CN112767388A - Wood microspur characteristic image acquisition and AI identification system - Google Patents
Wood microspur characteristic image acquisition and AI identification system Download PDFInfo
- Publication number
- CN112767388A CN112767388A CN202110134276.9A CN202110134276A CN112767388A CN 112767388 A CN112767388 A CN 112767388A CN 202110134276 A CN202110134276 A CN 202110134276A CN 112767388 A CN112767388 A CN 112767388A
- Authority
- CN
- China
- Prior art keywords
- wood
- image
- identification
- macro
- user
- 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.)
- Granted
Links
- 239000002023 wood Substances 0.000 title claims abstract description 248
- 238000000034 method Methods 0.000 claims abstract description 29
- 238000012545 processing Methods 0.000 claims abstract description 26
- 241000894007 species Species 0.000 claims description 33
- 238000012549 training Methods 0.000 claims description 10
- 238000007781 pre-processing Methods 0.000 claims description 8
- 238000012937 correction Methods 0.000 claims description 5
- 230000004044 response Effects 0.000 claims description 5
- 230000006835 compression Effects 0.000 claims description 4
- 238000007906 compression Methods 0.000 claims description 4
- 238000013527 convolutional neural network Methods 0.000 claims description 4
- 235000002566 Capsicum Nutrition 0.000 claims description 3
- 239000006002 Pepper Substances 0.000 claims description 3
- 235000016761 Piper aduncum Nutrition 0.000 claims description 3
- 235000017804 Piper guineense Nutrition 0.000 claims description 3
- 235000008184 Piper nigrum Nutrition 0.000 claims description 3
- 150000003839 salts Chemical class 0.000 claims description 3
- 241001465382 Physalis alkekengi Species 0.000 claims description 2
- 244000203593 Piper nigrum Species 0.000 claims 1
- 238000007726 management method Methods 0.000 description 16
- 230000006870 function Effects 0.000 description 13
- 238000010586 diagram Methods 0.000 description 9
- 230000008569 process Effects 0.000 description 8
- 238000011161 development Methods 0.000 description 6
- 238000003825 pressing Methods 0.000 description 5
- 239000013589 supplement Substances 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000007619 statistical method Methods 0.000 description 4
- 238000003860 storage Methods 0.000 description 4
- WHXSMMKQMYFTQS-UHFFFAOYSA-N Lithium Chemical compound [Li] WHXSMMKQMYFTQS-UHFFFAOYSA-N 0.000 description 3
- 230000000903 blocking effect Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 3
- 238000001914 filtration Methods 0.000 description 3
- 229910052744 lithium Inorganic materials 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 241000238631 Hexapoda Species 0.000 description 2
- 241000722363 Piper Species 0.000 description 2
- 230000003321 amplification Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 238000013079 data visualisation Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000010437 gem Substances 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000003199 nucleic acid amplification method Methods 0.000 description 2
- 230000001502 supplementing effect Effects 0.000 description 2
- 102100040401 DNA topoisomerase 3-alpha Human genes 0.000 description 1
- 101000611068 Homo sapiens DNA topoisomerase 3-alpha Proteins 0.000 description 1
- 238000013475 authorization Methods 0.000 description 1
- 238000005282 brightening Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 235000016709 nutrition Nutrition 0.000 description 1
- 230000035764 nutrition Effects 0.000 description 1
- 238000004806 packaging method and process Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 238000004171 remote diagnosis Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- 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
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
- G06F21/44—Program or device authentication
-
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformation in the plane of the image
- G06T3/40—Scaling the whole image or part thereof
- G06T3/4007—Interpolation-based scaling, e.g. bilinear interpolation
-
- G06T5/80—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20021—Dividing image into blocks, subimages or windows
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30161—Wood; Lumber
Abstract
The invention discloses a wood microspur characteristic image acquisition and AI identification system, which comprises a client, a database, a server and a wood image AI identification module, wherein the client is used for receiving user operation, the database is used for storing user data, the server comprises a security authentication module and a business processing module, the wood image AI identification module is used for identifying a wood image to be identified to obtain wood type information, the client receives a login request and sends the login request to the server, after the server receives the login request, the security authentication module verifies whether the login information of a user is correct or not by comparing the user data in the user database, and the business processing module identifies the wood image through the wood image AI identification module and feeds back an identification result to the client. The invention provides a method for acquiring clearer wood images through the microspur image acquisition terminal, improves the identification accuracy, simplifies the system workload by adopting the load balancer, and has simple operation and convenient use.
Description
Technical Field
The invention relates to the field of wood identification, in particular to a wood microspur characteristic image acquisition and AI identification system.
Background
At present, the wood species identification is mainly performed by observing macroscopic and microscopic characteristics of wood by means of instruments such as a magnifier, a microscope and the like by professional staff in a laboratory environment, and determining the wood species after comparing the macroscopic and microscopic characteristics with a standard sample. Every kind of wood forms unique macro and micro structure in the growing process, but the kinds of wood are various, and even if the same kind of wood, the macro and micro structure have certain difference due to the situations of place, climate, nutrition and the like. The inter-species similarity and intra-species difference phenomena increase the difficulty of wood species identification, and in addition, because the first line of wood trade and supervision in China has the shortage of experts with wood taxonomy knowledge, wood or wood products are frequently traded, the artificial wood species identification method has the problems of strong specialization, heavy task, long period, high risk, non-real-time property and the like, the requirements of real-time property and high efficiency of wood supervision cannot be met, and a rapid and accurate wood species identification method is urgently needed.
Disclosure of Invention
In view of the above, it is necessary to provide a wood macro characteristic image acquisition and AI identification system, which can accurately and quickly complete wood identification, and the technical solution provided by the present invention is as follows:
the invention provides a wood microspur characteristic image acquisition and AI identification system, comprising:
the client is used for receiving user operation;
a database comprising a user database for storing user data;
the server comprises a security authentication module and a service processing module;
the wood image AI identification module is used for identifying the wood image to be identified to obtain the wood species information of the wood;
the client and the database are respectively connected with the server, the client receives a login request and sends the login request to the server, after the server receives the login request, the security authentication module verifies whether the login information of the user is correct or not by comparing the user data in the user database,
if not, the security authentication module rejects the login request and sends a re-login instruction to the client;
if the wood image identification request is correct, the business processing module can receive an image identification request sent by the client, wherein the image identification request comprises the wood image to be identified;
and the business processing module identifies the wood image through a wood image AI identification module and feeds back an identification result to the client.
Further, the system also comprises a wood database, and the business processing module searches the wood database according to the identification result returned by the wood image AI identification module and feeds back the search result to the client.
Furthermore, the system also comprises a macro image acquisition terminal, wherein the macro image acquisition terminal can be arranged on a camera lens of the mobile terminal, so that the camera lens of the mobile terminal can shoot the detail image of the wood to be identified.
Preferably, the macro image capturing terminal is an assembled macro image capturing terminal, the assembled macro image capturing terminal includes a macro lens, an adapter disposed at an outer end of the macro lens, and a capturing head disposed at an outer end of the adapter,
the adapter is inside to be equipped with the confession the perspective chamber of macro lens, the collection head including be formed with the collection base of the annular chamber that the perspective chamber is linked together and set up lamps and lanterns in the collection base, wherein, the collection base the adapter reaches and waits to detect sample surface and form confined image acquisition district, perhaps, the collection base the adapter reaches and places the platform surface that detects the sample and form confined image acquisition district.
Preferably, the macro image acquisition terminal is an embedded macro image acquisition terminal, the embedded macro image acquisition terminal comprises a macro lens, the macro lens comprises a lens base and a lens, the lens base comprises a base body, a press ring detachably arranged at the front end of the base body, and a lens external connection part arranged at the rear end of the base body, wherein the press ring is annular, and a closed image acquisition area is formed by the inner cavity of the base body, the inner cavity of the press ring and the surface of a sample to be detected, or the inner cavity of the base body, the inner cavity of the press ring and a platform surface for placing the sample to be detected form a closed image acquisition area;
the embedded micro-distance image acquisition terminal is characterized in that the embedded micro-distance image acquisition terminal is also arranged on the seat body or a lamp used for compensating exposure and arranged on the inner wall of the compression ring, the lamp is annular, and the central line of the lamp and the central line of the lens are superposed.
Further, the wood image AI recognition module recognizing the image includes the steps of:
s1, inputting a wood image and preprocessing the wood image;
s2, processing the wood image in blocks;
s3, training the sub-image blocks through a convolutional neural network model;
s4, using different gradient values from the edge to the center of the wood image as the weight of the sub-image classification score of different areas, increasing the proportion of the center area in the whole wood image classification score, converting the weighted score into a final probability value, obtaining the wood species information of the wood image, wherein,
the step of preprocessing in S1 includes:
s11, performing color correction on the wood image;
s12, performing data enhancement on the corrected wood image;
the step of block-processing the wood image in S2 includes:
s21, segmenting the preprocessed wood image to obtain a plurality of sub-images;
and S22, unifying the size pixels of each sub-image by a bilinear interpolation method.
Further, step S11 includes color-correcting the wood cross-sectional image by a gray world method, calculating R, G, B three-channel gain coefficients and a gray-scale average value from the whole wood cross-sectional image and R, G, B three-channel pixel expectation values, and performing respective R, G, B three-channel component adjustment on each pixel in the wood cross-sectional image according to the gain coefficients and the gray-scale average value;
step S12 includes performing data enhancement on the corrected wood cross-sectional images by using horizontal inversion, vertical inversion and adding salt and pepper noise so that the training sample size of each wood image is within a prescribed number range.
Furthermore, the system also comprises a load balancer which is arranged between the client and the server, and the data sent by the client is processed by the load balancer and then is uniformly distributed on the server.
Further, the system also comprises an expert identification support module used for sending an expert identification support request, wherein the expert identification support request comprises the wood image to be identified and/or the identification result returned by the wood image AI identification module, and receiving expert identification information returned in response to the expert identification support request.
Further, the database comprises MySQL and/or Oracle and/or SQLite.
The invention has the following advantages:
a. a clearer wood image is obtained through the microspur image acquisition terminal, so that more fine characteristics can be extracted conveniently, wood identification is facilitated, and the wood identification accuracy is improved;
b. the load balancer is adopted to evenly distribute the data to the server, so that the workload of development and deployment is greatly simplified, the wood image identification time is shortened, and the long-term operation of the system is facilitated;
c. the system is simple to operate and convenient to use.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a block diagram of a wood macro characteristic image acquisition and AI identification system according to an embodiment of the present invention;
fig. 2 is a flowchart of the work flow of the Web end of the wood macro characteristic image acquisition and AI identification system according to the embodiment of the present invention;
fig. 3 is a flowchart of a Web-side image recognition process of a wood macro characteristic image acquisition and AI identification system according to an embodiment of the present invention;
fig. 4 is a schematic internal storage diagram of a user database of a wood macro characteristic image acquisition and AI identification system according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of a wood image AI identification module of the wood microspur characteristic image acquisition and AI identification system according to the embodiment of the present invention;
fig. 6 is a schematic structural diagram of an assembled macro image acquisition terminal of the wood macro characteristic image acquisition and AI identification system according to the embodiment of the present invention;
fig. 7 is an enlarged schematic view of an adapter of an assembled macro image acquisition terminal of the wood macro characteristic image acquisition and AI identification system according to the embodiment of the present invention;
fig. 8 is a schematic structural diagram of an embedded macro image acquisition terminal of the wood macro characteristic image acquisition and AI identification system according to the embodiment of the present invention;
fig. 9 is an exploded view of an embedded macro image capturing terminal of the wood macro characteristic image capturing and AI identifying system according to the embodiment of the present invention;
fig. 10 is an exploded schematic view of an embedded macro image acquisition terminal of the wood macro characteristic image acquisition and AI identification system according to the embodiment of the present invention after a compression ring is replaced;
fig. 11 is a schematic top view of a lamp of an embedded macro image capturing terminal of the wood macro characteristic image capturing and AI identifying system according to the embodiment of the present invention.
Wherein the reference numerals include: the lens comprises a 1-macro lens, a 10-lens, a 12-lens holder, a 120-holder body, a 120 a-first split body, a 120 b-second split body, a 121-press ring, a 122-lens external connection part, a 2-adapter, a 21-first connection part, a 22-second connection part, a 3-collecting head, a 30-collecting base, a 31-soft light ring, a 32-lamp, a 320-annular lamp strip, a 321-brightening body, a 322-external power supply connector, a 1-first annular cavity, a 2-second annular cavity, a 3-third annular cavity, b 1-annular clamping groove, a c-charging interface, a q-perspective cavity, a q 1-first cavity, a q 2-second cavity and a k-embedding notch.
Detailed Description
In order to make the technical solutions of the present invention better understood and more clearly understood by those skilled in the art, the technical solutions of the embodiments of the present invention will be described below in detail and completely with reference to the accompanying drawings. It should be noted that the implementations not shown or described in the drawings are in a form known to those of ordinary skill in the art. Additionally, while exemplifications of parameters including particular values may be provided herein, it is to be understood that the parameters need not be exactly equal to the respective values, but may be approximated to the respective values within acceptable error margins or design constraints. It is to be understood that the described embodiments are merely exemplary of a portion of the invention and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention. In addition, the terms "comprises" and "comprising," and any variations thereof, in the description and claims of this invention, are intended to cover a non-exclusive inclusion, such that a process, method, apparatus, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
In an embodiment of the present invention, a wood macro characteristic image collection and AI identification system is provided, as shown in fig. 1, including a client, a database, a server, a wood image AI identification module, and a macro image collection terminal.
The client is used for receiving user operation; the database comprises a user database and a wood database, wherein the user database is used for storing user data, the wood database is used for storing wood information, the server comprises a security authentication module and a business processing module, the security authentication module is used for verifying the user information, and the business processing module is used for processing related businesses; the wood image AI identification module is used for identifying a wood image to be identified to obtain the wood species information of the wood; the macro image acquisition terminal is installed on a camera lens of the mobile terminal and used for shooting a detailed image of wood to be identified.
The client and the database are respectively connected with the server, the client receives a login request and sends the login request to the server, after the server receives the login request, the security authentication module verifies whether the login information of the user is correct or not by comparing the user data in the user database,
if not, the security authentication module rejects the login request and sends a re-login instruction to the client;
if the wood image identification result is correct, the business processing module can receive an image identification request sent by the client, the image identification request comprises a wood image to be identified, the business processing module identifies the wood image through a wood image AI identification module, the wood image AI identification module receives the wood image to be identified, the identification result of the wood image to be identified is obtained through a convolutional neural network and fed back to the business processing module, and the business processing module retrieves the wood database according to the identification result returned by the wood image AI identification module and feeds back the retrieval result to the client.
The wood macro characteristic image acquisition and AI identification system also comprises an expert identification support module used for sending an expert identification support request, wherein the expert identification support request comprises the wood image to be identified and/or the identification result returned by the wood image AI identification module, and receives expert identification information returned in response to the expert identification support request.
In an embodiment of the present invention, as shown in fig. 2, taking the Web end as an example for specific description, the development and deployment of the Web end of the wood macro characteristic image acquisition and AI identification system are as follows:
the system is developed by utilizing HTML, CSS and JavaScript technologies, and a system frame is built by adopting Vue-CLI; designing a system interface by using an Element UI, modularizing functions at the front end, using an ES2015+ standard, being downward compatible with an ES5, and adapting to all mainstream browsers; the Vue Router is used as a routing manager, and functions of nesting of routing/view charts, modularized component configuration, fine-grained control of navigation and the like are supported; the network request uses an Axios network request based on an HTTP protocol, and functions of XMLHttpRequests, request data and response data conversion, JSON data automatic conversion, XSRF attack prevention and the like are supported; the identification information is stored through Token, so that unnecessary interaction information is reduced, and the robustness of the system is improved; and the front end is developed by using mock. js simulation data response, so that the complete separation of the front end development and the back end development and the high decoupling of the whole system are realized.
As shown in fig. 2, after the user logs in successfully, the server returns a corresponding function list and Token according to the user permissions, it should be noted that the user is divided into a management user, an expert user, and a general user, the user permissions are divided into a management user permission, an expert user permission, and a general user permission, the general user permission is relatively minimum, the expert user permission is relatively high, the management user permission includes all permissions, the management user can operate information, permissions, and the like of all users, and other users cannot operate, which is described in detail below.
And when the Web terminal initiates a user request, the Token is encapsulated in the HTTP request header, and the server terminal carries out security and authority verification on the user information through the Token information. During image recognition, a user packages an image into an HTTP request body in a Form of a Form, the server receives the image, returns wood related information, a related inquiry and issue case and customs information after recognition is completed, and renders the information on a webpage for the user to refer. If the user needs to browse the wood information, the Web end sends a related request to the server end, because the quantity of the wood information is large, the information needs to be paged for the user to inquire and browse, pageNum and pageSize fields need to be carried in request parameters and respectively represent the current page number and the number of information items contained in one page, and the server end returns corresponding data to the Web end for the user to browse and refer. If the user has a question about the identification result, the user can ask the expert through an expert online support function, and the question asking mode includes a picture-text mode and an audio-video mode, wherein in the picture-text mode, the user fills in question description and uploads related images, Web end packaging information is uploaded to the server end in an HTTP request body, the server end stores data into the database, the expert gives guidance opinions after checking the question, and the user can check the expert opinions in the question record. If the user needs to obtain expert support in a short time, the audio-video mode can be selected, and the problems can be solved through face-to-face communication with the experts.
In an embodiment of the present invention, when the client and the server perform data transmission, one or more servers have overload faults, so the wood macro characteristic image collection and AI identification system further includes a load balancer disposed between the client and the server, and data sent by the client is processed by the load balancer and then uniformly distributed to the server.
For the server, a Spring frame may be selected, and other frames may also be selected, such as a Struts2 frame, a Wicket frame, a Tapestry frame, and a strips frame, which does not limit the scope of the present invention. In an embodiment of the present invention, as shown in fig. 3, a Spring Boot frame is taken as an example for specific description:
the Spring Boot framework realizes automatic configuration of a plurality of frameworks on the basis of the Spring framework, and meanwhile, the Tomcat server is arranged in the Spring Boot framework, so that the workload of development and deployment is greatly simplified. And the Web end and the server end interact to use a RESTFul format request. And the user request is sent to the server through an HTTP (hyper text transport protocol), the request is firstly subjected to load balancing through Nginx and then is forwarded to the server background in a polling mode. After receiving a user request, the background uses Spring Security first and performs user authentication and authorization by combining JWT. The Spring Security is a Security management framework provided by the Spring framework, and can be used for filtering illegal requests and verifying whether user information is correct, and performing relevant service processing and returning to the Web end after the user is successfully verified. When a user identifies a wood image, the background of the server side receives the wood image, then the image validity is verified, and the wood image is forwarded to the Django server side after no problem is confirmed. The Django server is a Web framework developed based on Python, and the wood macro characteristic image acquisition and AI identification system uses the Django framework to avoid performance loss caused by direct language conversion. The method comprises the steps that after a Django server receives a wood image to be identified, an identification result is obtained by calling a wood image identification model and returned to the Django server, the Django server extracts relevant information after receiving the identification result and inquires the relevant information of the wood from a wood database, the relevant information of the wood comprises information such as a corresponding Latin name, a wood production place and a classification position, the information is packaged and then returned to a Web end in a JSON format, a user can see the identification result at the Web end, and the whole process is about 1-2 seconds.
For the database, one or more of MySQL, Oracle and SQLite may be adopted, and other databases may also be adopted, and the database is selected according to actual requirements, which does not limit the scope of the present invention. In an embodiment of the present invention, as shown in fig. 4, MySQL is taken as an example for specific description:
the data of the wood micro-distance characteristic image acquisition and AI identification system is divided into two parts of user information and wood information.
The user information data storage adopts a RBAC (Role-based Access Control) model, and manages user related information in a Role-associated user and Role-associated authority mode. Fig. 4 shows a structure of a user-related information table in the wood macro characteristic image acquisition and AI identification system, where a sys _ user table stores user information, such as user id, job number, user name, password, etc., a sys _ role table stores role authority information, such as authority name, authority keyword, etc., and a sys _ menu table stores all function interfaces in the system, and the function interfaces are associated with the sys _ role table to determine authority information to which a user belongs.
The storage of the wood information refers to the RBAC model, the wood information table comprises information such as wood names, classification status, production places and the like, the information of the record case and the information of the customs internal information are respectively stored in other two tables, and the wood, the information of the record case and the information of the customs internal information are associated through the two association tables. Meanwhile, Redis is combined as a cache database, hot spot information is stored in Redis, the capability of the wood macro characteristic image acquisition and AI identification system for processing high concurrent user requests can be enhanced, and the system stability is improved.
As for the wood image AI identification module, as shown in fig. 5, the specific steps of the operation are as follows:
s1, inputting a wood image and preprocessing the wood image;
s2, processing the wood image in blocks;
s3, training the sub-image blocks through a convolutional neural network model;
s4, using different gradient values from the edge to the center of the wood image as the weight of the sub-image classification score of different areas, increasing the proportion of the center area in the whole wood image classification score, converting the weighted score into a final probability value, obtaining the wood species information of the wood image, wherein,
the step of preprocessing in S1 includes:
s11, performing color correction on the wood image;
s12, performing data enhancement on the corrected wood image;
the step of block-processing the wood image in S2 includes:
s21, segmenting the preprocessed wood image to obtain a plurality of sub-images;
and S22, unifying the size pixels of each sub-image by a bilinear interpolation method.
Step S11 includes performing color correction on the wood cross-sectional image by a gray world method, calculating a gain coefficient and a gray average value of R, G, B three channels from the entire wood cross-sectional image and a pixel expectation value of R, G, B three channels, and performing adjustment of respective R, G, B three-channel components for each pixel in the wood cross-sectional image according to the gain coefficient and the gray average value.
Step S12 includes performing data enhancement on the corrected wood cross-sectional images by using horizontal inversion, vertical inversion and adding salt and pepper noise so that the training sample size of each wood image is within a prescribed number range.
The preprocessing method is only an example, and the preprocessing method may also use methods such as contrast widening, logarithmic transformation, density layering, histogram equalization, gaussian filtering, and laplacian filtering, and the like, which does not limit the scope of the present invention.
In one embodiment of the present invention, as shown in fig. 5, in a first step, a wood image is preprocessed; secondly, processing the wood image in a 7 x 7 blocking mode, and thirdly, training the image through a ResNet101 convolution neural network model; and fourthly, adopting different gradient values from the edge to the center of the wood image as the weights of the classification scores of the images of different sub-regions, increasing the proportion of the central region in the classification score of the whole wood image, converting the weighted scores of all the categories into final probability values by utilizing a Softmax method, obtaining the probability values of all the categories of the wood by utilizing the Softmax method so as to automatically identify the wood image, and inquiring the information of the wood category of the wood image by combining with the wood database.
It should be noted that the blocking method is only an example, and other blocking methods, such as 5 × 5 and 10 × 10, may also be adopted, and similarly, the training model is also only an example, and other training models, such as VggNet16, GoogleNet, densneet, MobileNetv3, ResNet50, and ResNet152, may also be adopted, without limiting the protection scope of the present invention.
In one embodiment of the invention, a Web end is built by utilizing HTML, CSS, JavaScript technology and Vue-CTI; adopting MySQL to establish a wood database for storing information such as wood information, case searching and issuing, historical record identification and the like; establishing a ResNet 101-based wood identification model, and deploying the model at a server developed based on a Spring Boot framework; the method comprises the steps of calling a deep learning model by using a Django framework, calling a wood recognition model and returning a recognition result when a Spring Boot server sends a wood image recognition request, searching information such as corresponding wood information and related checking and sending cases in a wood database by the Spring Boot server through the recognition result, feeding the information back to a user and displaying the information on a Web end, and storing the recognition result in the database. When a user encounters unknown wood, the wood image is obtained through a microscope or a mobile phone and the like and is guided into a computer, the wood image is uploaded through a Web end, the server feeds back information after the identification is completed and displays the information on the Web end, and the identification process is about 1-2 s. If the user has a question about the recognition result, the expert can be requested to perform remote diagnosis. The wood microspur characteristic image acquisition and AI identification system provides an auxiliary tool for quickly identifying wood species by AI, so that the economic cost and the time cost are greatly saved.
For the macro image acquisition terminal, an assembled macro image acquisition terminal, an embedded macro image acquisition terminal or other macro image acquisition terminals may be adopted, and the macro image acquisition terminal is selected according to actual requirements, which does not limit the protection scope of the present invention. Now, the assembled macro image capturing terminal and the embedded macro image capturing terminal are respectively explained:
(1) assembled macro image acquisition terminal
As shown in fig. 6, the assembled macro image capturing terminal includes a macro lens 1, an adapter 2 connected to an outer end of the macro lens 1, and a capturing head 3 disposed at an outer end of the adapter 2 and capable of compensating exposure.
The collecting head 3 comprises a collecting base 30 formed with an annular cavity, a soft light ring 31 and a lamp 32 which are arranged on the collecting base 30, wherein a closed image collecting area is formed by the adapter 2, the soft light ring 31, the surface of a sample to be detected or the adapter 2, the soft light ring 31 and the surface of a platform for placing the sample to be detected.
The end surface of the collecting base 30 far away from the adapter 2 is arranged in parallel with the lens 10 of the macro lens 1. After the mobile phone is connected to the shooting equipment (mobile phone), the lens of the shooting equipment and the end face of the collecting base far away from the adapter can be ensured to be parallel, and therefore the structure amplification picture of the surface of the sample to be detected is obtained under the same focal length.
The soft light ring 31 is arranged at the end part of the collecting base 30 far away from the adapter 2, and the soft light ring 31 is arranged in parallel with the lens 10 of the macro lens 1 far away from the end surface of the collecting base 30. Therefore, when the fixed-focus picture is taken, the lens is ensured to be parallel to the surface of the sample to be detected, and a necessary condition is provided for obtaining a satisfactory picture.
In one embodiment of the present invention, the interior of the soft light ring 31 is sequentially formed with a first annular cavity a1, a second annular cavity a2, and a third annular cavity a3 formed by extending the second annular cavity a2 from the outer end surface to the outside from the inside to the outside.
The section of the first annular cavity a1 is rectangular, the sections of the second annular cavity a2 and the third annular cavity a3 are isosceles trapezoids, and the soft light ring 31 spans the surface of the sample to be detected or the surface of the platform for placing the sample to be detected from the end face of the third annular cavity a 3. By the arrangement, better illumination conditions are created, and focusing is more accurate when the macro lens takes a picture.
In one embodiment of the present invention, the center lines of the first annular cavity a1, the second annular cavity a2, and the third annular cavity a3 coincide with the center line of the lens 10.
Meanwhile, in one embodiment of the present invention, the inner diameter of the first annular cavity a1 is larger than the outer diameter of the lens 10, and the inner diameter of the inner end of the second annular cavity a2 is larger than the inner diameter of the first annular cavity a 1. The exposure effect is optimal at this time.
The lighting fixture 32 comprises a circular light strip circumferentially arranged along the collecting base 30, a plurality of LED lamps arranged on the circular light strip, and a power supply and a switch, wherein the power supply is a rechargeable lithium battery. The annular light band is arranged, so that lamplight can be compensated in the all-directional collection area, and the photographing is convenient; meanwhile, the lithium battery is arranged, so that the lithium battery can directly supply power to the lamp 32, and the use is more convenient.
It should be noted that the center line of the circular strip coincides with the center line of the lens 10. Therefore, the annular light supplement effect is optimal.
In addition, a charging interface c communicated with a power supply is further arranged on the acquisition base 30. Therefore, the mobile phone can directly supply power to the bright body, and the practicability of the macro image acquisition terminal is further improved.
As shown in fig. 7, the adapter 2 includes a first connecting portion 21 connected to the macro lens 1 and a second connecting portion 22 detachably connected to the capturing base 30, wherein the first connecting portion 21 and the second connecting portion 22 are internally penetrated to form a perspective cavity q, an annular clamping groove b1 is formed at an end portion of the second connecting portion 22 far away from the macro lens 1, and an annular protrusion b2 matched with the annular clamping groove b1 is disposed on an outer periphery of the capturing base 30.
Therefore, the assembled micro-distance image acquisition terminal has the following advantages:
a. the annular light surrounding the camera can provide accurate annular light supplement (no annular light supplement lamp taking the mobile phone camera as the center at present) taking the shot object as the center for shooting the object, and the shadow of the shot object can be greatly eliminated relative to other lateral lights;
b. the light compensation cameras can be flexibly replaced aiming at the multi-camera mobile phone by matching with various fixing clamps;
c. the size is very small, the requirement of high flexibility of macro photography can be met, and the portability is high;
d. the power source is convenient to obtain, and the mobile phone can directly supply power;
e. the macro lens and the collecting head are quickly assembled through the adapter so as to be convenient to carry, and meanwhile, the micro lens can be attached to the surface of an object to shoot a planar microstructure and can also shoot tiny objects (such as insects, jewels, flowers and the like);
f. the assembled micro-distance image acquisition terminal and the micro-distance shooting equipment are matched for use, and the light is supplemented in the image acquisition area by the lamp, so that the focusing distance is shortened under the micro-distance lens, and the lens is stable and has no shadow light supplementing condition during shooting, so that a picture with good effect (such as high definition) can be quickly and accurately obtained, and the assembled micro-distance image acquisition terminal and the micro-distance shooting equipment can be operated by one person, and are very convenient.
(2) Embedded microspur image acquisition terminal
As shown in fig. 8, the in-line macro image capture terminal includes a macro lens 1 and a lamp 32 for compensating exposure.
As shown in fig. 9, the macro lens 1 includes a lens holder 12 and a lens 10, wherein the lens holder 12 includes a holder body 120, a press ring 121 detachably disposed at a front end portion of the holder body 120, and a lens external connection portion 122 disposed at a rear end portion of the holder body 120.
In an embodiment of the present invention, the pressing ring 121 is annular, and the inner cavity of the seat body 120, the inner cavity of the pressing ring 121, and the surface of the sample to be detected form a closed image capturing region.
The seat body 120 includes a first division body 120a and a second division body 120b, in which a first cavity q1 and a second cavity q2 are formed to penetrate.
Specifically, the first cavity q1 is cylindrical, the second cavity q2 is in a frustum shape with gradually increasing diameter from the first cavity forward, and the lens 10 is installed at the connection position of the first cavity q1 and the second cavity q2 in an intercepting manner.
The lens 10 is arranged in parallel with the front end face of the compression ring 121, so that a structure amplification picture of the surface of the sample to be detected can be obtained under the same focal length.
As shown in fig. 10, the pressing ring 121 is detachably connected to the second segment 120b, and an inner cavity of the pressing ring 121 is in a shape of a straight cylinder or a frustum with a free front inner diameter gradually increasing. The selection is made according to the object to be photographed, for example: for an object, the object can be in a common straight cylinder shape, and a small object is positioned in a straight cylinder area and is close to the lens, so that a clear magnified picture of the object tissue structure can be obtained more easily; once the shot object is large in size, the shot object adopts a frustum shape like a conventional wood section texture structure, so that the image acquisition area can be enlarged, and focusing can be facilitated.
The lens external connection portion 122 is a male screw provided on the outer periphery of the first division body 120 a.
The lamp 32 is annular, the lamp 32 is embedded in the second cavity q2, and the press ring 121 abuts against the lamp 32.
The diameter of the annular inner cavity of the lamp 32 is larger than the outer diameter of the lens 10, and the center line of the lamp 32, the center line of the lens 10 and the center line of the pressing ring 121 are overlapped.
As shown in fig. 11, the light fixture 32 includes a ring light strip 20, a plurality of luminary bodies 21 disposed on the ring light strip 20, and a power and control switch.
Specifically, in one embodiment of the present invention, the light body 21 is a conventional light bulb, such as: an LED lamp. One side of the second cavity q2 is provided with a fitting notch k, and the lamp 32 further includes an external power connector 22 extending outward from one side of the ring-shaped light strip 20, wherein the external power connector 21 is inserted from the fitting notch k, and the ring-shaped light strip 20 abuts against the wall surface of the second cavity q2 and is spaced from the lens. Through the arrangement of the external power supply connector, on one hand, the positioning of the annular lamp strip can be realized, and the displacement is prevented; on the other hand, the power supply of the mobile phone can be communicated with the power supply of the lamp through the transmission line, so that the mobile phone directly supplies power.
Therefore, the embedded macro image acquisition terminal has the following advantages:
a. the annular light surrounding the camera can provide accurate annular light supplement (no annular light supplement lamp taking the mobile phone camera as the center at present) taking the shot object as the center for shooting the object, and the shadow of the shot object can be greatly eliminated relative to other lateral lights;
b. the light compensation cameras can be flexibly replaced aiming at the multi-camera mobile phone by matching with various fixing clamps;
c. the size is very small, the requirement of high flexibility of macro photography can be met, and the portability is high;
d. the power source is convenient to obtain, and the mobile phone can directly supply power;
e. after the macro lens and the shooting equipment are combined, the micro lens can be attached to the surface of an object to shoot a planar microstructure, and can also shoot small objects (such as insects, jewels, flowers and the like), meanwhile, the light is supplemented into an image acquisition area by the lamp, the focusing distance is reduced under the macro lens, the lens is stable and has no shadow light supplementing condition during shooting, and further, a picture with good effect (such as high definition) can be quickly and accurately obtained, and meanwhile, the micro lens can be operated by one person, and the micro lens is very convenient.
Having described the development and deployment, users and user permissions of the Web end in detail, the following specific description is made on the page layout of the Web end:
the Web page comprises eight modules, namely a home page, intelligent identification, a database, online support, statistical analysis, learning management, authority setting and mobile terminal management. Wherein, for the management user, the authority of the management user relates to all the modules; for expert users, the authority of the expert users relates to six modules including 'home page', 'intelligent recognition', 'database', 'online support', 'statistical analysis' and 'learning management'; for a common user, the authority of the common user relates to three modules of 'home page', 'intelligent identification' and 'database'. Relatively speaking, the authority of the ordinary user is the least, and the authority of the expert user is relatively more.
The eight modules will now be described in detail:
(1) "front page"
The 'home page' is divided into three columns, the first column shows that yesterday users use App to identify wood conditions, and shows the number of active yesterday users, the yesterday identification times and the yesterday high-frequency identification wood species respectively; the second column displays tree species identification accuracy rate ranking, three modes of a data report form, a line graph and a bar graph are provided to realize data visualization, and a user can download visualized images; the third column shows the wood identification number ranking and the user identification number of wood per day.
(2) Intelligent recognition "
The intelligent identification is divided into two steps, wherein in the first step, pictures are added, the pictures can be selected from a picture library to be uploaded, at most 20 jpg or png format pictures can be uploaded at one time, the uploaded pictures can be checked on line again, and invalid images can be selectively removed or completely removed; and secondly, starting recognition, clicking 'starting recognition' after checking is correct to recognize the wood, displaying a recognition result immediately below, displaying the first three types of trees with higher confidence respectively by the recognition result, clicking 'wood type pictures and texts' to check the detailed information of the tree, and clicking 'checking and issuing cases' to check the detailed information of the checking and issuing cases of the tree if the checking and issuing cases exist. In addition, after the recognition is finished, the 'clear recognition result' can be clicked to clear the recognition information, and the tidiness of the page is kept.
(3) "database"
The "library" includes the following four parts:
the first part is a wood species graph library, at present, the wood species graph library totally comprises 122 trees, the interface supports a wood retrieval function, a user can input Chinese names, Daltin names, market popular names, family information, distribution origins and endangered grades of wood for retrieval, if the wood exists in the graph library, detailed information of the wood can be automatically popped up below the graph library, and otherwise, temporary data are output. In addition, the interface also supports four functions of 'adding', modifying ', deleting' and 'exporting' of wood, and a user can click a corresponding module to correct a wood graph library according to requirements.
The second part is department management, the related information of the ordering and the existence of the departments can be obtained by inputting the name search of the departments, and if the departments do not exist, the user can realize the online addition of the departments. In addition, the user can jump to the corresponding position according to the family to manage the 'adding', 'modifying' and 'deleting' of the corresponding family.
The third part is case-finding cases, users can inquire the cases according to case titles, finding time, related tree species, finding customs, origin places and endangered grades, and if the database does not have the case-finding cases, the users can newly add cases on line. The newly-increased user for searching the case needs to add information such as case titles, searching and issuing time, related tree species, searching and issuing customs, origin, detailed description, text and the like. The interface also supports "add", "modify", "delete" and "export" of the recipe cases.
The fourth part is front observation, and the user can input front observation titles, whether the front observation titles are used alternately or not and relevant tree species for information retrieval. If the library is not present, online addition is supported. The user can click a 'new adding' button, fill in the front-edge lookout title, select related tree species, pictures for front-edge lookout and front-edge lookout contents, and determine whether to add the pictures as the carousel pictures or not. The method supports the rotation play graph to realize the display of related information at the App mobile terminal, so that the user can know the latest information. The user can also modify the added front-edge lookout title, the related tree species, the picture of the front-edge lookout, the front-edge lookout content and whether the front-edge lookout title is added as the carousel graph related content.
(4) "Online support"
The online support comprises the following three parts:
the first part is image-text online support, and a common user can initiate a sticker online, input a subject title and subject contents, add related wood pictures and wait for experts or other users to give comments on answering; the part also supports the retrieval of historical posts, and a user can input a user name applying for posts or the title of the posts for retrieval; the part also supports the application of voice and video calls, can communicate with online experts in voice or video, and can also check historical call records.
The second part is audio and video online support, and the part supports online communication, can select users needing communication and communication, and also supports the addition of new users.
And the third part is expert identification support, which is a link of rechecking for machine identification, and the expert can see the images of the identified wood and the identification results uploaded by the common user, and can diagnose whether the identification is accurate on line and give an opinion. The section also supports the inquiry of historical identification information, and a user can provide a user name, a name of a wood species to be identified, an identification rate and the like to retrieve the historical information.
(5) "statistical analysis"
"statistical analysis" is divided into two parts:
the first part is user information statistics, which mainly displays the total number of users, the identification times in the last 30 days and the historical identification total times, and can select the last five months, five weeks and five days to check the identification times of the users. The part provides three data visualization modes of a data report, a line graph and a bar graph for a user to select, and supports the downloading of images. In addition, the part provides display of the ten numbers of the last thirty days or the top of the identification times of the historical tree species, visualizes the data by using a data report, a line graph and a bar graph, and supports the downloading of images.
The second part is identification information statistics, which respectively shows the identification times ranking (top ten kinds of tree species), the tree species identification accuracy ranking and the tree species identification accuracy ranking (reverse order). The user may select the recognition accuracy ranking at different time periods.
(6) Learning management "
The learning management comprises three parts:
the first part is checking and warehousing, an administrator can screen according to data uploaded by the user and select 'warehousing' or 'deleting', a retrieval interface is further provided, and the administrator can input a user name, identify the wood species type, the identification rate, whether to check and the creation time to retrieve.
The second part is data storage, the part supports retrieval and checking of storage conditions, and a user can input information such as wood species types and creation time to inquire the storage conditions.
And the third part is model training, an administrator can select to identify the wood model and add a new model, and when the administrator needs to add the new model and clicks 'new addition', the new model can be added.
(7) Permission setting "
The 'permission setting' comprises three parts:
the first part is authorized by the user, summarizes all customs departments in China, can narrow the search range by inputting the name of the customs department, and can search by inputting information such as user number, mobile phone number, user job number and the like. The retrieved user may choose to delete or enter a decommissioned state. Clicking on the modification can modify the relevant information of the user, such as name, home department, mobile phone number, user job number and the like. The administrator can click to add a new user by 'new establishment', the user is required to provide a real name, an affiliation department, a mobile phone number, a user work number and the like, when the user forgets the password, the administrator can help the user to reset the password, and the administrator can click to 'reset' to input the new password.
The second part is user authority modification, which provides on-line search, and the administrator can input user name, authority character, state and creating time, and modify the user authority after searching. The project is divided into four user roles at present, namely an administrator, a customs administrator, an expert and a common user. The administrator may click on "modify" and check in front of the corresponding function to indicate that the user may use the function.
The third part is department management, which provides on-line search function, and can search through inputting department name or department state, to correct the information of Chinese customs department, when the information of customs department changes, the manager can click 'modify', to maintain the basic information of department, when new customs department is added, the manager can click 'add' to add new customs department.
(8) Mobile terminal management "
The mobile terminal management comprises two parts:
the first part is version updating service, an administrator can click 'add new version' to upload App installation package of the latest version, and a user can download and install App software of the latest version on a webpage end in time.
The second part is a mobile terminal home page display diagram, the mobile terminal home page display diagram is provided, and an administrator can set an App terminal home page display diagram and can add a new display diagram.
The above is the page layout of the Web end, and the following is the page layout of the App mobile end:
the App mobile terminal page comprises five modules of 'front-edge observation', 'knowledge dictionary', 'intelligent recognition', 'online support' and 'my', and the five modules are specifically explained:
(1) front edge lookout "
The top carousel graph of the front edge lookout can realize the carousel content playing in turn. The carousel content needs to be set at a Web end; the corresponding front-edge observation information can be clicked to browse the details; the content can be glided down for watching; in addition, when browsing articles, the right upper 'collecting' button can be clicked to collect the contents, and the pictures in the article contents can be clicked to browse the details of the pictures.
(2) Knowledge dictionary "
The material species image-text page is entered by default, and the upper button for searching and issuing case can be clicked to enter the page for searching and issuing case. In the upper search bar, characters can be input for fuzzy search, for example, "wood" is input, and all the names of wood with "wood" can be searched. And (3) sequentially showing the wood macroscopic structure, the wood name, the wood Latin school name and the wood description from left to right in the wood species column, and clicking a corresponding button to enter a wood species detail page. The wood species image-text detail page comprises a wood name, wood details and a wood picture from top to bottom in sequence. In addition, the collection button at the upper right can be clicked during browsing to collect the wood. The wood pictures can be slid left and right to check a plurality of pictures, and the details of the pictures can be entered by clicking the wood pictures. The case searching and issuing page comprises a search column and a case searching and issuing column, and any case searching and issuing column can be clicked to enter the case searching and issuing details. The detailed information of the case finding and issuing is the title of the case finding and issuing, the collection button, the information of the case finding and issuing, the text of the case finding and issuing and the picture of the case finding and issuing from top to bottom in sequence.
(3) Intelligent recognition "
And clicking an intermediate picture button to take a picture or upload an album. In the uploading process, a progress bar prompt is given. The result page after successful uploading comprises TOP3 tree pictures and confidence. If the wood is the expert user, clicking a 'result warehousing' button at the upper right after 'confirming the wood' can be clicked, and storing the result in a database; if the user is a common user, the user can click a 'request expert identification button', the request is stored in a background, and the expert uniformly processes the identification request at a Web end.
(4) "Online support"
The same as the online support module in the Web end, which is not described herein again.
(5) "My"
"My" relates to user information and permission settings.
The wood microspur characteristic image acquisition and AI identification system comprises a client, a database, a server, a wood image AI identification module and a microspur image acquisition terminal, and has the functions of wood image intelligent identification, wood information browsing and inquiring, expert remote online support, user identification history browsing and the like. When the worker encounters wood which is difficult to distinguish, the collected wood image is uploaded to the server, the server feeds back the recognition result to the user within 1-2s, and the remote online identification can be carried out on the questioned place by applying for experts, so that wood type information reference is provided for the user quickly. Moreover, the wood species which have doubts in the identification result and are not in the database can be further identified by experts, so that the wood identification accuracy is ensured.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (10)
1. A wood microspur characteristic image acquisition and AI identification system is characterized by comprising:
the client is used for receiving user operation;
a database comprising a user database for storing user data;
the server comprises a security authentication module and a service processing module;
the wood image AI identification module is used for identifying the wood image to be identified to obtain the wood species information of the wood;
the client and the database are respectively connected with the server, the client receives a login request and sends the login request to the server, after the server receives the login request, the security authentication module verifies whether the login information of the user is correct or not by comparing the user data in the user database,
if not, the security authentication module rejects the login request and sends a re-login instruction to the client;
if the wood image identification request is correct, the business processing module can receive an image identification request sent by the client, wherein the image identification request comprises the wood image to be identified;
and the business processing module identifies the wood image through a wood image AI identification module and feeds back an identification result to the client.
2. The wood macro feature image acquisition and AI identification system according to claim 1, further comprising a wood database, wherein the business processing module retrieves the wood database according to the identification result returned by the wood image AI identification module, and feeds back the retrieval result to the client.
3. The wood macro feature image collection and AI identification system according to claim 1, further comprising a macro image collection terminal capable of being mounted on a camera lens of a mobile terminal, so that the camera lens of the mobile terminal captures a detailed image of wood to be identified.
4. The wood macro feature image collection and AI identification system of claim 3, wherein the macro image collection terminal is an assembled macro image collection terminal comprising a macro lens, an adapter disposed at an outer end of the macro lens, and a collection head disposed at an outer end of the adapter, wherein,
the adapter is inside to be equipped with the confession the perspective chamber of macro lens, the collection head including be formed with the collection base of the annular chamber that the perspective chamber is linked together and set up lamps and lanterns in the collection base, wherein, the collection base the adapter reaches and waits to detect sample surface and form confined image acquisition district, perhaps, the collection base the adapter reaches and places the platform surface that detects the sample and form confined image acquisition district.
5. The wood macro characteristic image acquisition and AI identification system according to claim 3, wherein the macro image acquisition terminal is an embedded macro image acquisition terminal, the embedded macro image acquisition terminal comprises a macro lens, the macro lens comprises a lens holder and a lens, the lens holder comprises a holder body, a press ring detachably disposed at a front end of the holder body, and a lens external connection portion disposed at a rear end of the holder body, wherein the press ring is annular, and an inner cavity of the holder body, an inner cavity of the press ring, and a surface of a sample to be detected form a closed image acquisition area, or an inner cavity of the holder body, an inner cavity of the press ring, and a platform surface on which the sample to be detected is placed form a closed image acquisition area;
the embedded micro-distance image acquisition terminal is characterized in that the embedded micro-distance image acquisition terminal is also arranged on the seat body or a lamp used for compensating exposure and arranged on the inner wall of the compression ring, the lamp is annular, and the central line of the lamp and the central line of the lens are superposed.
6. The wood macro feature image acquisition and AI identification system according to claim 1, wherein the wood image AI identification module identifies an image comprising the steps of:
s1, inputting a wood image and preprocessing the wood image;
s2, processing the wood image in blocks;
s3, training the sub-image blocks through a convolutional neural network model;
s4, using different gradient values from the edge to the center of the wood image as the weight of the sub-image classification score of different areas, increasing the proportion of the center area in the whole wood image classification score, converting the weighted score into a final probability value, obtaining the wood species information of the wood image, wherein,
the step of preprocessing in S1 includes:
s11, performing color correction on the wood image;
s12, performing data enhancement on the corrected wood image;
the step of block-processing the wood image in S2 includes:
s21, segmenting the preprocessed wood image to obtain a plurality of sub-images;
and S22, unifying the size pixels of each sub-image by a bilinear interpolation method.
7. The wood macro feature image collection and AI identification system of claim 6, wherein the step S11 includes performing color correction on the wood cross-sectional image through a gray world method, calculating a gain coefficient and a gray average of R, G, B three channels from pixel expectation values of R, G, B three channels of the whole wood cross-sectional image, and performing adjustment of respective R, G, B three-channel components on each pixel in the wood cross-sectional image according to the gain coefficient and the gray average;
step S12 includes performing data enhancement on the corrected wood cross-sectional images by using horizontal inversion, vertical inversion and adding salt and pepper noise so that the training sample size of each wood image is within a prescribed number range.
8. The wood macro characteristic image acquisition and AI identification system according to claim 1, further comprising a load balancer disposed between the client and the server, wherein data sent by the client is processed by the load balancer and then uniformly distributed to the server.
9. The wood macro feature image collection and AI identification system according to claim 6, further comprising an expert identification support module for sending an expert identification support request including the wood image to be identified and/or the identification result returned by the wood image AI identification module, and receiving expert identification information returned in response to the expert identification support request.
10. The wood macro feature image acquisition and AI identification system of claim 1, wherein the database comprises MySQL and/or Oracle and/or SQLite.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110134276.9A CN112767388B (en) | 2021-01-29 | 2021-01-29 | Wood microspur characteristic image acquisition and AI identification system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110134276.9A CN112767388B (en) | 2021-01-29 | 2021-01-29 | Wood microspur characteristic image acquisition and AI identification system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112767388A true CN112767388A (en) | 2021-05-07 |
| CN112767388B CN112767388B (en) | 2023-12-26 |
Family
ID=75704399
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110134276.9A Active CN112767388B (en) | 2021-01-29 | 2021-01-29 | Wood microspur characteristic image acquisition and AI identification system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112767388B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| BRPI0704329A2 (en) * | 2007-10-23 | 2009-06-23 | De Oliveira Luiz Eduardo Soares | computational process of wood image analysis |
| CN108648191A (en) * | 2018-05-17 | 2018-10-12 | 吉林大学 | Pest image-recognizing method based on Bayes's width residual error neural network |
| US20190227049A1 (en) * | 2017-03-13 | 2019-07-25 | Lucidyne Technologies, Inc. | Method of board lumber grading using deep learning techniques |
| CN110503051A (en) * | 2019-08-27 | 2019-11-26 | 西南林业大学 | A kind of precious timber identifying system and method based on image recognition technology |
| CN211352280U (en) * | 2019-12-23 | 2020-08-25 | 联想(北京)有限公司 | Light-emitting device and electronic equipment |
| CN111652307A (en) * | 2020-05-29 | 2020-09-11 | 广西大学 | Intelligent nondestructive identification method and device for redwood furniture based on convolutional neural network |
| CN111696033A (en) * | 2020-05-07 | 2020-09-22 | 中山大学 | Real image super-resolution model and method for learning cascaded hourglass network structure based on angular point guide |
| CN111898649A (en) * | 2020-07-07 | 2020-11-06 | 西南林业大学 | Wood tree species classification and identification method based on visible light or near infrared spectrum analysis |
-
2021
- 2021-01-29 CN CN202110134276.9A patent/CN112767388B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| BRPI0704329A2 (en) * | 2007-10-23 | 2009-06-23 | De Oliveira Luiz Eduardo Soares | computational process of wood image analysis |
| US20190227049A1 (en) * | 2017-03-13 | 2019-07-25 | Lucidyne Technologies, Inc. | Method of board lumber grading using deep learning techniques |
| CN108648191A (en) * | 2018-05-17 | 2018-10-12 | 吉林大学 | Pest image-recognizing method based on Bayes's width residual error neural network |
| CN110503051A (en) * | 2019-08-27 | 2019-11-26 | 西南林业大学 | A kind of precious timber identifying system and method based on image recognition technology |
| CN211352280U (en) * | 2019-12-23 | 2020-08-25 | 联想(北京)有限公司 | Light-emitting device and electronic equipment |
| CN111696033A (en) * | 2020-05-07 | 2020-09-22 | 中山大学 | Real image super-resolution model and method for learning cascaded hourglass network structure based on angular point guide |
| CN111652307A (en) * | 2020-05-29 | 2020-09-11 | 广西大学 | Intelligent nondestructive identification method and device for redwood furniture based on convolutional neural network |
| CN111898649A (en) * | 2020-07-07 | 2020-11-06 | 西南林业大学 | Wood tree species classification and identification method based on visible light or near infrared spectrum analysis |
Non-Patent Citations (2)
| Title |
|---|
| ANINDITA SAFNA ET AL.: "Wood species identification using convolutional neural network(CNN) architectures on macroscopic images", 《JITECS》, vol. 4, no. 3, pages 274 - 283 * |
| 曾铮等: "基于轻量化卷积神经网络的木材种类识别", 《现代计算机》, pages 59 - 63 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112767388B (en) | 2023-12-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10387938B2 (en) | Automated color processing and selection platform | |
| CN104854859B (en) | Recommend transformation for photography | |
| CA2890158C (en) | Automated color processing and selection platform | |
| US9558401B2 (en) | Scanbox | |
| US8639041B2 (en) | Product identification using image analysis and user interaction | |
| JP5790094B2 (en) | Image management system, image management system control method, control program, and recording medium | |
| US20150149454A1 (en) | Preferred image retrieval | |
| US20070070217A1 (en) | Image analysis apparatus and image analysis program storage medium | |
| CN109711443A (en) | Floor plan recognition methods, device, equipment and storage medium neural network based | |
| CN110083762A (en) | Source of houses searching method, device, equipment and computer readable storage medium | |
| CN109523499A (en) | A kind of multi-source fusion full-view modeling method based on crowdsourcing | |
| Li et al. | Lighting measurement with a 360° panoramic camera: Part 2–Applications | |
| CN112767388B (en) | Wood microspur characteristic image acquisition and AI identification system | |
| CN111598874A (en) | Mangrove canopy density survey method based on intelligent mobile terminal | |
| CN111563759A (en) | Agricultural product traceability process identification and analysis system based on AI technology | |
| CN107945353B (en) | Self-service tour guide system based on deep learning | |
| CA3112703A1 (en) | Image management with region-based metadata indexing | |
| JP2006101156A (en) | Information processing device and program | |
| Carboni et al. | Data provenance in photogrammetry through documentation protocols | |
| KR101580034B1 (en) | Color extraction and storing method using mobile terminal | |
| CN106954009A (en) | Collection of illustrative plates acquisition device, method and portable terminal device | |
| CN113824874A (en) | Auxiliary shooting method and device, electronic equipment and storage medium | |
| John et al. | drpToolkit: An automated workflow for aligning and analysing vegetation and ground surface time‐series imagery | |
| CN103198162B (en) | A kind of picture browsing exchange method | |
| CN111222392A (en) | Intelligent courtyard control system |
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 |