CN113252584B - Crop growth detection method and system based on 5G transmission - Google Patents
Crop growth detection method and system based on 5G transmission Download PDFInfo
- Publication number
- CN113252584B CN113252584B CN202110438384.5A CN202110438384A CN113252584B CN 113252584 B CN113252584 B CN 113252584B CN 202110438384 A CN202110438384 A CN 202110438384A CN 113252584 B CN113252584 B CN 113252584B
- Authority
- CN
- China
- Prior art keywords
- crop
- characteristic
- growth condition
- crops
- information
- 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
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
-
- 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
-
- 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
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G08—SIGNALLING
- G08C—TRANSMISSION SYSTEMS FOR MEASURED VALUES, CONTROL OR SIMILAR SIGNALS
- G08C17/00—Arrangements for transmitting signals characterised by the use of a wireless electrical link
- G08C17/02—Arrangements for transmitting signals characterised by the use of a wireless electrical link using a radio link
-
- 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
- G06T2207/10032—Satellite or aerial image; Remote sensing
- G06T2207/10036—Multispectral image; Hyperspectral 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30181—Earth observation
- G06T2207/30188—Vegetation; Agriculture
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A40/00—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
- Y02A40/10—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in agriculture
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Bioinformatics & Computational Biology (AREA)
- Medical Informatics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Evolutionary Biology (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The application relates to a crop growth detection method and system based on 5G transmission. The method comprises the following steps: collecting hyperspectral images of crops and identifying the variety types of the crops by utilizing a crop identification model; finding out the characteristic wave band type according to the variety type; processing the hyperspectral image according to the characteristic wave band type of the hyperspectral image to obtain a characteristic image under the characteristic wave band; analyzing according to the characteristic image to obtain the growth condition information of the crops; in addition, the system includes: the intelligent hyperspectral camera and the cloud server can realize image acquisition and image preprocessing through the intelligent hyperspectral camera; therefore, according to the scheme provided by the application, the data volume to be processed can be reduced on the premise of ensuring the reliability of the data, the calculation pressure of the cloud server is reduced, and the data transmission and processing speed is increased.
Description
Technical Field
The application relates to the technical field of agricultural machine vision, in particular to a crop growth detection method and system based on 5G transmission.
Background
The facility horticulture is an industry of high and cross integration of biological high and new technology, environmental control technology and greenhouse engineering technology, refers to a modern agricultural production mode for efficiently producing crops by adopting engineering technology under the condition of relatively controllable environment, is a knowledge and technology highly intensive industry, has the characteristics of high technological content, high investment, high yield, high benefit, easiness in intensive production and the like, is one of important marks of the agricultural modernization level, and is an important component of modern agriculture. With the wide application of information technology in the agricultural field, the information amount and the processing requirement of the growth condition information of crops are far beyond the traditional processing method and the transmission capability. How to effectively collect and utilize the information is one of the problems to be solved urgently in the field of facility gardening.
In the related technology, the hyperspectral imaging provides an effective way for real-time, nondestructive and rapid detection of the physiological indexes of crops by virtue of strong continuity of wave bands and high spectral resolution.
Because the hyperspectral imaging technology acquires more information data, the data volume of the hyperspectral imaging technology is dozens of times or even hundreds of times of that of the traditional imaging technology, when the traditional hyperspectral imaging technology is used for detecting crops, huge data volume causes great calculation pressure to a computer, and the data processing process is slow in speed and poor in timeliness.
Disclosure of Invention
In order to overcome the problems in the related art, the application provides a crop growth detection method and system based on 5G transmission, which can reduce the data volume to be processed and reduce the calculation pressure, thereby improving the speed of data transmission and processing.
The application provides a crop growth detection method based on 5G transmission in a first aspect, which comprises the following steps:
collecting a hyperspectral image of a crop;
identifying the hyperspectral image through a crop identification model to obtain the variety type of the crop;
searching according to the variety type to obtain the characteristic wave band type of the crop; under the characteristic wave band corresponding to the characteristic wave band type, the characteristic quantity of the hyperspectral image of the crop is sequenced into the first N bits in the characteristic quantity sequence; the characteristic quantity sequence is a sequence obtained by sequentially arranging characteristic quantities of the hyperspectral images from high to low under M wave bands;
extracting and obtaining N characteristic images of the crops under corresponding characteristic wave bands based on the characteristic wave band types; wherein N is a positive integer and N is less than M;
and analyzing the characteristic image through a growth condition monitoring model to obtain the growth condition information of the crops.
In one embodiment, the identifying the hyperspectral image by the crop identification model to obtain the variety type of the crop includes:
segmenting the hyperspectral image by using an image local gravity center algorithm to obtain a single crop image;
extracting the single crop image by using a Canny operator to obtain the edge information of the crop;
performing pseudo-color synthesis on the hyperspectral image to obtain color information of the crop;
and matching the crop species based on the edge information and the color information to obtain the variety type of the crop.
In one embodiment, before the retrieving the characteristic band type of the crop according to the variety type, the method includes: collecting the characteristic wave band types of different crops, and establishing a characteristic wave band database;
the obtaining of the characteristic wave band type of the crop according to the variety type retrieval comprises: and matching the crop variety type with the characteristic waveband type according to a characteristic waveband database to obtain the characteristic waveband type of the crop.
In one embodiment, the acquiring characteristic band types of different crops and establishing a characteristic band database includes:
collecting the crop spectral image;
obtaining the characteristic quantity of the crops under different characteristic wave bands by adopting a continuous selection projection algorithm;
sorting the characteristic quantity from high to low;
and selecting N characteristic wave bands corresponding to the first N bits of the characteristic quantity in the sequence as the characteristic wave band types of the crops.
In one embodiment, the growth monitoring model is trained using a crop training set; the input of the growth condition monitoring model is a hyperspectral image of the crop, and the output is the growth condition information of the crop;
the growth condition information comprises: crop ingredients and ingredient content;
the crop training set comprises: a spectral data set of the crop; the spectral data set is a hyperspectral image with growth condition information as a label.
In one embodiment, before analyzing the characteristic image by the growth monitoring model to obtain the growth information of the crop, the method includes:
acquiring a characteristic waveband spectrum data set of the crop;
and fitting the regression relationship between the characteristic wave band spectrum data set of the crops and the growth condition information by using a partial least square method to obtain the growth condition monitoring model.
In an embodiment, after the analyzing the characteristic image by the intelligent growth monitoring model to obtain the growth condition information of the crop, the method further includes:
performing data analysis according to the growth condition information to obtain a cultivation strategy of the crop;
the culture strategy comprises the following steps: elements to be supplemented and recommended supplement amounts.
In one embodiment, the analyzing the data according to the growth status information to obtain the cultivation strategy of the crop comprises:
grading the growth condition of the crops by using a grading model to obtain the growth condition grade of the crops;
combining the growth condition information and the growth condition grade to obtain a cultivation strategy of the crop;
the grading model is obtained by establishing a classification model by taking the growth condition information as a data set characteristic and utilizing an SVM (support vector machine) algorithm and a Gaussian core to check the growth condition information and the growth condition grade.
A second aspect of the present application provides a crop growth detection system based on 5G transmission, comprising:
the system comprises an intelligent hyperspectral camera and a cloud server;
the intelligent hyperspectral camera comprises: the device comprises an acquisition module, a storage module, a calculation module and a transmission module;
the acquisition module is used for acquiring hyperspectral images of crops;
the storage module is used for storing a characteristic waveband database and a crop identification model code;
the calculation module is used for executing the crop identification model code to identify and obtain the variety type of the crop, extracting the characteristic wave band type of the crop according to the characteristic wave band database in the storage module, and obtaining the characteristic image of the crop by utilizing the characteristic wave band type;
the transmission module is used for transmitting the characteristic image obtained in the calculation module to the cloud server;
and the cloud server is deployed with an intelligent growth condition monitoring model and used for processing data transmitted by the intelligent hyperspectral camera to obtain growth condition information of the crops.
In one embodiment, the crop growth detection system based on 5G transmission further comprises: an interactive front end;
the interactive front end performs data transmission with the intelligent hyperspectral camera and the cloud server by using a WebSocket protocol; through the interactive front end, the intelligent hyperspectral camera can be opened and closed, and the camera parameter setting and cloud server data are visualized.
In one embodiment, the transmission module is a 5G transmission module; the 5G transmission module transmits data and instructions in a JSON format by using a WebSocket protocol, an HTTP protocol, a TCP protocol or a UDP protocol so as to realize full-duplex communication between the intelligent hyperspectral camera and the cloud server.
The technical scheme provided by the application can comprise the following beneficial effects:
according to the scheme, the variety type of the crop is identified by using the crop identification model, the collected hyperspectral image is extracted according to the variety type, and the hyperspectral information of the crop under the characteristic wave band is obtained, so that the characteristic image with the main characteristics of the crop is obtained, on the premise that the characteristic image can reliably represent the crop, the data volume to be processed is reduced, the calculation pressure is reduced, the data transmission and processing speed is increased, and the timeliness of crop detection is improved; in addition, the intelligent growth condition monitoring model improves the accuracy of crop detection results, improves the data analysis speed and improves the detection timeliness.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the application.
Drawings
The foregoing and other objects, features and advantages of the application will be apparent from the following more particular descriptions of exemplary embodiments of the application, as illustrated in the accompanying drawings wherein like reference numbers generally represent like parts throughout the exemplary embodiments of the application.
Fig. 1 is a schematic flow chart of a crop growth detection method based on 5G transmission according to an embodiment of the present application;
fig. 2 is a schematic flow chart of a crop variety type identification method according to an embodiment of the present disclosure;
FIG. 3 is a diagram illustrating a process of building a characteristic band database according to an embodiment of the present application;
fig. 4 is a schematic flow chart of a crop cultivation strategy acquisition method according to an embodiment of the present application.
Detailed Description
Preferred embodiments of the present application will be described in more detail below with reference to the accompanying drawings. While the preferred embodiments of the present application are shown in the drawings, it should be understood that the present application may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in this application and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.
It should be understood that although the terms "first," "second," "third," etc. may be used herein to describe various information, these information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of the present application. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the present application, "a plurality" means two or more unless specifically limited otherwise.
When the traditional hyperspectral imaging technology is used for detecting crops, huge data volume causes great calculation pressure to a computer, and the data processing process is slow and has poor timeliness.
Example one
In view of the above problems, embodiments of the present application provide a crop growth detection method based on 5G transmission, which can reduce the amount of data to be processed and increase the data processing speed of crop growth detection.
The technical solutions of the embodiments of the present application are described in detail below with reference to the accompanying drawings.
Fig. 1 is a schematic flow chart of a crop growth detection method based on 5G transmission according to an embodiment of the present application.
Referring to fig. 1, the crop growth detection method based on 5G transmission includes:
101. collecting a hyperspectral image of a crop;
in the embodiment of the application, the intelligent hyperspectral camera is used for collecting the hyperspectral image of the crop. A crop identification model and a characteristic waveband database are loaded in the intelligent hyperspectral camera, and the collected hyperspectral image can be subjected to characteristic image extraction through the following steps 102 and 103, so that the data transmitted by the intelligent hyperspectral camera is finally the characteristic image of the crop under the characteristic waveband, compared with the original hyperspectral image, the data size is reduced, and the transmission speed is improved.
102. Identifying the hyperspectral image through a crop identification model to obtain the variety type of the crop;
in the embodiment of the application, the crop identification model performs data matching of crop variety types according to the edge information and the color information of crops, a mapping table of the crop edge information and the color information and the crop variety types is stored in the intelligent hyperspectral camera, and the variety types of the crops can be found based on the crop edge information and the color information according to the mapping relation of the crop edge information and the color information and the crop variety types.
It should be noted that, in the actual application process, based on the actual situation, the information with the minimum distance may be found by calculating the collected variety identification information of the crop to be detected, that is, the euclidean distance between the edge information and the color information, and the variety identification information of each crop in the mapping table, and the corresponding crop variety type is the variety type of the crop to be detected.
It is understood that the above-mentioned process for matching the crop variety type data is only an example in the embodiment of the present application, and should not be taken as a limitation to the present invention.
103. Searching according to the variety type to obtain the characteristic wave band type of the crop;
under the characteristic wave band corresponding to the characteristic wave band type, the characteristic quantity of the hyperspectral image of the crop is sequenced into the first N bits in the characteristic quantity sequence; the characteristic quantity sequence is a sequence obtained by sequentially arranging the characteristic quantities of the hyperspectral images from high to low under M wave bands; wherein N is a positive integer and N is less than M.
In the embodiment of the present application, the obtaining process of the characteristic band type of the crop is as follows: and matching the crop variety type with the characteristic waveband type according to a characteristic waveband database to obtain the characteristic waveband type of the crop.
In the embodiment of the application, the intelligent hyperspectral camera is provided with a characteristic waveband database, the characteristic waveband database stores a mapping table of the crop variety type and the characteristic waveband type, and the characteristic waveband type of the crop to be detected can be obtained by matching from the mapping table based on the variety type of the crop to be detected obtained in the step 102.
104. Extracting and obtaining N characteristic images of the crops under corresponding characteristic wave bands based on the characteristic wave band types; wherein N is a positive integer and N is less than M;
in the embodiment of the application, the characteristic band types of the N crops to be detected are obtained through step 103, and accordingly, the hyperspectral images of the crops to be detected are subjected to extraction of the characteristic images under the bands corresponding to the characteristic band types, so that the N characteristic images are obtained.
105. And analyzing the characteristic image through a growth condition monitoring model to obtain the growth condition information of the crops.
In the embodiment of the application, the growth condition monitoring model is formed by training a mathematical model established based on a partial least square method through a crop training set; the hyperspectral image of the crop is input, and the growth condition information of the crop is output.
In the practical application process, the growth condition monitoring model can also adopt a convolutional neural network; and training the convolutional neural network by using a gradient cutting method or a regularization method based on a crop training set to obtain a growth condition monitoring model.
In an embodiment of the present application, the growth status information includes: crop ingredients and ingredient content; the above crop training set comprises: a set of spectral data of the crop; the spectral data set is a hyperspectral image with growth condition information as a label.
Further, the growth status information further includes: the shape parameter; the growth conditions of the crops can be graded by combining the appearance parameters and the component contents of the crops, and then a corresponding crop cultivation strategy is formulated.
In the embodiment of the application, before analyzing the characteristic image through the growth condition monitoring model to obtain the growth condition information of the crop, the growth condition monitoring model needs to be established. The establishment of the growth condition monitoring model can be expressed as follows: acquiring a characteristic waveband spectrum data set of the crop; and fitting the regression relationship between the characteristic wave band spectrum data set of the crops and the growth condition information by using a partial least square method to obtain the growth condition monitoring model.
Taking ganoderma lucidum as an example: after a hyperspectral image of the ganoderma is collected, the edge of the ganoderma is obtained through a Sobel operator, and therefore the area of a ganoderma pileus is calculated; based on the content of the ganoderma lucidum polysaccharide obtained by pre-detection, a data set of the growth condition of the ganoderma lucidum can be formed, wherein the data set comprises ganoderma lucidum hyperspectral image data, ganoderma lucidum pileus area and polysaccharide content; and establishing a ganoderma lucidum hyperspectral image and growth condition information, namely a regression relation between the ganoderma lucidum hyperspectral image and polysaccharide content and pileus area by utilizing a partial least square method based on the ganoderma lucidum growth condition data group, thereby completing model establishment and obtaining a growth condition monitoring model.
The scheme shown in the embodiment of the application identifies the variety type of the crop by using the crop identification model, extracts the collected hyperspectral image according to the variety type to obtain the hyperspectral information of the crop under the characteristic wave band, thereby obtaining the characteristic image retaining the main characteristics of the crop, reduces the data volume to be processed and reduces the calculation pressure on the premise that the characteristic image can reliably represent the crop, thereby improving the speed of data transmission and processing and improving the timeliness of crop detection; in addition, the intelligent growth condition monitoring model improves the accuracy of crop detection results, improves the data analysis speed and improves the detection timeliness.
Example two
The embodiment of the present application designs step 102 in the first embodiment.
Fig. 2 is a schematic flow chart of a crop variety type identification method according to an embodiment of the present application.
Referring to fig. 2, the method for identifying the type of crop variety includes:
201. segmenting the hyperspectral image by using an image local gravity center algorithm to obtain a single crop image;
in an embodiment of the application, the hyperspectral images include images of crops collected under light of wavelength bands of 450nm, 550nm and 650 nm.
In the embodiment of the application, the contour of the image is extracted by using a FindContour function in OpenCV, then the centroid of the contour is obtained by adopting the first moment positioning of the image, and the hyperspectral image is segmented by combining the image contour and the contour centroid, so that the image of the single crop can be obtained.
In the practical application process, the hyperspectral image can be processed by adopting an example segmentation model to obtain a single crop image.
202. Extracting the single crop image by using a Canny operator to obtain the edge information of the crop;
in this embodiment of the present application, the purpose of processing the individual crop image by using a Canny operator is to extract edge features of the crop image, and a specific process may be described as follows: smoothing the image with a gaussian filter; calculating the gradient amplitude and direction by using a first-order partial derivative finite difference method; carrying out non-maximum suppression on the gradient amplitude; edges are detected and connected using a dual threshold algorithm.
It should be noted that, in the actual application process, the edge detection algorithm may be adjusted according to the actual situation, for example, the edge information of the hyperspectral image of the crop may also be extracted by using a laplacian operator.
It is to be understood that the above description of the edge detection algorithm should not be taken as limiting the invention.
203. Performing pseudo-color synthesis on the hyperspectral image to obtain color information of the crop;
in the embodiment of the application, a color synthesizer made according to an additive method principle can be used for processing the hyperspectral image to synthesize a pseudo-color image, and the color information of the crop can be extracted based on the obtained pseudo-color image.
204. And matching the crop species based on the edge information and the color information to obtain the variety type of the crop.
In the embodiment of the present application, the crop species matching process is as follows: and matching the crop species in a crop image database by combining the edge information and the color information and using a Haar feature operator, so as to identify the crop species. And storing sample data corresponding to the crop variety type, the crop edge information and the color information in the crop image database.
After the scheme shown in the embodiment of the application segments the single crop image, extracting the edge characteristic information of the crop by using a Canny operator; acquiring color information of the crop by utilizing pseudo color synthesis; the method combines the edge characteristic information and the color information, utilizes the Haar characteristic operator to match the crop variety types, thereby identifying the crop variety types, and provides a reliable basis for the subsequent identification of the crop characteristic wave band types, so that the characteristic image extracted based on the characteristic wave band not only retains the main characteristics of the crops, but also reduces the irrelevant data quantity.
EXAMPLE III
According to the characteristic band type obtaining process described in the first embodiment, before step 103, characteristic band types of different crops need to be collected to establish a characteristic band database. The embodiment of the application designs the establishment of the characteristic waveband database.
Fig. 3 is a schematic diagram illustrating a characteristic band database establishment process according to an embodiment of the present application.
Referring to fig. 3, the establishing process of the characteristic band database includes:
301. collecting the crop spectral image;
in the embodiment of the application, the crop spectral image is used as a sample for establishing a characteristic waveband database, and standard normal variable transformation processing can be performed on the crop spectral image to reduce the influence of non-uniform particle size and non-specific scattering on the particle surface on spectral data.
Specifically, the transformation process is performed on each spectral image using the following formula:
wherein x is snv Is the result of the standard normal variable transformation process, x (i，m) Spectral data at the m-th wavelength band representing the ith sample,
represents the average of the spectral data of the ith sample at all bands.
It should be noted that the above description of the standard normal variable transformation is only an example in the embodiment of the present application, and does not limit the present invention.
302. Obtaining characteristic quantities of the crops under different characteristic wave bands by adopting a continuous selection projection algorithm;
303. sorting the characteristic quantities from high to low;
304. and selecting N characteristic wave bands corresponding to the first N bits of the characteristic quantity in the sequence as the characteristic wave band types of the crops.
In the embodiment of the application, in the experimental stage, the value of N is 30; in the experiment, a continuous selection projection algorithm is adopted for carrying out characteristic wave band selection and characteristic quantity proportion evaluation on a hyperspectral image of a crop, the spectrum wave band of 30 bits before the crop characteristic quantity is sequenced is selected from the wave bands obtained in the experiment as the characteristic wave band type of the crop, and the characteristic wave band database is established after the processing is carried out on different crops.
The characteristic wave band database is stored in a storage module of the intelligent hyperspectral camera, and after the variety type of the crop to be detected is identified, the characteristic wave band database is called to match the characteristic wave band type.
It should be noted that the value of N is a value adopted in an experimental stage, and in an actual application process, the value of N may be adjusted according to an application scenario.
It is to be understood that the above values for N should not be construed as limiting the invention.
The characteristic wave band database shown in the embodiment of the application is established based on sample data obtained after characteristic wave band types of different crops are identified, wherein the sample data comprises a plurality of crops and corresponding characteristic wave band types, and when the growth of the crops to be detected is detected, the characteristic wave band types of the crops to be detected can be obtained by directly calling the data in the characteristic wave band database, so that the characteristic images of the crops to be detected are extracted, the process is simple and quick, and the data processing speed is accelerated; meanwhile, based on the scheme shown in the embodiment of the application, the data of the characteristic waveband database can be expanded, so that the application range of the characteristic waveband database is widened.
Example four
On the basis of the crop growth detection method based on 5G transmission described in the first embodiment, the embodiment of the present application further designs an acquisition method of the crop cultivation strategy, that is, after the growth condition information is obtained according to the method described in the first embodiment, data analysis is performed according to the growth condition information to obtain the cultivation strategy of the crop.
Fig. 4 is a schematic flow chart of a crop cultivation strategy acquisition method according to an embodiment of the present application.
Referring to fig. 4, the method for obtaining the crop cultivation strategy comprises the following steps:
401. grading the growth condition of the crops by using a grading model to obtain the growth condition grade of the crops;
in the embodiment of the present application, the grading model is a classification model of growth condition information and growth condition grade, which is established by using the growth condition information as a data set feature and combining an SVM algorithm with a gaussian kernel. The input of the grading model is the growth condition information of the crop to be detected, and the output is the growth condition grade of the crop to be detected.
In the embodiment of the present application, in the experimental stage, the growth condition grades include four grade criteria of good, medium and poor, it should be noted that, in the practical application process, the division of the growth condition grades may be adjusted according to the practical requirements, for example, the growth condition grades are divided into I grade, II grade and III grade. Taking dendrobium as an example: and in the experimental stage, the standard nitrogen content of the dendrobium is set as S, and the grade of the growth condition is defined as excellent when the nitrogen content of the dendrobium is higher than S, good when the nitrogen content of the dendrobium is lower than S and is higher than 1/2S, medium when the nitrogen content of the dendrobium is lower than 1/2S and is higher than 1/3S, and poor when the nitrogen content of the dendrobium is lower than 1/3S.
402. And combining the growth condition information and the growth condition grade to obtain a culture strategy of the crop.
In the examples of the present application, the cultivation strategy comprises: elements to be supplemented and recommended supplement amounts.
In the embodiment of the application, a crop cultivation strategy database can be established in advance, wherein the cultivation strategy database comprises a plurality of sets of cultivation strategies of a plurality of crops under a plurality of growth condition grades for a plurality of growth condition information; and carrying out data calling and data matching on the culture strategy database by combining the growth condition information and the growth condition grade of the crop to be detected, so as to obtain the culture strategy corresponding to the crop to be detected.
Furthermore, the plurality of sets of cultivation strategies in the cultivation strategy database can also reference the appearance parameters of the crops to form a crop cultivation strategy which is formulated based on four-dimensional data of the types, the component contents, the growth condition grades and the appearance parameters of the crops.
It should be noted that the above description of the crop cultivation strategy database is only an example shown in the embodiments of the present application, and is not necessarily taken as a limitation of the present invention.
The crop cultivation strategy acquisition method shown in the embodiment of the application is based on a crop growth detection method, and the growth condition information of the crops obtained by the growth detection method is used for grading the growth condition of the crops, so that a cultivation strategy adaptive to the current growth condition of the crops is matched, and the intelligent detection and cultivation of the crops are realized.
EXAMPLE five
Corresponding to the embodiment of the application function implementation method, the application also provides a crop growth detection system based on 5G transmission and a corresponding embodiment.
The crop growth detection system based on 5G transmission comprises:
the system comprises an intelligent hyperspectral camera and a cloud server;
the intelligent hyperspectral camera comprises: the device comprises an acquisition module, a storage module, a calculation module and a transmission module;
the acquisition module is used for acquiring hyperspectral images of crops;
the storage module is stored with a characteristic wave band database and a crop identification model code;
the calculation module is used for executing the crop identification model code to identify and obtain the variety type of the crop, extracting the characteristic wave band type of the crop according to the characteristic wave band database in the storage module, and obtaining the characteristic image of the crop by utilizing the characteristic wave band type;
the transmission module is used for transmitting the characteristic image obtained in the calculation module to the cloud server;
and the cloud server is deployed with an intelligent growth condition monitoring model for processing the data transmitted by the intelligent hyperspectral camera to obtain the growth condition information of the crops.
In the embodiment of the application, the transmission module is a 5G transmission module developed based on a fifth generation mobile communication technology; the 5G transmission module transmits data and instructions in a JSON format by using a WebSocket protocol, an HTTP protocol, a TCP protocol or a UDP protocol so as to realize full-duplex communication between the intelligent hyperspectral camera and the cloud server.
It should be noted that the communication protocol adopted by the 5G transmission module in this embodiment may be any one of a WebSocket protocol, an HTTP protocol, a TCP protocol, and a UDP protocol, and specifically, the selected communication protocol may be determined according to actual conditions, that is, the communication protocol adopted by the 5G transmission module is not necessarily taken as a limitation to the present invention.
In embodiments of the present application, the storage module may include various types of storage units, such as a system memory, a Read Only Memory (ROM), and a permanent storage device. Wherein the ROM may store static data or instructions for other modules. The persistent storage device may be a read-write storage device. The persistent storage may be a non-volatile storage device that does not lose stored instructions and data even after the computer is powered off. In some embodiments, the persistent storage device employs a mass storage device (e.g., magnetic or optical disk, flash memory) as the persistent storage device. The system memory may be a read-write memory device or a volatile read-write memory device, such as a dynamic random access memory. The system memory may store instructions and data that some or all of the processors require at runtime.
Further, the crop growth detection system based on 5G transmission further includes: an interactive front end;
the interactive front end uses a WebSocket protocol to perform data transmission with the intelligent hyperspectral camera and the cloud server; through the interactive front end, can realize the opening and closing of intelligence hyperspectral camera, the visual of camera parameter setting and high in the clouds server data.
With regard to the apparatus in the above-described embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.
The aspects of the present application have been described in detail hereinabove with reference to the accompanying drawings. In the above embodiments, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments. Those skilled in the art should also appreciate that the acts and modules referred to in the specification are not necessarily required in the present application. In addition, it can be understood that the steps in the method of the embodiment of the present application may be sequentially adjusted, combined, and deleted according to actual needs, and the modules in the device of the embodiment of the present application may be combined, divided, and deleted according to actual needs.
Furthermore, the method according to the present application may also be implemented as a computer program or computer program product comprising computer program code instructions for performing some or all of the steps of the above-described method of the present application.
Alternatively, the present application may also be embodied as a non-transitory machine-readable storage medium (or computer-readable storage medium, or machine-readable storage medium) having stored thereon executable code (or a computer program, or computer instruction code) which, when executed by a processor of an electronic device (or electronic device, server, etc.), causes the processor to perform part or all of the various steps of the above-described method according to the present application.
Those of skill would further appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the applications disclosed herein may be implemented as electronic hardware, computer software, or combinations of both.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems and methods according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
Having described embodiments of the present application, the foregoing description is intended to be exemplary, not exhaustive, and not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen in order to best explain the principles of the embodiments, the practical application, or improvements made to the technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.
Claims (4)
1. A crop growth detection method based on 5G transmission is characterized by comprising the following steps:
collecting a hyperspectral image of a crop; the hyperspectral images comprise crop images collected under the light with wave bands of 450nm, 550nm and 650 nm;
identifying the hyperspectral image through a crop identification model to obtain the variety type of the crop;
the crop identification model carries out data matching of crop variety types through the edge information and the color information of crops, and finds the crop variety types based on the edge information and the color information of the crops according to a mapping table of the edge information and the color information of the crops and the crop variety types and through a mapping relation between the edge information and the color information of the crops and the crop variety types;
calculating Euclidean distance between the collected edge information and color information and variety identification information of each crop in the mapping table, and finding out information with the minimum distance, wherein the corresponding crop variety type is the variety type of the crop to be detected;
searching according to the variety type to obtain the characteristic wave band type of the crop; under the characteristic wave band corresponding to the characteristic wave band type, the characteristic quantity of the hyperspectral image of the crop is sequenced into the first N bits in the characteristic quantity sequence; the characteristic quantity sequence is a sequence obtained by sequentially arranging the characteristic quantities of the hyperspectral images from high to low under M wave bands;
extracting and obtaining N characteristic images of the crops under corresponding characteristic wave bands based on the characteristic wave band types; wherein N is a positive integer and N is less than M;
acquiring a characteristic waveband spectrum data set of the crop;
fitting a regression relation between the characteristic waveband spectrum data set of the crops and the growth condition information by using a partial least square method to obtain the growth condition monitoring model;
the growth condition monitoring model is formed by training a mathematical model established based on a partial least square method through a crop training set; the hyperspectral image of the crop is input, and the growth condition information of the crop is output; adopting a convolutional neural network; training the convolutional neural network by using a gradient cutting method or a regularization method based on a crop training set to obtain a growth condition monitoring model;
the crop training set comprises: a set of spectral data of the crop; the spectral data group is a hyperspectral image with growth condition information as a label;
analyzing the characteristic image through a growth condition monitoring model to obtain growth condition information of the crop;
the growth condition information comprises: appearance parameters, crop ingredients and ingredient contents; grading the growth condition of the crops by combining the appearance parameters and the component content of the crops;
performing data analysis according to the growth condition information to obtain a cultivation strategy of the crop;
the culture strategy comprises the following steps: elements to be supplemented and suggested supply amount;
grading the growth condition of the crops by using a grading model to obtain the growth condition grade of the crops;
combining the growth condition information and the growth condition grade to obtain a culture strategy of the crop;
a crop cultivation strategy database is established in advance, and the cultivation strategy database comprises a plurality of sets of cultivation strategies of a plurality of crops under a plurality of growth condition grades for a plurality of growth condition information; combining the growth condition information and the growth condition grade of the crop to be detected, and performing data calling and data matching on the culture strategy database to obtain a corresponding culture strategy of the crop to be detected;
the shape parameters of the crops are taken into reference by a plurality of sets of cultivation strategies in the cultivation strategy database to form a crop cultivation strategy which is made based on four-dimensional data of the crop species, the component content, the growth condition grade and the shape parameters;
the grading model is obtained by establishing a classification model by taking the growth condition information as a data set characteristic and utilizing an SVM algorithm and a Gaussian core to check the growth condition information and the growth condition grade.
2. The crop growth detection method based on 5G transmission according to claim 1, wherein the identifying the hyperspectral image by a crop identification model to obtain the variety type of the crop comprises:
segmenting the hyperspectral image by using an image local gravity center algorithm to obtain a single crop image;
extracting the single crop image by using a Canny operator to obtain the edge information of the crop;
performing pseudo-color synthesis on the hyperspectral image to obtain color information of the crop;
and matching the crop species based on the edge information and the color information to obtain the variety type of the crop.
3. The method for detecting crop growth based on 5G transmission according to claim 1,
before the characteristic wave band type of the crop is obtained according to the variety type retrieval, the method comprises the following steps: collecting the characteristic wave band types of different crops, and establishing a characteristic wave band database;
the obtaining of the characteristic wave band type of the crop according to the variety type retrieval comprises: and matching the crop variety type with the characteristic waveband type according to a characteristic waveband database to obtain the characteristic waveband type of the crop.
4. The crop growth detection method based on 5G transmission according to claim 3, wherein the collecting of the characteristic wave band types of different crops and establishing the characteristic wave band database comprises the following steps:
collecting the crop spectral image;
obtaining characteristic quantities of the crops under different characteristic wave bands by adopting a continuous selection projection algorithm;
sorting the characteristic quantity from high to low;
and selecting N characteristic wave bands corresponding to the first N bits of the characteristic quantity in the sequence as the characteristic wave band types of the crops.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110438384.5A CN113252584B (en) | 2021-04-22 | 2021-04-22 | Crop growth detection method and system based on 5G transmission |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110438384.5A CN113252584B (en) | 2021-04-22 | 2021-04-22 | Crop growth detection method and system based on 5G transmission |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113252584A CN113252584A (en) | 2021-08-13 |
| CN113252584B true CN113252584B (en) | 2022-09-27 |
Family
ID=77221504
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110438384.5A Active CN113252584B (en) | 2021-04-22 | 2021-04-22 | Crop growth detection method and system based on 5G transmission |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113252584B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114545833B (en) * | 2022-03-02 | 2023-05-02 | 深圳市丰农数智农业科技有限公司 | Intelligent interactive processing system of facility agriculture based on internet of things |
| CN116907576A (en) * | 2023-07-13 | 2023-10-20 | 广东省农业科学院设施农业研究所 | Automatic seedling detection system, method and electronic equipment |
| CN116897818A (en) * | 2023-07-18 | 2023-10-20 | 广东省农业科学院设施农业研究所 | Water planting nutrient solution regulating and controlling method and regulating and controlling system |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101915738A (en) * | 2010-06-23 | 2010-12-15 | 江苏大学 | Method and device for rapidly detecting nutritional information of tea tree based on hyperspectral imaging technique |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102385694B (en) * | 2010-09-06 | 2014-04-16 | 中国科学院遥感与数字地球研究所 | Hyperspectral identification method for land parcel-based crop variety |
| CN109345555B (en) * | 2018-10-15 | 2020-08-25 | 中科卫星应用德清研究院 | Method for identifying rice based on multi-temporal multi-source remote sensing data |
| CN111474117B (en) * | 2020-04-24 | 2022-11-01 | 中国科学院空天信息创新研究院 | Method and device for monitoring crop diseases |
| CN112434569B (en) * | 2020-11-09 | 2024-03-12 | 吉林化工学院 | Unmanned aerial vehicle thermal imaging system |
| CN112668406B (en) * | 2020-12-10 | 2022-07-12 | 哈尔滨师范大学 | Effective screening method for soybean varieties |
-
2021
- 2021-04-22 CN CN202110438384.5A patent/CN113252584B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101915738A (en) * | 2010-06-23 | 2010-12-15 | 江苏大学 | Method and device for rapidly detecting nutritional information of tea tree based on hyperspectral imaging technique |
Non-Patent Citations (2)
| Title |
|---|
| 高光谱技术在茶叶品种检测中的应用;王化 等;《科技创新导报》;20170331;第14卷(第9期);第101-103页 * |
| 高光谱遥感图像降维技术研究;张悦 等;《中国优秀硕士学位论文全文数据库 信息科技辑》;20190115(第12期);第I140-993页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113252584A (en) | 2021-08-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Tu et al. | Detection of passion fruits and maturity classification using Red-Green-Blue Depth images | |
| CN113252584B (en) | Crop growth detection method and system based on 5G transmission | |
| Amatya et al. | Detection of cherry tree branches with full foliage in planar architecture for automated sweet-cherry harvesting | |
| Bargoti et al. | Image segmentation for fruit detection and yield estimation in apple orchards | |
| Li et al. | Immature green citrus fruit detection and counting based on fast normalized cross correlation (FNCC) using natural outdoor colour images | |
| Shirzadifar | Automatic weed detection system and smart herbicide sprayer robot for corn fields | |
| Fu et al. | Fast detection of banana bunches and stalks in the natural environment based on deep learning | |
| Oppenheim et al. | Detecting tomato flowers in greenhouses using computer vision | |
| Wang et al. | Recognition and localization of occluded apples using K-means clustering algorithm and convex hull theory: a comparison | |
| Kierdorf et al. | Behind the leaves: Estimation of occluded grapevine berries with conditional generative adversarial networks | |
| Ji et al. | Target recognition method of green pepper harvesting robot based on manifold ranking | |
| Ge et al. | Three dimensional apple tree organs classification and yield estimation algorithm based on multi-features fusion and support vector machine | |
| Liu et al. | Small unopened cotton boll counting by detection with MRF-YOLO in the wild | |
| Li et al. | An improved binocular localization method for apple based on fruit detection using deep learning | |
| Wang et al. | Combining SUN-based visual attention model and saliency contour detection algorithm for apple image segmentation | |
| CN111950391A (en) | Fruit tree bud recognition method and device | |
| Nagar et al. | Pest detection on leaf using image processing | |
| Zhang et al. | Yield estimation of citrus fruit using rapid image processing in natural background | |
| Liu et al. | SE-Mask R-CNN: An improved Mask R-CNN for apple detection and segmentation | |
| Septiarini et al. | Image processing techniques for tomato segmentation applying k-means clustering and edge detection approach | |
| CN114596274A (en) | Natural background citrus greening disease detection method based on improved Cascade RCNN network | |
| Liu et al. | An improved lightweight network based on deep learning for grape recognition in unstructured environments | |
| Wang et al. | A transformer-based mask R-CNN for tomato detection and segmentation | |
| Zhang et al. | An improved target detection method based on YOLOv5 in natural orchard environments | |
| CN113723833B (en) | Method, system, terminal equipment and storage medium for evaluating quality of forestation actual results |
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 |