in the aspects of shape, color and the like are rapidly compared with the characteristic library in the brain of the expert, so that the result is identified; therefore, the construction method is an intelligent and standardized innovation in the traditional Chinese medicine form identification, and solves the problems that the quality of the traditional Chinese medicine cannot be effectively maintained and improved due to the fact that the traditional Chinese medicine identification relies on a manual identification mode, has no unified standard, is low in speed, low in efficiency and low in accuracy;
5) the traditional Chinese medicine identification system is based on a machine vision technology, acquires images of all traditional Chinese medicine shape identification characteristic points (namely shape characteristics), combines an artificial intelligence technology, performs teaching and deep learning on the shape identification characteristic points of different traditional Chinese medicines, finally generates a traditional Chinese medicine variety standard image library, and generates traditional Chinese medicine shape characteristic identification standardization by continuously verifying the identification efficiency and accuracy of the equipment system and utilizing the artificial intelligence technology; the method is innovative application of intellectualization and standardization on Chinese medicine form identification, and has important functions and significance for promoting Chinese medicine identification accuracy, improving Chinese medicine quality and ensuring clinical curative effect;
6) the application range of the traditional Chinese medicine identification system can comprise traditional Chinese medicinal materials and traditional Chinese medicine decoction pieces, and the identification range can comprise two directions of authenticity and quality;
7) the identification result of the traditional Chinese medicine identification system is judged by yes or no, so that the traditional Chinese medicine identification system is more accurate.
Drawings
Fig. 1 is a schematic structural diagram of a traditional Chinese medicine shape feature acquisition device of the present application;
wherein:
a carrying plate-11 and a multi-degree-of-freedom moving mechanism-12;
a camera-2;
a bottom light source-31, a top light source-32, a side light source-33;
a PC machine-4;
FIG. 2 is a schematic structural diagram of a multi-degree-of-freedom moving mechanism part in the traditional Chinese medicine shape feature collecting device of the present application;
wherein:
an X-axis adjusting mechanism-51 of the object carrying plate; a Y-axis adjusting mechanism-52 of the object carrying plate; a left and right rotating mechanism-53 for the loading plate; a loading rectangular plate-54; a carrier plate rotation shaft 55; a support slide-56; a fixing plate-57;
fig. 3 is a flowchart of a method for constructing a database of shape and characteristics of traditional Chinese medicine according to the present application.
Detailed Description
In order to make the implementation objects, technical solutions and advantages of the present application clearer, the technical solutions in the embodiments of the present application will be described in more detail below with reference to the drawings in the embodiments of the present application. The described embodiments are a subset of the embodiments in the present application and not all embodiments in the present application. The embodiments described below with reference to the drawings are exemplary and intended to be used for explaining the present application and should not be construed as limiting the present application.
The device for collecting shape characteristics of Chinese herbs, the method for constructing the database of shape characteristics of Chinese herbs, the database of shape characteristics of Chinese herbs and the system for identifying Chinese herbs according to the present application will be further described with reference to the accompanying drawings 1 to 3.
In a first aspect, the application discloses a traditional Chinese medicine shape feature acquisition device; as shown in fig. 1, the apparatus for acquiring shape and characteristics of traditional Chinese medicine comprises a stage, a camera 2, a light source, a PC 4 (at least comprising a host and a display), and the like.
Specifically, the top of the object stage is provided with an object plate 11, and generally, the upper surface of the object plate 11 is used for bearing at least one traditional Chinese medicine standard sample; in addition, the bottom of the stage comprises a multi-degree-of-freedom moving mechanism 12, and the multi-degree-of-freedom moving mechanism 12 is used for supporting the object plate 1 and controllably driving the object plate 1 to move along multiple directions. It should be noted that the standard traditional Chinese medicine samples carried on the object carrying plate 11 may be multiple standard traditional Chinese medicine samples of the same traditional Chinese medicine, and the multiple standard traditional Chinese medicine samples are randomly placed on the object carrying table during collection, so as to expose all features of the traditional Chinese medicine as much as possible.
The camera 2 can be fixed independently of the loading plate 1 by a corresponding fixing and supporting device (not shown in the figure), and is located right above the traditional Chinese medicine standard sample for taking a picture of the traditional Chinese medicine standard sample from right above. It should be noted that, in order to respond to the control of the subsequent camera control unit and to achieve multi-target shooting better and more clearly, the camera 2 is preferably an industrial camera capable of achieving automatic focusing, which can select a proper lens and size and resolution of the camera sensor according to the need.
The light source is arranged around the traditional Chinese medicine standard sample and is used for providing illumination when the camera 2 takes a picture.
Further, in order to realize the automatic movement of the stage plate 11 and the automatic focusing of the camera 2, the PC 4 of the present application is provided with a plurality of corresponding control software (i.e., control units), specifically including a stage control unit, a camera control unit, a storage unit, and a processing unit; the processing unit is used as a master control unit and is used for generating a corresponding control instruction according to the shape characteristics of the pre-collected traditional Chinese medicine standard sample.
Specifically, the stage control unit is configured to respond to a shooting angle command received when shooting the standard traditional Chinese medicine sample on the object carrying plate 11, and automatically control the multiple degree of freedom moving mechanism 12 to drive the object carrying plate 11 and the standard traditional Chinese medicine sample on the object carrying plate 11 to move.
The camera control unit is used for responding to a definition instruction received when the traditional Chinese medicine standard sample on the carrying plate 11 is shot, and controlling the camera 2 to carry out automatic focusing, so that the automatic collection of the traditional Chinese medicine standard sample image including shape characteristics is realized.
The processing unit is used for generating a corresponding shooting angle instruction and a corresponding definition instruction according to the shape characteristics of the pre-collected traditional Chinese medicine standard sample and sending the instructions to the objective table control unit and the camera control unit; the shape characteristics of the pre-collected traditional Chinese medicine standard sample at least comprise the plane size, thickness, shape, surface characteristics, cross-sectional texture, color and luster of the traditional Chinese medicine standard sample. For example, when the shape characteristic adopted at this time is thickness, the shooting angle is the side of the traditional Chinese medicine standard sample, and after the shooting angle command generated at this time is transmitted to the stage control unit, the multi-degree-of-freedom moving mechanism 12 is controlled to drive the object carrying plate 11 to rotate as much as possible so that the included angle between the upper surface of the object carrying plate 11 and the top camera 2 is the smallest, so that the thickness of the side of the sample can be shot more conveniently, and meanwhile, the generated definition command controls the camera 2 to automatically focus the thickness of the side of the narrow sample.
Further, the processing unit is also used for controlling the camera 2 to transmit the acquired traditional Chinese medicine standard sample image (through a graphic card and a communication interface card) including the shape characteristics to the storage unit for storage for subsequent use.
Further, in the traditional Chinese medicine shape characteristic acquisition device, the light source can be arranged at a plurality of suitable positions around the traditional Chinese medicine standard sample as required; in this embodiment, the light source can include independent control and the bottom light source 31, top light source 32 and the side light source 33 that set up around traditional chinese medicine standard sample to carry out automatic adjustment through the luminous intensity to being located traditional chinese medicine standard sample upper and lower and the different light sources all around, make traditional chinese medicine standard sample's different shape characteristics can be highlighted.
Specifically, as shown in fig. 1, the bottom light source 31 is a straight plate, and is laid in parallel on the carrying plate 11 or integrated on the carrying plate 11 at a suitable position on the upper surface, in which case, the standard traditional Chinese medicine sample may be fixed on the top surface of the bottom light source 31.
Top light source 32 is the straight plate form equally, and its parallel arrangement is at the top of bottom light source 31, and runs through on top light source 32 and corresponds position department with traditional chinese medicine standard sample and seted up the through-hole (the purpose is for camera 2 takes a picture and stays the position) to camera 2 can pass this through-hole and take a picture to traditional chinese medicine standard sample directly over from.
The side light source 33 is in a bar shape, and two side light sources 33 are arranged between the bottom light source 31 and the top light source 32 and symmetrically arranged along the left and right sides of the traditional Chinese medicine standard sample in the embodiment.
Accordingly, the PC 4 of the apparatus for collecting shape and characteristics of traditional Chinese medicine is further provided with a light source control unit for controlling the light source. Specifically, the light source control unit is configured to respond to an illumination intensity instruction received when the traditional Chinese medicine standard sample on the loading plate 11 is photographed, and automatically control the illumination intensity of the corresponding light source. In a similar way, the processing unit at this time can also generate a corresponding illumination intensity instruction according to the shape characteristics of the pre-collected traditional Chinese medicine standard sample, and send the instruction to the light source control unit.
Further, in the apparatus for acquiring shape and characteristics of traditional Chinese medicine of the present application, the multi-degree-of-freedom moving mechanism 12 may select various suitable degrees of freedom as required; in this embodiment, as shown in fig. 2, the multi-degree-of-freedom moving mechanism 12 is a three-degree-of-freedom moving mechanism, and includes an object carrying plate X-axis adjusting mechanism 51, an object carrying plate Y-axis adjusting mechanism 52, and an object carrying plate left-right rotating mechanism 53, so that the object carrying plate 11 can move along the X-axis or Y-axis direction in the same two-dimensional coordinate system to realize translation, and can rotate with the X-axis thereof as a rotation axis to realize oblique rotation.
It should be noted that, in the three-degree-of-freedom moving mechanisms known in the prior art, there are many mechanisms capable of implementing the operations of the left-right translation and turning of the object carrying plate 11 in the present application, which are not illustrated here, and the driving method thereof may also adopt a lead screw and slide rail mechanism or a gear and rack mechanism. In this embodiment, the object carrying plate X-axis adjusting mechanism 51 and the object carrying plate Y-axis adjusting mechanism 52 both adopt a screw-slide rail mechanism as shown in fig. 2, and may include a servo motor, a motor reducer, a screw, and a slide rail, and the screw is driven to rotate by the servo motor and the motor reducer, so as to drive a corresponding mechanism (equivalent to a slider, for example, the object carrying plate X-axis adjusting mechanism 51 located on the slide rail of the object carrying plate Y-axis adjusting mechanism 52) connected to the slide rail to move.
In addition, a supporting slide block 56 is arranged at the top of the carrying plate X-axis adjusting mechanism 51 (capable of moving on a slide rail of the carrying plate X-axis adjusting mechanism 51), two fixing plates 57 perpendicular to the top surface of the slide block are arranged at intervals at the top of the supporting slide block 56, further, the carrying plate left-right rotating mechanism 53 comprises an H-shaped carrying plate rotating support frame, aluminum alloy carrying rectangular plates 54 arranged in parallel are arranged at the left side and the right side of the carrying plate rotating support frame, a circular carrying plate rotating shaft 55 is vertically connected between the two carrying rectangular plates 54, and two ends of the carrying plate rotating shaft 55 extend through the fixing plates 57 at the two sides, so that the carrying plate rotating support frame is arranged on the top surface of the supporting slide block 56, and can rotate around the carrying plate rotating shaft 55 in the forward and reverse directions; further, one end of the carrying plate rotating shaft 55 also extends to be rigidly connected with an output shaft of a motor reducer, the motor reducer is connected with a servo motor, and the motor reducer and the servo motor are also fixed on the top surface of the supporting sliding block 56 at the same time, so that the carrying plate rotating supporting frame can be driven by the servo motor to rotate in the positive and negative directions. In this embodiment, it is further preferable that the rotation angle of the front and back directions of the rotating support frame of the loading plate is not more than 15 °.
It can be understood that the traditional manual identification of morphological characteristics of traditional Chinese medicine is mainly based on the characteristics of shape, size, surface characteristics, cross-sectional characteristics, texture, color and glossiness of different decoction pieces. The Chinese herbs can be processed and cut into pieces according to different sources, planting methods, processing and cutting methods, and producing areas. For example, the outline of the external shape of the traditional Chinese medicine is recognized, classified and distinguished, and the side length, the perimeter and the diameter are accurately measured. The surface characteristics comprise the shape of a stem base, the shape and the size of fibrous root marks of a root-stem medicinal material, the surface smoothness, the color and the shape and the size of a skin hole of a skin medicinal material, vein textures, hair and hair of a leaf medicinal material, a receptacle shape of a flower medicinal material, a color of a corolla and the like. Cross section texture shape, color and crack condition, and scale characteristics of the texture. Therefore, the shape feature points need to be identified, positioned and measured, and all the operations cannot be influenced by the placing position, the posture, the existence of superposition and the like of decoction pieces, so that the feature image acquisition system is required to have higher automation and intelligence, can well adapt to the changes, and acquires accurate and clear Chinese medicine identification features.
Therefore, in the traditional Chinese medicine shape feature acquisition device, the shape feature acquisition has higher automation and intelligence through controlling the multi-degree-of-freedom movement of the objective table and the automatic exposure of the camera, is not influenced by the placement position, the posture, the existence of superposition and the like of the traditional Chinese medicines, and can acquire accurate and clear shape features; in addition, the light sources are distributed around the traditional Chinese medicine standard sample, and the intensity of the light sources around the traditional Chinese medicine standard sample can be automatically adjusted, so that the automation and the intellectualization of the acquisition device are further improved, and image data with higher quality can be acquired; furthermore, all the technologies of the application realize control operation through control management software installed in a PC, and through the application of the automation technologies, image acquisition of morphological characteristics of all traditional Chinese medicines can be realized, high-quality image data is obtained, and the image data is automatically transmitted to the PC for data storage.
In the second and third aspects, the application also discloses a traditional Chinese medicine shape characteristic database construction method and a traditional Chinese medicine shape characteristic database obtained by the database construction method.
Specifically, as shown in fig. 3, the method for constructing the database of shape and characteristics of traditional Chinese medicine comprises the following steps:
step one, acquiring a traditional Chinese medicine standard sample image of various traditional Chinese medicines collected and stored by the traditional Chinese medicine shape characteristic collecting device in any one of the first aspect.
And secondly, positioning, measuring and calibrating the shape features in the traditional Chinese medicine standard sample image of any traditional Chinese medicine by a traditional Chinese medicine shape identification method, defining the successfully calibrated shape features as shape identification feature points, and teaching each shape identification feature point.
It should be noted that the traditional Chinese medicine shape identification method belongs to a known mature identification method, wherein the positioning refers to that software identifies a characteristic position; the calibration refers to manually marking the characteristic position and enabling software to learn.
Further, in order to ensure the successful teaching ratio in the step two, in the step one, at least 100 different images of the standard Chinese medicine samples are collected and stored for any one of the Chinese medicines, and then the shape characteristics of at least 100 different images of the standard Chinese medicine samples of any one of the Chinese medicines are processed by the processing method in the step two.
Step three, deep learning is carried out on each shape recognition characteristic point taught in the step two; specifically, deep learning refers to learning morphological feature rules of traditional Chinese medicine, such as size range, color gray scale value range, texture morphology and color.
In addition, in order to improve the identification success rate and the reliability of equipment, when deep learning is carried out on each shape recognition feature point of each traditional Chinese medicine standard sample image, the adopted learning sample data is classified at least according to the following steps:
harvesting time, cutting mode, producing area, wild or family species, counterfeit products and the like.
And step four, processing the traditional Chinese medicine standard sample images of all the traditional Chinese medicines in the step one by using the method of the step two and the method of the step three to obtain shape recognition characteristic data with various traditional Chinese medicine standard sample images being perfect, so as to obtain a traditional Chinese medicine shape characteristic database.
In summary, in the method for constructing the traditional Chinese medicine shape feature database, the traditional Chinese medicine standard sample image data (which can be offline data) with accurate and clear shape features is obtained based on the traditional Chinese medicine shape feature acquisition device, and the traditional Chinese medicine shape feature database (or called as a standard image library) is constructed, so that people who need to identify traditional Chinese medicines can conveniently compare and identify the detected traditional Chinese medicines and the standard image library through the system no matter professionals or non-professionals, and the detected variety is identified in terms of authenticity and quality.
In addition, the construction method of the traditional Chinese medicine variety standard image library is a mode of truly simulating an expert identifying traditional Chinese medicine in the traditional Chinese medicine identification aspect, and is actually a process of carrying out identification on the identified traditional Chinese medicine by an old expert with abundant experience, and the characteristics of the identified traditional Chinese medicine in the aspects of shape, color and the like are rapidly compared with the characteristic library in the brain of the expert, so that the result is identified; therefore, the construction method is an innovation of intellectualization and standardization in the traditional Chinese medicine form identification, and solves the problems that the traditional Chinese medicine quality cannot be effectively maintained and improved due to the fact that the traditional Chinese medicine identification relies on a manual identification mode, and has no unified standard, low speed, low efficiency and low accuracy.
In a fourth aspect, the application also discloses a traditional Chinese medicine identification system, which comprises a shape feature database, an identification unit and an identification result display unit.
Specifically, the shape feature database is the traditional Chinese medicine shape feature database in the third aspect; the identification unit is used for comparing and identifying at least one shape feature of the traditional Chinese medicine to be identified (the image of the identified traditional Chinese medicine can be acquired by the traditional Chinese medicine shape feature acquisition device in the first aspect, or other existing traditional Chinese medicine images) with the corresponding shape feature of the corresponding traditional Chinese medicine in the shape feature database, and outputting the comparison and identification result; and the identification result display unit is used for displaying the comparison identification result of the identification unit.
Further, when the traditional Chinese medicine identification system performs comparison identification in the identification unit, if a certain shape feature of the traditional Chinese medicine to be identified is the same as the corresponding shape feature comparison identification in the shape feature database, outputting a yes result and displaying the identification similarity percentage; if not, outputting a 'no' result; and displaying the traditional Chinese medicine to be identified, the shape characteristics of the traditional Chinese medicine to be identified and the 'yes' or 'no' in the identification result display unit.
In summary, the traditional Chinese medicine identification system of the application is based on a machine vision technology, collects images of all traditional Chinese medicine shape identification feature points (namely shape features), combines an artificial intelligence technology, performs teaching and deep learning on the shape identification feature points of different traditional Chinese medicines, finally generates a traditional Chinese medicine variety standard gallery, and generates traditional Chinese medicine morphological feature identification standardization by continuously verifying the identification efficiency and accuracy of the equipment system and utilizing the artificial intelligence technology; the method is innovative application of intellectualization and standardization on Chinese medicine form identification, and has important functions and significance for promoting Chinese medicine identification accuracy, improving Chinese medicine quality and ensuring clinical curative effect; in addition, the application range of the traditional Chinese medicine identification system can comprise traditional Chinese medicinal materials and traditional Chinese medicine decoction pieces, and the identification range can comprise true and false directions and quality and inferiority directions; and the identification result is judged by yes or no, so that the method is more accurate.
The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present application should be covered within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (6)
1. A traditional Chinese medicine shape characteristic acquisition device is characterized by comprising:
the top of the objective table is provided with an objective plate used for bearing at least one traditional Chinese medicine standard sample, and the bottom of the objective table comprises a multi-degree-of-freedom moving mechanism used for supporting the objective plate and controllably driving the objective plate to move along multiple directions;
the camera is fixed independently of the carrying plate, is positioned right above the traditional Chinese medicine standard sample and is used for photographing the traditional Chinese medicine standard sample from right above;
the light source is arranged around the traditional Chinese medicine standard sample and used for providing illumination when the camera takes a picture;
the objective table control unit is used for responding to a shooting angle instruction received by the traditional Chinese medicine standard sample on the objective plate during shooting, and automatically controlling the multi-degree-of-freedom moving mechanism to drive the objective plate and the traditional Chinese medicine standard sample on the objective plate to move;
the camera control unit is used for responding to a definition instruction received when the traditional Chinese medicine standard sample on the carrying plate is shot and controlling the camera to carry out automatic focusing so as to realize automatic acquisition of an image of the traditional Chinese medicine standard sample with shape characteristics;
the storage unit is used for storing the traditional Chinese medicine standard sample image which is acquired by the camera and comprises shape characteristics;
the processing unit is used for generating a corresponding shooting angle instruction and a corresponding definition instruction according to the shape characteristics of the pre-collected traditional Chinese medicine standard sample, sending the instructions to the objective table control unit and the camera control unit, and meanwhile, controlling the camera to transmit the collected traditional Chinese medicine standard sample image including the shape characteristics to the storage unit for storage;
the light source comprises the following independently controlled:
the bottom light source is in a straight plate shape and is laid on the carrying plate in parallel, wherein the traditional Chinese medicine standard sample is correspondingly fixed on the top surface of the bottom light source;
the top light source is arranged on the top of the bottom light source in parallel, and a through hole is formed in the position, corresponding to the traditional Chinese medicine standard sample, of the top light source in a penetrating mode, so that the camera can penetrate through the through hole to photograph the traditional Chinese medicine standard sample from the right top;
the two side surface light sources are arranged between the bottom light source and the top light source and are symmetrically arranged along the left side and the right side of the traditional Chinese medicine standard sample;
correspondingly, the traditional Chinese medicine shape characteristic acquisition device further comprises:
the light source control unit is used for responding to an illumination intensity instruction received by the traditional Chinese medicine standard sample on the carrying plate during shooting and automatically controlling the illumination intensity of the corresponding light source;
correspondingly, the processing unit is also used for generating a corresponding illumination intensity instruction according to the shape characteristics of the pre-collected traditional Chinese medicine standard sample and sending the instruction to the light source control unit;
the multi-degree-of-freedom moving mechanism is a three-degree-of-freedom moving mechanism and can move along the x-axis or y-axis direction in the same two-dimensional coordinate system to realize translation of the object carrying plate, and meanwhile, the multi-degree-of-freedom moving mechanism can rotate by taking the x-axis as a rotation axis to realize rotation of the object carrying plate;
the shape characteristics of the pre-collected traditional Chinese medicine standard sample at least comprise the plane size, thickness, shape, surface characteristics, cross-section texture, color and luster of the traditional Chinese medicine standard sample.
2. A method for constructing a traditional Chinese medicine shape characteristic database is characterized by comprising the following steps:
acquiring a traditional Chinese medicine standard sample image of various traditional Chinese medicines collected and stored by the traditional Chinese medicine shape characteristic collecting device of claim 1;
secondly, positioning, measuring and calibrating the shape characteristics in the traditional Chinese medicine standard sample image of any traditional Chinese medicine by a traditional Chinese medicine shape identification method, defining the successfully calibrated shape characteristics as shape identification characteristic points, and teaching each shape identification characteristic point;
step three, deep learning is carried out on each shape recognition characteristic point taught in the step two;
step four, processing the traditional Chinese medicine standard sample images of all the traditional Chinese medicines in the step one by using the method of the step two and the method of the step three to obtain shape recognition characteristic data with various traditional Chinese medicine standard sample images being perfect, so as to obtain a traditional Chinese medicine shape characteristic database;
in the first step, at least 100 different images of standard Chinese medicine samples are collected and stored for any Chinese medicine; and
in the second step, the shape characteristics of at least 100 different Chinese medicine standard sample images of any Chinese medicine need to be correspondingly processed.
3. The method for constructing a database of shape and features of chinese herbs according to claim 2, wherein in the third step, when performing deep learning on each shape recognition feature point of each image of a chinese herb standard sample, the learning sample data used is classified at least as follows:
time of harvest, cutting mode, place of production, wild or family species and counterfeits.
4. A database of shape and characteristics of traditional Chinese medicine, which is constructed by the method for constructing the database of shape and characteristics of traditional Chinese medicine according to any one of claims 2 to 3.
5. A traditional Chinese medicine identification system is characterized by comprising the following components:
a shape feature database using the traditional Chinese medicine shape feature database of claim 4;
the identification unit is used for comparing and identifying at least one shape feature of the traditional Chinese medicine to be identified with the corresponding shape feature of the corresponding traditional Chinese medicine in the shape feature database and outputting a comparison identification result;
and the identification result display unit is used for displaying the comparison identification result of the identification unit.
6. The system according to claim 5, wherein when performing comparison recognition in the recognition unit, if a certain shape feature of the Chinese medicine to be recognized is the same as the corresponding shape feature comparison recognition in the shape feature database, a "yes" result is output and a recognition similarity percentage is displayed; if not, outputting a 'no' result; and
and in the identification result display unit, displaying the traditional Chinese medicine to be identified, the shape characteristics of the traditional Chinese medicine to be identified and the 'yes' or 'no' in the identification result.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111521236.6A CN114255334B (en) | 2021-12-13 | 2021-12-13 | Shape feature acquisition device, database and identification system for traditional Chinese medicine |
| CN202211266416.9A CN115700780A (en) | 2021-12-13 | 2021-12-13 | Traditional Chinese medicine shape feature database construction method, database and identification system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111521236.6A CN114255334B (en) | 2021-12-13 | 2021-12-13 | Shape feature acquisition device, database and identification system for traditional Chinese medicine |
Related Child Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211266416.9A Division CN115700780A (en) | 2021-12-13 | 2021-12-13 | Traditional Chinese medicine shape feature database construction method, database and identification system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114255334A CN114255334A (en) | 2022-03-29 |
| CN114255334B true CN114255334B (en) | 2022-08-30 |
Family
ID=80794949
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211266416.9A Pending CN115700780A (en) | 2021-12-13 | 2021-12-13 | Traditional Chinese medicine shape feature database construction method, database and identification system |
| CN202111521236.6A Active CN114255334B (en) | 2021-12-13 | 2021-12-13 | Shape feature acquisition device, database and identification system for traditional Chinese medicine |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211266416.9A Pending CN115700780A (en) | 2021-12-13 | 2021-12-13 | Traditional Chinese medicine shape feature database construction method, database and identification system |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN115700780A (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114841986B (en) * | 2022-05-24 | 2022-11-11 | 亳州学院 | Herbal pieces-quality monitoring analytic system based on machine vision |
| CN116843352B (en) * | 2023-06-26 | 2023-12-19 | 安徽省食品药品检验研究院(安徽国家农副加工食品质量监督检验中心) | Block chain-based traditional Chinese medicine identification method, system, device and storage medium |
| CN116597969B (en) * | 2023-07-14 | 2023-11-10 | 张家港市绿色中药饮片有限公司 | Curing method and system for in-store traditional Chinese medicine decoction pieces |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101266131A (en) * | 2008-04-08 | 2008-09-17 | 长安大学 | Volume measurement device based on image and its measurement method |
| CN103439781A (en) * | 2013-09-06 | 2013-12-11 | 成都西图科技有限公司 | Polarizing microscopic image automatic acquisition and analysis device |
| CN104749180A (en) * | 2015-03-25 | 2015-07-01 | 北京浩辰星月科技有限公司 | System for visually identifying and analyzing as well as identifying drug allergy and drug allergy detecting method |
| CN110850576A (en) * | 2019-10-29 | 2020-02-28 | 北京临近空间飞行器系统工程研究所 | Microscope image scanning control method and device and computer storage medium |
| CN113567356A (en) * | 2021-07-21 | 2021-10-29 | 浙江大学 | Hyperspectral atlas instrument for genuine medicinal material identification and application |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN204154284U (en) * | 2014-06-23 | 2015-02-11 | 汕头大学 | A kind of high-accuracy multiple-degree-of-freedom vision platform |
| CN204536211U (en) * | 2015-03-25 | 2015-08-05 | 北京浩辰星月科技有限公司 | Visual identification is analyzed and susceptibility identification systems |
| CN113358654B (en) * | 2018-07-05 | 2022-07-19 | 深圳辉煌耀强科技有限公司 | Image acquisition and analysis system |
| CN109978018A (en) * | 2019-03-06 | 2019-07-05 | 颐保医疗科技(上海)有限公司 | A kind of Chinese medicine artificial intelligence identification system and its discrimination method |
| CN110609039B (en) * | 2019-09-23 | 2021-09-28 | 上海御微半导体技术有限公司 | Optical detection device and method thereof |
| CN110850020B (en) * | 2019-11-11 | 2022-03-29 | 中国药科大学 | Traditional Chinese medicine identification method based on artificial intelligence |
| CN112115921A (en) * | 2020-09-30 | 2020-12-22 | 北京百度网讯科技有限公司 | True and false identification method and device and electronic equipment |
| CN112861724A (en) * | 2021-02-06 | 2021-05-28 | 安徽理工大学 | Traditional Chinese medicine identification system |
| CN113305885A (en) * | 2021-04-22 | 2021-08-27 | 中国科学院沈阳自动化研究所 | Micro-operation robot system for space standing on-orbit life science experiment |
-
2021
- 2021-12-13 CN CN202211266416.9A patent/CN115700780A/en active Pending
- 2021-12-13 CN CN202111521236.6A patent/CN114255334B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101266131A (en) * | 2008-04-08 | 2008-09-17 | 长安大学 | Volume measurement device based on image and its measurement method |
| CN103439781A (en) * | 2013-09-06 | 2013-12-11 | 成都西图科技有限公司 | Polarizing microscopic image automatic acquisition and analysis device |
| CN104749180A (en) * | 2015-03-25 | 2015-07-01 | 北京浩辰星月科技有限公司 | System for visually identifying and analyzing as well as identifying drug allergy and drug allergy detecting method |
| CN110850576A (en) * | 2019-10-29 | 2020-02-28 | 北京临近空间飞行器系统工程研究所 | Microscope image scanning control method and device and computer storage medium |
| CN113567356A (en) * | 2021-07-21 | 2021-10-29 | 浙江大学 | Hyperspectral atlas instrument for genuine medicinal material identification and application |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115700780A (en) | 2023-02-07 |
| CN114255334A (en) | 2022-03-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114255334B (en) | Shape feature acquisition device, database and identification system for traditional Chinese medicine | |
| EP3467432B1 (en) | Method for operating a device for detecting the three-dimensional geometry of objects | |
| CN109977813A (en) | A kind of crusing robot object localization method based on deep learning frame | |
| CN109906456A (en) | Automation trimming or harvesting system for complicated form branches and leaves | |
| DE602004005496T2 (en) | Localizing teeth in dental images | |
| CN112967285B (en) | Chloasma image recognition method, system and device based on deep learning | |
| JP2019532352A (en) | System for histological examination of tissue specimens | |
| JP4953092B2 (en) | Viable cell discrimination method in cell observation, image processing program for cell observation, and image processing apparatus | |
| CN108846828A (en) | A kind of pathological image target-region locating method and system based on deep learning | |
| CN110287895B (en) | Method for measuring emotion based on convolutional neural network | |
| JP5439096B2 (en) | Bacteria classification device and bacteria test pretreatment device | |
| CN113159227A (en) | Acne image recognition method, system and device based on neural network | |
| CN107209092A (en) | System and method for centre dissection | |
| CN107945156A (en) | A kind of method of automatic Evaluation numeral pathology scan image image quality | |
| CN111855578B (en) | Pathological section scanner | |
| CN104318565B (en) | Interactive method for retinal vessel segmentation based on bidirectional region growing of constant-gradient distance | |
| CN113607659A (en) | Conveyor belt type crop phenotype acquisition method, system and device | |
| CN106706643A (en) | Liver cancer comparison section detection device and detection method | |
| CN112529948A (en) | Mature pomegranate positioning method based on Mask R-CNN and 3-dimensional sphere fitting | |
| RU2726257C1 (en) | Apparatus for automatic measurement of plant parameters | |
| CN114659463B (en) | Plant phenotype acquisition device and acquisition method thereof | |
| CN109241948A (en) | A kind of NC cutting tool visual identity method and device | |
| CN111275754B (en) | Face acne mark proportion calculation method based on deep learning | |
| CN108027968A (en) | information conversion in digital pathology | |
| CN106248634A (en) | Fruit exocuticle glossiness measurement apparatus and method |
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 |