CN113575388A - Agricultural intelligent irrigation system based on artificial intelligence and big data - Google Patents
Agricultural intelligent irrigation system based on artificial intelligence and big data Download PDFInfo
- Publication number
- CN113575388A CN113575388A CN202110873969.XA CN202110873969A CN113575388A CN 113575388 A CN113575388 A CN 113575388A CN 202110873969 A CN202110873969 A CN 202110873969A CN 113575388 A CN113575388 A CN 113575388A
- Authority
- CN
- China
- Prior art keywords
- humidity
- image
- obtaining
- area
- soil
- 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.)
- Withdrawn
Links
- 230000002262 irrigation Effects 0.000 title claims abstract description 25
- 238000003973 irrigation Methods 0.000 title claims abstract description 25
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 14
- 238000009826 distribution Methods 0.000 claims abstract description 102
- 239000002689 soil Substances 0.000 claims abstract description 83
- 238000010586 diagram Methods 0.000 claims abstract description 34
- 239000003621 irrigation water Substances 0.000 claims abstract description 25
- 238000001514 detection method Methods 0.000 claims abstract description 17
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 15
- 238000000034 method Methods 0.000 claims description 15
- 239000011159 matrix material Substances 0.000 claims description 9
- 238000004519 manufacturing process Methods 0.000 claims description 7
- 239000013598 vector Substances 0.000 claims description 6
- 238000006243 chemical reaction Methods 0.000 claims description 2
- 239000002699 waste material Substances 0.000 abstract description 4
- 230000011218 segmentation Effects 0.000 description 8
- 230000007613 environmental effect Effects 0.000 description 4
- 230000008901 benefit Effects 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 238000005192 partition Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 241000209140 Triticum Species 0.000 description 1
- 235000021307 Triticum Nutrition 0.000 description 1
- 241001464837 Viridiplantae Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- NNBFNNNWANBMTI-UHFFFAOYSA-M brilliant green Chemical group OS([O-])(=O)=O.C1=CC(N(CC)CC)=CC=C1C(C=1C=CC=CC=1)=C1C=CC(=[N+](CC)CC)C=C1 NNBFNNNWANBMTI-UHFFFAOYSA-M 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01G—HORTICULTURE; CULTIVATION OF VEGETABLES, FLOWERS, RICE, FRUIT, VINES, HOPS OR SEAWEED; FORESTRY; WATERING
- A01G25/00—Watering gardens, fields, sports grounds or the like
- A01G25/16—Control of watering
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01G—HORTICULTURE; CULTIVATION OF VEGETABLES, FLOWERS, RICE, FRUIT, VINES, HOPS OR SEAWEED; FORESTRY; WATERING
- A01G25/00—Watering gardens, fields, sports grounds or the like
- A01G25/16—Control of watering
- A01G25/167—Control by humidity of the soil itself or of devices simulating soil or of the atmosphere; Soil humidity sensors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Forestry; Mining
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Environmental Sciences (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Water Supply & Treatment (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Marine Sciences & Fisheries (AREA)
- Animal Husbandry (AREA)
- Agronomy & Crop Science (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Primary Health Care (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Mining & Mineral Resources (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Soil Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of artificial intelligence, in particular to an agricultural intelligent irrigation system based on artificial intelligence and big data. The system comprises a growth state detection unit: acquiring a color distribution characteristic image and a texture characteristic image of the crop according to the RGB image of the crop to obtain a growth state distribution map of the crop; a humidity distribution acquisition unit: combining the humidity sensor with the growth state distribution map to obtain the soil humidity of each position in the farmland, and obtaining the humidity distribution map of all the positions in the farmland according to the soil humidity; water consumption prediction unit: and obtaining the drought degree of the drought area in the farmland according to the growth state distribution diagram and the humidity distribution diagram so as to obtain the irrigation water consumption. The drought area is irrigated by combining the growth state of crops and the soil humidity of the position, the irrigation pertinence is guaranteed, the irrigation water consumption is controlled, and the waste of water resources is avoided.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to an agricultural intelligent irrigation system based on artificial intelligence and big data.
Background
With the development of agriculture and the popularization and application of various agricultural equipment, large-area farmlands and large-scale agricultural planting can be uniformly managed at present or in the future, and an intelligent agricultural management system frees the labor force of farmers and improves the yield of crops.
The agricultural irrigation system of intelligence is the indispensable agricultural system in large-scale farmland, and current irrigation system can be through laying reasonable pipeline, water waste when avoiding traditional farmland irrigation such as reasonable water source management, blind flood irrigation scheduling problem when irrigating crops, but current irrigation system still lacks intelligent control system, and is high inadequately to the utilization ratio of water resource, does not come reasonable utilization water resource to the growing environment and the state of crops, exists the space that can improve.
Disclosure of Invention
In order to solve the technical problems, the invention aims to provide an agricultural intelligent irrigation system based on artificial intelligence and big data, and the adopted technical scheme is as follows:
the embodiment of the invention provides an agricultural intelligent irrigation system based on artificial intelligence and big data, which comprises:
the growth state detection unit is used for acquiring RGB images of crops in a farmland, obtaining a color distribution characteristic image and a texture characteristic image of the crops according to the RGB images, and combining the color distribution characteristic image and the texture characteristic image to obtain a growth state distribution diagram of the crops; the growth status distribution map refers to the growth status of the crop at each location in the field;
the humidity distribution acquisition unit is used for obtaining the soil humidity of each position in the farmland by combining a humidity sensor and the growth state distribution map, and acquiring the humidity distribution maps of all the positions in the farmland according to the soil humidity;
and the water consumption prediction unit is used for obtaining the drought degree of the drought area in the farmland according to the growth state distribution diagram and the humidity distribution diagram, and obtaining the irrigation water consumption of the drought area by combining the drought degree, the environment temperature and an irrigation water consumption model constructed by the historical drought degree, the historical environment temperature and the historical irrigation water consumption.
Preferably, the method for acquiring the color feature distribution feature map in the growth state detection unit includes:
and performing HSV space conversion on the RGB image to obtain a hue channel image and a brightness image, obtaining a color contrast gain coefficient from a blue channel image, a red channel image, a green channel image and the brightness image of the RGB image, and obtaining the color distribution characteristic diagram of the crop by combining the color contrast gain coefficient and the hue channel image.
Preferably, the method for acquiring the texture feature image map in the growth state detection unit includes:
converting the RGB image into a gray image, performing Gaussian blur on the gray image to obtain a first gray image, and performing difference between the gray image and the first gray image to obtain an edge texture image;
and fusing the gray level image and the edge texture image to obtain an edge binary image, and further acquiring the texture feature image map of the edge binary image based on a gray level co-occurrence matrix.
Preferably, the humidity distribution acquisition unit includes a humidity detection unit, the humidity detection unit is used for combining a humidity sensor and obtaining the soil humidity of each position in the farmland according to the growth state distribution diagram, and then the humidity detection unit includes:
and performing super-pixel division on the production state distribution diagram to obtain a plurality of divided areas, and obtaining the soil humidity of each position according to the indication number of the humidity sensors arranged in the divided areas.
Preferably, the method of deriving the soil moisture at each location from indications of the moisture sensors arranged in the partitioned areas in the moisture detection unit comprises:
taking the partitioned area containing the humidity sensor as a known humidity area, otherwise, an unknown humidity area; obtaining a plurality of known humidity areas adjacent to the unknown humidity area, and fitting a quadratic polynomial curve equation between the pixel value mean value and the soil humidity according to the soil humidity of each known humidity area and the pixel value mean value of the known humidity area on the production state distribution diagram;
predicting the soil humidity of the unknown humidity area by combining the pixel value mean value of the unknown humidity area on the production state distribution diagram and the quadratic polynomial curve equation; deriving the soil moisture for each location from the soil moisture of the area of known moisture and the area of unknown moisture.
Preferably, the humidity distribution acquiring unit includes an image acquiring unit, the image acquiring unit is configured to acquire a humidity distribution map of all positions in the farmland according to the soil humidity, and the image acquiring unit includes:
obtaining a predicted humidity distribution map of all positions in the farmland according to the soil humidity of each position;
respectively obtaining a predicted humidity value of each known humidity area by using the quadratic polynomial curve equation, and obtaining an error value of a curve corresponding to the quadratic polynomial curve equation according to the predicted humidity values of the plurality of known humidity areas and the corresponding soil humidity; the error value refers to the accuracy of predicting the unknown humidity region by using a plurality of adjacent known humidity regions;
determining the humidity accuracy rate of each known humidity area and each unknown humidity area according to the error value, and obtaining a humidity accuracy rate distribution map of all positions in the farmland according to the humidity accuracy rates;
acquiring the central point position of each known humidity area and each unknown humidity area, and obtaining the soil humidity of all positions by utilizing a linear difference according to the soil humidity of the central point position to form a prior humidity distribution map;
and combining the prior humidity distribution map, the predicted humidity distribution map and the humidity accuracy rate distribution map to obtain the humidity distribution map.
Preferably, when the soil humidity of all the areas with unknown humidity is obtained in the image obtaining unit, a plurality of error values exist in each of the areas with known humidity and the areas with unknown humidity, and the minimum error value is reserved.
Preferably, the method for obtaining the drought degree of the drought area in the farmland in the water consumption prediction unit according to the growth state distribution diagram and the humidity distribution diagram comprises the following steps:
merging the growth state distribution map and the humidity distribution map into a dual-channel first image, forming a two-dimensional vector of each pixel point by pixel values of two channels in the first image, calculating the modular length sum of the two-dimensional vectors of all the pixel points in the drought region, and taking the modular length sum as the drought degree.
The embodiment of the invention has at least the following beneficial effects: the growth state of crops is obtained according to the color and the texture of the crops in the collected image, the growth condition of the crops can be accurately described, the soil humidity of all positions in a farmland is obtained according to the soil humidity of the known position and by combining the growth state of the crops, the drought region is irrigated by combining the growth state of the crops and the soil humidity of the positions, the irrigation pertinence is guaranteed, the irrigation water consumption is controlled, and the waste of water resources is avoided.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions and advantages of the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a block diagram of an agricultural intelligent irrigation system based on artificial intelligence and big data according to an embodiment of the present invention;
fig. 2 is a block diagram of a humidity distribution acquiring unit according to an embodiment of the present invention.
Detailed Description
In order to further explain the technical means and effects of the present invention adopted to achieve the predetermined invention purpose, the following detailed description, the structure, the features and the functions of an agricultural intelligent irrigation system based on artificial intelligence and big data according to the present invention are described in detail with reference to the accompanying drawings and the preferred embodiments. In the following description, different "one embodiment" or "another embodiment" refers to not necessarily the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following describes a specific scheme of the agricultural intelligent irrigation system based on artificial intelligence and big data in detail with reference to the accompanying drawings.
The embodiment of the invention aims at the following specific scenes: in large-scale and open-air cultivated farmlands, crops refer to green plants such as wheat.
It should be noted that the growing environment of the crops in the embodiment of the present invention includes soil humidity and ambient temperature.
Referring to fig. 1, an embodiment of the present invention provides an agricultural intelligent irrigation system based on artificial intelligence and big data, which includes a growth state detection unit 10, a humidity distribution acquisition unit 20, and a water consumption prediction unit 30, wherein,
the growth state detection unit 10 is used for acquiring RGB images of crops in a farmland, obtaining a color distribution characteristic image and a texture characteristic image of the crops according to the RGB images, and combining the color distribution characteristic image and the texture characteristic image to obtain a growth state distribution diagram of the crops; the growth state profile refers to the growth state of the crop at each location in the field.
Specifically, unmanned aerial vehicle carries on high definition RGB camera, and the camera visual angle is overlooked downwards, and unmanned aerial vehicle low-altitude flight gathers the image data of crops in the farmland, every image data of gathering, calculates the growth state of every position department crops in the image field of vision according to image data, and wherein the method of obtaining the growth state of crops according to image data is:
the method comprises the steps of obtaining RGB images of crops in a farmland, and carrying out HSV space transformation on the RGB images to obtain hue channel images H and brightness images s, wherein the hue channel images H are gray images and represent hue distribution of the crops in the farmland. Since the minimum gray value of the hue channel image H is 0 and the maximum gray value is 359, the invention performs thresholding on the hue channel image H: setting the gray value of the pixel point with the gray value larger than 110 on the hue channel image H as 110, setting the gray value of the pixel point with the gray value smaller than 25 as 25, and keeping the gray values of other pixel points unchanged, namely, the gray values are [25,110 ]]Pixel points of interval in which the gray value [25, 110%]Belongs to the hue interval of crops and land. Subtracting 25 from the gray value of all pixel points in the hue channel image H, and dividing by (110-25), namely
So that the maximum gray value is not more than 1.0.
Based on a priori knowledge: the color of crops with good growth state is emerald green, and the brightness is not too high; the crops with bad growth state are yellow, and have high brightness, even the areas without crops are completely earthy yellow, and have high brightness. Because yellow and green have larger gray values on a green channel, green has smaller gray values on a red channel and a blue channel, and the gray value of yellow on the red channel is far larger than that on the blue channel, a blue channel image b, a red channel image r and a green channel image g of an RGB image are obtained, and color contrast increase is calculated by utilizing the blue channel image b, the red channel image r, the green channel image g and a brightness image sCoefficient of benefit
Where b, r, g, s are all matrices, then
And exp (-2s) are also a matrix; an e indicates a hadamard product of two matrices, i.e. multiplying corresponding elements of the two matrices; the color contrast gain factor alpha is also a matrix, i.e. an image.
It should be noted that the smaller b-r, the smaller the difference between the red channel image and the blue channel image, the closer the color of the crop is to green, and the larger b-r, the closer the color of the crop is to yellow, therefore
A larger indicates that the crop is closer to green in color, otherwise the color is closer to yellow. It is known that although the gray scale values of green and yellow are both larger on the green channel, the gray scale value of yellow on the green channel is larger than that of green because the gray scale value of yellow on the green channel is larger when the growth state of the crop is good, so the embodiment of the invention introduces exp (-2s) to make the dark green crop have a higher color contrast gain coefficient alpha, and the yellow crop have a lower color contrast gain coefficient alpha, wherein the purpose of the color contrast gain coefficient alpha is to increase the color discrimination at the positions of the crops with good growth state and with poor growth state to ensure that the accurate growth state of the crop is obtained.
Normalizing the color contrast gain coefficient alpha, and then performing Gaussian filtering on the color contrast gain coefficient alpha and the hue channel image H by utilizing a Gaussian kernel of 3 multiplied by 3, wherein the filtering does not change the size of the color contrast gain coefficient alpha; and combining the filtered color contrast gain coefficient and the color channel image to obtain a color distribution characteristic graph F1 ═ α &.
Further, the RGB image is converted into a grayscale image I, the grayscale image is gaussian-blurred by using a 3 × 3 gaussian kernel to obtain a first grayscale image I1, and the grayscale image I and the first grayscale image I1 are subtracted to obtain an edge texture image I2, i.e., I2 — I1, where I2 represents a high-frequency texture of the grayscale image I, i.e., an edge texture. Fusing the gray level image I and the edge texture image I2 to obtain an edge binary image, wherein the method for obtaining the edge binary image comprises the following steps: let I3 be I + α I2, where I3 represents a fused image of the grayscale image I and the edge texture image I2, and the color contrast gain coefficient α is used to focus more on the edge texture of a crop with a better growth state, and the focus of the edge texture at a position with a poor growth state or even without the growth of the crop is low, so that on one hand, the texture of the fused image I3 is clearer, and on the other hand, the good growth state and the bad crop have stronger contrast characteristics, which is helpful for obtaining a more accurate growth state of the crop; and (5) carrying out normalization processing on the fused image I3, and acquiring an edge binary image I4 of the fused image I3 by using a Canny operator.
Acquiring a texture feature image F2 of the edge binary image I4 based on the gray level co-occurrence matrix, wherein the acquiring method of the texture feature image F2 comprises the following steps: on the edge binary image I4, a sub-image is formed with a small window. Calculating the entropy values of the gray level co-occurrence matrix and the gray level co-occurrence matrix of the sub-images in the small window, assigning the entropy value to the central point of the small window, thus completing the calculation of the texture characteristics of the first small window, then moving the small window by the distance of one pixel point to form the sub-image of another small window, repeatedly calculating the entropy values of the gray level co-occurrence matrix and the gray level co-occurrence matrix of the sub-images in the small window, assigning the entropy values to the central point of the small window, traversing the whole edge binary image I4 in this way, enabling the whole edge binary image I4 to generate a texture characteristic image F2, wherein the texture characteristic image is as large as the edge binary image, each pixel value of the texture characteristic image indicates the complexity or the chaos degree of the texture, the larger the pixel value indicates that the texture distribution is more complex, and the better the growth state of crops is indicated, namely the crops have luxuriant branches and leaves.
The color profile F1 and the texture profile F2 were combined to obtain a growth status profile F0 ═ F1 ═ F2, in which the growth status profile indicates the growth status of the crop at each position in the field, and the larger the pixel value in the growth status profile F0 indicates the better the growth status of the crop at the position of this pixel point.
It should be noted that, the growth state of the crops of every position in the farmland in this image field of vision can be acquireed to a RGB image, and unmanned aerial vehicle splices all RGB images at the in-process of taking photo by plane and can obtain the growth state of the crops of every position in whole farmland, and wherein, unmanned aerial vehicle accessible carries on laser radar and obtains the height of unmanned aerial vehicle apart from ground, and unmanned aerial vehicle is through the positioner who carries on or utilize SLAM system to obtain unmanned aerial vehicle's position.
The humidity distribution obtaining unit 20 is configured to obtain soil humidity at each position in the farmland by combining the humidity sensor and the growth state distribution map, and obtain the humidity distribution map at all positions in the farmland according to the soil humidity.
Specifically, referring to fig. 2, the embodiment of the present invention provides a block diagram of a humidity distribution acquisition unit, and the humidity distribution acquisition unit 20 includes a humidity detection unit 21 and an image acquisition unit 22.
The humidity detection unit 21 is used for obtaining soil humidity of each position in the farmland by combining the humidity sensor and the growth state distribution diagram, and the soil humidity obtaining method comprises the following steps:
a humidity sensor is inserted into soil, and a humidity sensor is inserted at regular intervals and used for obtaining the soil humidity at the position of the humidity sensor.
Preferably, the interval in the embodiment of the present invention is set to 50 meters.
Performing superpixel segmentation on the growth state distribution diagram F0 by using SLIC (simple linear iterative clustering) to obtain a plurality of segmentation areas, wherein crops in each segmentation area have approximate growth states, and the soil humidity of each position is obtained by indicating numbers of humidity sensors arranged in the segmentation areas, namely the segmentation areas where each humidity sensor is located are obtained, and the soil humidity of each position in the segmentation areas is indicated number of the humidity sensors; if a plurality of humidity sensors are corresponding to one divided area, the soil humidity of each position in the divided area is the indicated average value of the humidity sensors.
Acquiring all the divided areas containing the humidity sensors, and taking the divided areas containing the humidity sensors as known humidity areas, wherein the soil humidity of each position in the known humidity areas is the reading of the humidity sensors; if the partitioned area without the humidity sensor is used as the unknown humidity area, the soil humidity at each position in the unknown humidity area cannot be acquired.
For an unknown humidity area, assuming that the unknown humidity area has the most adjacent known humidity areas, acquiring K known humidity areas with the shortest distance to the unknown humidity area, and fitting a quadratic polynomial curve equation between the pixel mean value and the soil humidity according to the soil humidity of each known humidity area and the pixel value mean value of the known humidity area on the production state distribution diagram, namely: respectively acquiring soil humidity of K areas with known humidity to form a sequence a ═ a1,a2,…,ak…,aKIn which a iskThe soil humidity corresponding to the Kth known humidity area; respectively obtaining the pixel value mean values of all pixel points in the growth state distribution diagram F0 of the K known humidity regions to form a sequence b ═ b1，b2，...，bk...，bKIn which b isKThe K & ltth & gt known humidity area is the mean value of the pixel values, and the mean value of the pixel values represents the average growth state of crops; fitting a quadratic polynomial curve equation according to the sequence a and the sequence b by using a RANSAC algorithm
Wherein,
denotes bkThe result of the mapping on the curve.
Preferably, in the embodiment of the present invention, K is 9.
Obtaining the pixel value mean value b of all pixel points of the unknown humidity region in the growth state distribution diagram F00Combined with the mean value b of the pixel values0Obtaining the soil humidity a of the unknown humidity area by a quadratic polynomial curve equation0＝f(b0)。
After the soil humidity of an unknown humidity area is predicted according to K known humidity areas around the unknown humidity area, the unknown humidity area is used as the known humidity area to predict the soil humidity of other unknown humidity areas, further the soil humidity of all the unknown humidity areas is obtained, and the soil humidity of each position is obtained according to the soil humidity of the known humidity area and the soil humidity of the unknown humidity area.
The image obtaining unit 22 is configured to obtain a humidity distribution map of all positions in the farm field according to the soil humidity, where the humidity distribution map is obtained by:
obtaining a predicted humidity distribution diagram F3 of all positions in the farmland according to the soil humidity of each position; in the step of obtaining the soil humidity value of any one unknown humidity area, one unknown humidity area and the corresponding K known humidity areas correspond to the error value of the same curve, namely, the predicted humidity value of each known humidity area is respectively obtained by utilizing a quadratic polynomial curve equation, and the error value of the curve corresponding to the quadratic polynomial curve equation is obtained according to the predicted humidity values of the K known humidity areas and the corresponding soil humidity, namely, the error value of the curve corresponding to the quadratic polynomial curve equation is obtained, namely
The error value e is the accuracy of predicting the unknown humidity area by using a plurality of adjacent known humidity areas, if the error value is small, it is feasible to set the soil humidity of each position in the K known humidity areas as the indication number of the corresponding humidity sensor in each area, and the soil humidity of each position in the unknown humidity area can be accurately predicted.
After obtaining the soil humidity of all the unknown humidity areas, an error value corresponding to a plurality of curves exists in each of the known humidity areas and the unknown humidity areas, the minimum error value x is taken as the error value of the area, and the smaller the x, the greater the reliability or the greater the accuracy of the soil humidity of each position calculated in each area is; determining the humidity accuracy of each known humidity region and each unknown humidity region according to the error value, namely regarding exp (-x) as the humidity accuracy of each region; and obtaining the humidity accuracy distribution map of all positions in the farmland according to the humidity accuracy, and after obtaining the humidity accuracy of each region, distributing an accuracy for predicting the soil humidity of all positions on the humidity distribution map F3 to represent the reliability of the soil humidity of each position, so that the accuracy of the soil humidity of all positions is called as a humidity accuracy distribution map beta.
Further, the central point position of each partition area is obtained, each central point position corresponds to one soil humidity, the soil humidity of all the positions is interpolated according to the soil humidity of the central point positions of all the partition areas by using linear interpolation, the soil humidity of all the positions obtained by using the linear interpolation is called as a prior humidity distribution diagram F4, and the prior humidity distribution diagram F4 shows an empirical soil humidity obtained only according to different central point positions after the growth state of crops in the area is ignored.
Combining the predicted humidity profile F3, the humidity accuracy profile β and the prior humidity profile F4 to obtain a humidity profile F5, i.e., F5 ═ β ≥ GasBlur (F3) + (1-. beta.) F4, wherein GasBlur (F3) indicates that the predicted humidity profile F3 is gaussian blurred with a gaussian blur kernel of 11 × 11, in order to soften the edge of the predicted humidity profile F3; larger β indicates a greater interest in the predicted humidity distribution in the predicted humidity profile F3, and smaller β indicates a greater interest in the empirical humidity distribution in the prior humidity profile F4.
The water consumption prediction unit 30 is configured to obtain a drought degree of the drought region in the farmland according to the growth state distribution map and the humidity distribution map, and obtain an irrigation water consumption of the drought region by combining the drought degree, the environmental temperature, and an irrigation water consumption model constructed by the historical drought degree, the historical environmental temperature, and the historical irrigation water consumption.
Specifically, the growth state distribution diagram F0 and the humidity distribution diagram F5 are combined into a dual-channel first image, the first image is input into a semantic segmentation network to obtain a drought semantic region or a plurality of drought semantic regions of the crop, the drought semantic region refers to a connected region where the growth state and the soil humidity of the crop are small and the local area is large, that is, the growth state of most of the crop in the drought semantic region is poor, the soil humidity is low, but the growth state of a small part of the crop is good, and the soil humidity is high. In the embodiment of the invention, only crops in the drought semantic region need to be irrigated, and other regions do not irrigate or irrigate less.
Preferably, the semantic segmentation network in the embodiment of the present invention is MaskRCNN.
Taking a drought semantic area as an example to obtain the irrigation water consumption of the drought semantic area: acquiring all pixel points of the drought semantic region in the first image, forming a two-dimensional vector of each pixel point by pixel values of two channels in the first image, calculating the sum of the modular lengths of the two-dimensional vectors of all the pixel points, and taking the sum of the modular lengths as the drought degree G. Obtaining the irrigation water consumption of the drought semantic region by an irrigation water consumption model according to the environment temperature T and the drought degree G during image acquisition, wherein the irrigation water consumption model is a ═ gamma GT + delta, and a is the irrigation water consumption; gamma and delta are optimization parameters.
It should be noted that the irrigation water consumption model is constructed based on historical data, the growth state distribution of crops in the drought semantic region and the irrigation water consumption of each drought semantic region after the crops are irrigated for a period of time are firstly obtained before each irrigation in the history, and the data are stored in the big data system, and the data amount in the big data system is increased along with the increase of the irrigation times.
Based on data of each irrigation in the big data system, if the average value of the growth state distribution in the drought semantic region is larger than a certain threshold value during a certain irrigation, the growth state of crops after the irrigation is good is shown, the irrigation water consumption a, the environment temperature T and the drought degree b of the drought semantic region are obtained, multiple groups of data (a, T and b) are obtained in the big data system, and the optimal parameters gamma and delta in the irrigation water consumption model are fitted by utilizing a SAC RANalgorithm according to the data (a, T and b).
To sum up, the embodiment of the invention provides an agricultural intelligent irrigation system based on artificial intelligence and big data, the system collects RGB images of crops in a farmland at a growth state detection unit 10, obtains a color distribution characteristic image and a texture characteristic image map of the crops according to the RGB images, and combines the color distribution characteristic image and the texture characteristic image map to obtain a growth state distribution map of the crops; the humidity distribution obtaining unit 20 combines the humidity sensor and the growth state distribution map to obtain the soil humidity of each position in the farmland, and obtains the humidity distribution map of all the positions in the farmland according to the soil humidity; the water consumption prediction unit 30 obtains the drought degree of the drought region in the farmland according to the growth state distribution diagram and the humidity distribution diagram, and obtains the irrigation water consumption of the drought region by combining the drought degree, the environmental temperature and an irrigation water consumption model constructed by the historical drought degree, the historical environmental temperature and the historical irrigation water consumption. The drought area is irrigated by combining the growth state of crops and the soil humidity of the position, the irrigation pertinence is guaranteed, the irrigation water consumption is controlled, and the waste of water resources is avoided.
It should be noted that: the precedence order of the above embodiments of the present invention is only for description, and does not represent the merits of the embodiments. And specific embodiments thereof have been described above. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments.
The present invention is not limited to the above preferred embodiments, and any modifications, equivalent replacements, improvements, etc. within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (8)
1. An agricultural intelligent irrigation system based on artificial intelligence and big data, characterized in that, the system comprises:
the growth state detection unit is used for acquiring RGB images of crops in a farmland, obtaining a color distribution characteristic image and a texture characteristic image of the crops according to the RGB images, and combining the color distribution characteristic image and the texture characteristic image to obtain a growth state distribution map of the crops; the growth status distribution map refers to the growth status of the crop at each location in the field;
the humidity distribution acquisition unit is used for obtaining the soil humidity of each position in the farmland by combining a humidity sensor and the growth state distribution map, and acquiring the humidity distribution maps of all the positions in the farmland according to the soil humidity;
and the water consumption prediction unit is used for obtaining the drought degree of the drought area in the farmland according to the growth state distribution diagram and the humidity distribution diagram, and obtaining the irrigation water consumption of the drought area by combining the drought degree, the environment temperature and an irrigation water consumption model constructed by the historical drought degree, the historical environment temperature and the historical irrigation water consumption.
2. The system according to claim 1, wherein the method for obtaining the color feature distribution feature map in the growth status detection unit comprises:
and performing HSV space conversion on the RGB image to obtain a hue channel image and a brightness image, obtaining a color contrast gain coefficient from a blue channel image, a red channel image, a green channel image and the brightness image of the RGB image, and obtaining the color distribution characteristic diagram of the crop by combining the color contrast gain coefficient and the hue channel image.
3. The system of claim 1, wherein the method for obtaining the texture feature image map in the growth status detection unit comprises:
converting the RGB image into a gray image, performing Gaussian blur on the gray image to obtain a first gray image, and performing difference between the gray image and the first gray image to obtain an edge texture image;
and fusing the gray level image and the edge texture image to obtain an edge binary image, and further acquiring the texture feature image map of the edge binary image based on a gray level co-occurrence matrix.
4. The system as claimed in claim 1, wherein the moisture distribution acquiring unit includes a moisture detecting unit for obtaining soil moisture at each position in the agricultural field in combination with a moisture sensor and the growth status map, the moisture detecting unit includes:
and performing super-pixel division on the production state distribution diagram to obtain a plurality of divided areas, and obtaining the soil humidity of each position according to the indication number of the humidity sensors arranged in the divided areas.
5. The system of claim 4, wherein the method of deriving the soil moisture for each location from indications of the moisture sensors disposed in the partitioned areas in the moisture detection unit comprises:
taking the partitioned area containing the humidity sensor as a known humidity area, otherwise, an unknown humidity area; obtaining a plurality of known humidity areas adjacent to the unknown humidity area, and fitting a quadratic polynomial curve equation between the pixel value mean value and the soil humidity according to the soil humidity of each known humidity area and the pixel value mean value of the known humidity area on the production state distribution diagram;
predicting the soil humidity of the unknown humidity area by combining the pixel value mean value of the unknown humidity area on the production state distribution diagram and the quadratic polynomial curve equation; deriving the soil moisture for each location from the soil moisture of the area of known moisture and the area of unknown moisture.
6. The system of claim 1 or 5, wherein the moisture profile acquiring unit comprises an image acquiring unit for acquiring a moisture profile of all locations in the agricultural field according to the soil moisture, and the image acquiring unit comprises:
obtaining a predicted humidity distribution map of all positions in the farmland according to the soil humidity of each position;
respectively obtaining a predicted humidity value of each known humidity area by using the quadratic polynomial curve equation, and obtaining an error value of a curve corresponding to the quadratic polynomial curve equation according to the predicted humidity values of the plurality of known humidity areas and the corresponding soil humidity; the error value refers to the accuracy of predicting the unknown humidity region by using a plurality of adjacent known humidity regions;
determining the humidity accuracy rate of each known humidity area and each unknown humidity area according to the error value, and obtaining a humidity accuracy rate distribution map of all positions in the farmland according to the humidity accuracy rates;
acquiring the central point position of each known humidity area and each unknown humidity area, and obtaining the soil humidity of all positions by utilizing a linear difference according to the soil humidity of the central point position to form a prior humidity distribution map;
and combining the prior humidity distribution map, the predicted humidity distribution map and the humidity accuracy rate distribution map to obtain the humidity distribution map.
7. The system of claim 6, wherein a plurality of said error values exist for each of said known moisture area and said unknown moisture area while obtaining said soil moisture for all of said unknown moisture areas in said image acquisition unit, a minimum of said error values remaining.
8. The system of claim 1, wherein the method of deriving the extent of drought in the arid region of the field from the growth status map and the humidity map in the water usage prediction unit comprises:
merging the growth state distribution map and the humidity distribution map into a dual-channel first image, forming a two-dimensional vector of each pixel point by pixel values of two channels in the first image, calculating the modular length sum of the two-dimensional vectors of all the pixel points in the drought region, and taking the modular length sum as the drought degree.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110873969.XA CN113575388A (en) | 2021-07-30 | 2021-07-30 | Agricultural intelligent irrigation system based on artificial intelligence and big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110873969.XA CN113575388A (en) | 2021-07-30 | 2021-07-30 | Agricultural intelligent irrigation system based on artificial intelligence and big data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113575388A true CN113575388A (en) | 2021-11-02 |
Family
ID=78252874
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110873969.XA Withdrawn CN113575388A (en) | 2021-07-30 | 2021-07-30 | Agricultural intelligent irrigation system based on artificial intelligence and big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113575388A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114431124A (en) * | 2022-03-03 | 2022-05-06 | 郑州麦佳农业科技有限公司 | Intelligent spraying control method for agricultural planting |
| CN114868640A (en) * | 2022-07-12 | 2022-08-09 | 成都秦川物联网科技股份有限公司 | Smart city green space irrigation water distribution method and system based on Internet of things |
| CN116975789A (en) * | 2023-09-21 | 2023-10-31 | 北京建工环境修复股份有限公司 | Intelligent farmland field analysis method, system and medium based on big data |
| CN117370823A (en) * | 2023-12-05 | 2024-01-09 | 恒健达(辽宁)医学科技有限公司 | Spraying control method and system for agricultural planting |
| CN117496208A (en) * | 2023-12-29 | 2024-02-02 | 山东朝辉自动化科技有限责任公司 | Method for acquiring stockpiling information in stock yard in real time |
| CN117743975A (en) * | 2024-02-21 | 2024-03-22 | 君研生物科技(山西)有限公司 | Hillside cultivated land soil environment improvement method |
-
2021
- 2021-07-30 CN CN202110873969.XA patent/CN113575388A/en not_active Withdrawn
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114431124A (en) * | 2022-03-03 | 2022-05-06 | 郑州麦佳农业科技有限公司 | Intelligent spraying control method for agricultural planting |
| CN114431124B (en) * | 2022-03-03 | 2022-12-23 | 郑州麦佳农业科技有限公司 | Intelligent spraying control method for agricultural planting |
| CN114868640A (en) * | 2022-07-12 | 2022-08-09 | 成都秦川物联网科技股份有限公司 | Smart city green space irrigation water distribution method and system based on Internet of things |
| US11693380B1 (en) | 2022-07-12 | 2023-07-04 | Chengdu Qinchuan Iot Technology Co., Ltd. | Distribution methods and distribution systems for greenspace irrigation water of smart cities based on internet of things |
| US11966208B2 (en) | 2022-07-12 | 2024-04-23 | Chengdu Qinchuan Iot Technology Co., Ltd. | Methods and systems for greenspace cultivation and management in smart cities based on Internet of Things |
| CN116975789A (en) * | 2023-09-21 | 2023-10-31 | 北京建工环境修复股份有限公司 | Intelligent farmland field analysis method, system and medium based on big data |
| CN116975789B (en) * | 2023-09-21 | 2023-12-05 | 北京建工环境修复股份有限公司 | Intelligent farmland field analysis method, system and medium based on big data |
| CN117370823A (en) * | 2023-12-05 | 2024-01-09 | 恒健达(辽宁)医学科技有限公司 | Spraying control method and system for agricultural planting |
| CN117370823B (en) * | 2023-12-05 | 2024-02-20 | 恒健达(辽宁)医学科技有限公司 | Spraying control method and system for agricultural planting |
| CN117496208A (en) * | 2023-12-29 | 2024-02-02 | 山东朝辉自动化科技有限责任公司 | Method for acquiring stockpiling information in stock yard in real time |
| CN117496208B (en) * | 2023-12-29 | 2024-03-29 | 山东朝辉自动化科技有限责任公司 | Method for acquiring stockpiling information in stock yard in real time |
| CN117743975A (en) * | 2024-02-21 | 2024-03-22 | 君研生物科技(山西)有限公司 | Hillside cultivated land soil environment improvement method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113575388A (en) | Agricultural intelligent irrigation system based on artificial intelligence and big data | |
| US11317562B2 (en) | Generating a yield map for an agricultural field using classification and regression methods | |
| CN109416413B (en) | Solar energy forecast | |
| Majeed et al. | Apple tree trunk and branch segmentation for automatic trellis training using convolutional neural network based semantic segmentation | |
| CN110765916B (en) | Farmland seedling ridge identification method and system based on semantics and example segmentation | |
| Hočevar et al. | Flowering estimation in apple orchards by image analysis | |
| CN114239756B (en) | Insect pest detection method and system | |
| CN113255434B (en) | Apple identification method integrating fruit characteristics and deep convolutional neural network | |
| CN102542560B (en) | Method for automatically detecting density of rice after transplantation | |
| CN110648020A (en) | Greenhouse crop water demand prediction method and device | |
| CN108629289A (en) | The recognition methods in farmland and system, applied to the unmanned plane of agricultural | |
| CN113065562A (en) | Crop ridge row extraction and leading route selection method based on semantic segmentation network | |
| CN113569772A (en) | Remote sensing image farmland instance mask extraction method, system, equipment and storage medium | |
| CN115687850A (en) | Method and device for calculating irrigation water demand of farmland | |
| CN114898216A (en) | Remote sensing image deep learning segmentation method and system based on superpixels and watershed | |
| Loresco et al. | Segmentation of lettuce plants using super pixels and thresholding methods in smart farm hydroponics setup | |
| CN117876823B (en) | Tea garden image detection method and model training method and system thereof | |
| Yan et al. | High-resolution mapping of paddy rice fields from unmanned airborne vehicle images using enhanced-TransUnet | |
| WO2024178904A1 (en) | Crop water and fertilizer stress decision-making method and apparatus, and mobile phone terminal | |
| Loresco et al. | Vision-based lettuce growth stage decision support system using artificial neural networks | |
| CN115526927A (en) | Rice planting method integrating phenological data and remote sensing big data and area estimation method thereof | |
| CN111753687A (en) | Seedling shortage identification method and device, electronic equipment and computer readable storage medium | |
| Mei et al. | Combing color index and region growing with simple non-iterative clustering for plant segmentation | |
| CN115126686B (en) | Be used for unmanned aerial vehicle of plant protection to carry diaphragm pump control system | |
| Kumar et al. | Detection and counting of tassels for maize crop monitoring using multispectral images |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20211102 |