KR101936586B1 - System for mapping river water-bloom map using data for detecting by gps-based random sampling, and method for the same - Google Patents

Abstract

According to the present invention, provided are a green algae map generating system using data measured by a GPS-based random sampling method and a method thereof. The present invention reasons a tide index by using a GPS-based optimal estimation model equation to calculate the concentration and area of green algae distribution by the random sampling on an autonomous movement path and an autonomous flight path which are not at a predetermined measurement point of a mobile floating body and an autonomous navigation unmanned aerial vehicle. Accordingly, the present invention is able to: interpret the distribution of green algae temporally and spatially; grasp the spatial distribution and temporal movement phenomenon of green algae; generate the area and concentration distribution of an area where green algae occur; and quickly evaluate the performance of green algae treatment technology according to the measured result. The present invention calculates the spatial occurrence area and concentration features of green algae in an area of a measurement object based on a green algae prediction model algorithm and a model thereof by the tide index calculated by a GPS-based artificial neural network algorithm displayed in geometrical coordinates by the random sampling on the autonomous movement path or the autonomous flight path. As a result, the present invention is able to: reduce the hassle of carrying out a site water quality survey per every flight operation as the calculated tide index value is a value calculated by a neural network algorithm considering a site weather condition and a water surface reflectivity; use the optimally calculated tide index as an absolute value; and can be used as big data according to a random sampling. Moreover, the present invention is able to improve the convenience and efficiency of an operation method when operating green algae removal equipment for a measurement target area by using a tide removing vessel or water circulating equipment.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a system and method for generating a green algae map using data measured by a GPS-based random sampling method,

[0001] The present invention relates to a green tract mapping system, and more particularly to a GPS-based autonomous navigation unmanned aerial vehicle for measuring the spatial distribution characteristics and temporal movement characteristics of green tides occurring in reservoirs or rivers by random sampling Using the Algae Factor (AF) deduced from the measured ultrasound sensor image and the concentration of chlorophyll-a or phycocyanin measured by random sampling in the GPS-based mobile unmanned body, The present invention relates to a system and a method for mapping a green tide map using data measured by a GPS-based random sampling method.

Recently, the physical environment of rivers and lakes due to climate change, for example, sudden change in water retention time, has led to serious social problems such as algae occurring in dams dedicated for water supply or reservoirs. In particular, the green algae phenomenon refers to the mass proliferation of algae among algae living in the water. If these algae are mass-grown in rivers or reservoirs, it may lead to problems such as fisheries and fisheries accidents due to depletion of oxygen in water bodies, inducing this hobby problem in the source water of the water purification plant, deteriorating the filtering function of the sand filter and toxicity to human body due to algae toxic substances .

The most widely used method for studying this greenhouse distribution is to utilize Chlorophyll-a or Phycocyanin concentrations. These chlorophyll-a concentrations or phycocyanin concentrations are used as indices for indirectly confirming the occurrence of green tides in rivers or reservoirs.

In Korea, the government operates a bird alert system for 28 rivers and lakes across the country. The purpose of introducing the bird alert system is based on Article 21 of the Law on the Protection of Water Quality and Aquatic Ecosystems. To provide stable supply of tap water. It is a system in which water is distributed according to the extent of the occurrence of the green tide. Despite the operation of the algae warning system, the problem of green algae is increasing.

In general, river water quality monitoring is performed at a representative point of the river. Therefore, when the measurement range is widened, or when there is a lot of junction points of a small river adjacent to a large river, The efficiency of monitoring the water quality of the river can be reduced, and it is difficult to grasp the spatial distribution of the whole stream. Therefore, research on water quality monitoring for understanding the spatial distribution of green tide is underway.

On the other hand, as a prior art, Korean Patent No. 10-1328026 discloses an invention entitled " a water environment monitoring system and a water environment monitoring method utilizing a depth-dependent profiling technique " do.

FIG. 1 is a schematic configuration diagram of a water environment monitoring system utilizing a depth-profiling profiling technique according to the prior art.

Referring to FIG. 1, a water environment monitoring system utilizing a profiling technique according to the related art is a system for monitoring water bodies such as rivers, lakes, reservoirs, and waters where continuous monitoring is required, (10) which is fixed to the bottom (2) of the water body or which is installed downward from the water body top to a water depth close to the bottom, including a water tank (21), a strainer (22) and a water collecting valve (23); A water collection device (20) provided in the support (10) for taking samples by water depth; A water environment measuring device 30 for receiving a water sample collected by the water sampling device 20 and measuring a water environment influence factor; A control unit for controlling the water taking apparatus 20 and the water environment measuring apparatus 30; And a profile analysis device 40 for analyzing the measurement results of the water environment measurement device 30 and analyzing the profiles by depth and time series. Reference numeral 31 denotes a water level meter, 32 denotes an anemometer, 50 denotes an automatic watering device, 60 denotes a profile storage device, and 70 denotes a data transmitting / receiving device.

According to the water environment monitoring system utilizing the profiling technique according to the conventional technology, by constructing a real-time monitoring system utilizing the water depth profile for water bodies requiring monitoring according to the water depth, it is possible to prevent water pollution, It is possible to establish the cause and cause ecology / water environment forecasting and alarm system of our company, but it requires huge installation and operation cost due to the use of the fixed type support, and when the flow rate is rapidly increased, There is a problem that it may be lost.

Recently, a remote sensing (RS) technique using satellite imagery has been under way to monitor the algae, and it has been widely applied to the monitoring of water quality in lakes, reservoirs, rivers and oceans have. In particular, various chlorophyll-a concentration estimation algorithms have been developed and applied to concentration calculations for chlorophyll-a concentration analysis using remote sensing data combined with field observations.

Recently, we have developed a model that predicts the distribution of the occurrence of green vegetation by using satellite images from remote sensing techniques. In case of using such satellite images, the greenhouse distribution prediction model may become inaccurate due to the fact that the field survey is not performed at the same time, or when there is no geographical coordinates or inconsistency at the time of water quality measurement at the time of field survey. In order to solve these problems, there has been proposed an alternative method of observing the distribution of green algae using UAV (Unmanned Aerial Vehicles). However, There is a problem.

In addition, even if a separate spectral sensor is mounted on the unmanned aerial vehicle, it is only necessary to parallel the on-site water quality analysis at the same point in time to estimate the green tract index to ensure the accuracy of the green tract map expressing the green tract distribution. For example, in the case of a river, a green algae generating area is large, or a green algae is distributed in order to remove a green algae at a large occurrence level, or when a mechanical device such as an aeration device or a water circulating device is operated, It is difficult to promptly cope with the problem when the area and the moving characteristics of the area are not taken into consideration, and there is a problem that the chemicals must be used excessively.

As described above, according to the conventional technique, in order to monitor the green tide occurring in a river or a reservoir, point-based green tide monitoring is performed using sampling and automatic analysis equipment at a point 500 m away from the river structure or at a specific point . This method has a problem that it is difficult to grasp the spatial distribution of a large-scale river or an entire reservoir.

According to the conventional technology, remote sensing technique using satellite images is applied and the monitoring is applied to the characteristics of the green tide distribution by extending the surface unit at the point unit level. In the case of the water quality factor measurement technique using the satellite image according to the conventional technique, the characteristics of the greenhouse distribution were analyzed by various chlorophyll-a concentration estimation algorithm using the remote sensing data combined with the field measurement.

However, there are limitations in the method of measuring water quality factor using satellite images according to the conventional techniques, such as lack of field measurement data measured at the same time with measurement images, precise preprocessing, correction and resolution. In order to compensate for the disadvantages, UAVs are equipped with spectroscopic sensors. However, there is a limitation in that it is necessary to carry out a survey on site water quality at the same time point in the same area as the satellite image.

Korean Patent No. 10-1328026 filed on Feb. 6, 2013, entitled " Water Environment Monitoring System and Water Environment Monitoring Method Utilizing Depth Profiling Technique " Korean Patent No. 10-1616727 filed on October 28, 2014, entitled " Safety System-Integrated Greenhouse and Red Tide Remote Monitoring System " Korean Patent No. 10-1446037 filed on May 7, 2013, entitled " Water Temperature, Green Tide, and Red Tide Occurrence Prediction System by Rather < RTI ID = Korean Patent No. 10-1338038 filed on Sep. 5, 2012, entitled " Apparatus for monitoring the occurrence of red tide and green tide using laser " Korean Patent No. 10-1075561 filed on June 10, 2011, entitled " UHM Multi-Water Quality Measurement Device Based on Wireless Internet " Korean Patent No. 10-906654 filed on July 4, 2008, entitled " Automatic Remote Water Quality Monitoring System and Turbidity Observation Method for Small Reservoir & Reclamation Fresh Water Lake & Korean Patent No. 10-522764 filed on November 20, 2002, entitled " Real-Time Water Quality Monitoring Apparatus and Control Method Thereof " Korean Patent No. 10-477554 filed on Mar. 11, 2003, entitled " Water Quality Management System and Method "

In order to solve the above-mentioned problems, the technical problem to be solved by the present invention is to provide a GPS-based optimal estimation model formula by random sampling on an autonomous flight path and an autonomous moving path, not a predetermined measuring point of an autonomous navigation unmanned aerial vehicle A system and method of green algae mapping using data measured by GPS-based random sampling method that can generate reservoir and river greenhouse map by calculating concentration and area of algae by inferring algae index using neural network algorithm .

It is another object of the present invention to provide a method and apparatus for analyzing a spatial distribution of a green tide and a temporal migration phenomenon, The present invention is to provide a system and a method for mapping a green tide utilizing data measured by a GPS-based random sampling method.

According to an aspect of the present invention, there is provided a green map mapping system that utilizes data measured by a GPS-based random sampling method according to the present invention. The system includes an automatic navigation device for autonomous navigation, An autonomous navigation unmanned aerial vehicle equipped with a GPS module and flying so as to acquire ultra-spectroscopic image data by a random sampling method in accordance with an autonomous flight path over an area to be measured; And the water quality measurement sensor acquires water quality measurement data for each measurement point by a random sampling method along an autonomous movement path on the surface of the measurement target region, and the water quality measurement data is cross flowed with the autonomous navigation unmanned aerial vehicle A movable type unmanned body traveling in the direction of the arrow; A data collecting and correcting unit for collecting data from the autonomous traveling unmanned aerial vehicle and the moving type unmanned vehicle and correcting the geometrical errors and positions so as to be represented by measurement point geographical coordinates; A data merge processing unit for merging the measurement data of the autonomous traveling unmanned aerial vehicle and the moving type unmanned vehicle based on the measurement point geographical coordinates and analyzing the correlation with the chlorophyll-a concentration or the phycocyanin concentration, ; An algae index calculating unit for calculating the algae index by calculating an optimum value of the estimated model equation, a water surface reflectivity, and a field weather condition, and calculating the algae index by an artificial neural network algorithm; A correlation analysis and map mapping unit for mapping the data processed by the data merging unit and the values calculated by the correlation analysis between the algae index calculated by the algae index calculating unit and the water quality data to a map; And a green map mapping unit for calculating a green map concentration map and an area of a green map of the measurement target area to generate a point-based green map, wherein the data merging process unit performs a random sampling method on the measurement path And the data measured by the autonomous traveling unmanned aerial vehicle and the moving type unmanned aerial vehicle are merged based on the measurement point geographical coordinates determined through the position correction and the correlation between the measurement points and the water quality data is analyzed, And an optimal estimation model equation is calculated.

Here, the GPS module mounted on the autonomous traveling unmanned aerial vehicle and the GPS module mounted on the portable unmanned vehicle each convert GPS coordinates to geographical coordinates so that the green tide map maker creates a point-based green map, respectively .

Here, the ultra-spectral sensor can measure the visible light region wavelength band of 400 nm to 700 nm and the near-infrared region wavelength region of 700 nm to 1000 nm in several hundreds of bands.

delete

Here, the data merge processing unit may include: an image matching processing unit that performs image matching processing according to a ground reference point for coordinate accuracy of the measurement target region with respect to the hyperspectral image collected from the data collection and correction unit; An orthographic projection image generation unit for generating an orthographic projection image from a plurality of hyperspectral images so as to extract accurate feature points between individual images according to an object recognition algorithm; And an optimal estimation model calculation unit for calculating an optimal estimation model equation using the ultrasound image data and the water quality data of the measurement target area.

Here, the correlation analysis and map mapping unit analyzes the result of the regression analysis between the algae index and the field chlorophyll-a concentration or the phycocyanin concentration using the GIS region classification scheme together with the positionally corrected geographical coordinate information The greenhouse mapping unit is configured to determine a greenhouse concentration distribution diagram of the measurement target area and to select the greenhouse rainfall area, wherein the greenhouse rainfall area has a chlorophyll-a concentration of 50 / / L or more To calculate the area for the green-marine watershed area.

According to another aspect of the present invention, there is provided a greenhouse mapping method using data measured by a GPS-based random sampling method according to the present invention, comprising the steps of: a) A step of acquiring hyperspectral image data by measuring a measurement target area by a random sampling method along an autonomous flight path of an unmanned aerial vehicle; b) a method in which a GPS module and a water quality measurement sensor measure a chlorophyll- Measuring a phycocyanin concentration; c) examining the water surface reflectivity and the field weather condition of the measurement target area; d) collecting data from the autonomous navigation unmanned aerial vehicle and the moving type unmanned vehicle and correcting the geometric errors and positions so that they can be represented by measurement point geographical coordinates; e) merging the measurement data of the autonomous traveling unmanned aerial vehicle and the moving type unmanned vehicle according to the measuring point geographical coordinates; f) calculating an optimal estimation model equation according to the data measured from the autonomous traveling unmanned aerial vehicle and the moving type unmanned vehicle; g) calculating an algebraic index by calculating an optimal estimated model equation, a reflectivity value of a water surface, and a field weather condition, and calculating the algebraic index using an artificial neural network algorithm; h) analyzing the correlation between the optimal estimation model equation and the chlorophyll-a concentration or phycocyanin concentration measured from the mobile unattended subject; i) mapping the topographic map according to the analyzed correlations; And j) creating a point-based green map for the area to be measured, wherein step e) comprises collecting a plurality of measurements in a random sampling manner on a measurement path other than a predetermined measurement point, The data measured by the autonomous navigation unmanned aerial vehicle and the moving type unmanned vehicle are merged based on the measurement point geographical coordinates determined through the position correction and the optimum estimation model expression for each measurement point is calculated through correlation analysis with the water quality data; In the step (j), the greenhouse concentration distribution map and the greenhouse well zone area of the measurement target area are calculated.

According to the present invention, by using GPS-based optimal estimation model and artificial neural network algorithm by random sampling on autonomous flight paths and autonomous travel paths, rather than predetermined measurement points of autonomous traveling unmanned aerial vehicles and mobile subjects, It is possible to easily calculate the greenhouse distribution density and area and to analyze the greeneye distribution temporally and spatially.

According to the present invention, it is possible to grasp the spatial distribution of the green tide and the temporal migration phenomenon, to make an area and a concentration distribution diagram of the green tide occurrence center area, and to quickly evaluate the performance of the green tide processing technique according to the measured result.

According to the present invention, a green algae prediction model algorithm based on the algae index calculated by a GPS-based artificial neural network algorithm represented by geographical coordinates by random sampling on an autonomous flight path or autonomous movement path, The calculated algae index values are calculated by neural network algorithm considering water surface reflectivity and field weather conditions. Therefore, it is troublesome to perform field water quality survey in every flight operation. And the optimum calculated algae index can be used as an absolute value, and it can be utilized as big data according to random sampling.

According to the present invention, a green algae prediction model algorithm based on the algae index calculated by a GPS-based artificial neural network algorithm represented by geographical coordinates by random sampling on an autonomous flight path or autonomous movement path, By calculating the spatial occurrence area and concentration characteristics of the green tide in the area, it is possible to improve the convenience and efficiency of the operation method when operating the green tide removal device for the measurement target area by utilizing the algae elimination line or the water circulation device.

FIG. 1 is a schematic configuration diagram of a water environment monitoring system utilizing a depth-profiling profiling technique according to the prior art.
FIG. 2 is a schematic view for explaining the operation of an autonomous traveling unmanned aerial vehicle and a moving type unmanned vehicle for calculating an aviation index measured by a GPS-based random sampling method according to an embodiment of the present invention.
3 is a block diagram of a green tide mapping system that utilizes data measured by a GPS-based random sampling method according to an embodiment of the present invention.
FIG. 4A is a diagram illustrating an autonomous navigation unmanned aerial vehicle in a greenhouse mapping system utilizing data measured by a GPS-based random sampling method according to an embodiment of the present invention, and FIG. 4B is a block diagram of an autonomous navigation unmanned aerial vehicle .
5 is a diagram illustrating a moving type unattended subject in a green tide mapping system utilizing data measured by a GPS-based random sampling method according to an embodiment of the present invention.
FIG. 6 is a diagram illustrating a flight path of a self-driving unmanned aerial vehicle and a moving path of a moving unmanned vehicle in a green tide mapping system utilizing data measured by a GPS-based random sampling method according to an embodiment of the present invention.
7 is a block diagram of a data merge processing unit in a green tide mapping system utilizing data measured by a GPS-based random sampling method according to an embodiment of the present invention.
FIG. 8 is a flowchart illustrating a method of estimating an algae index based on geographical coordinates and data accumulated by random sampling in a greenhouse mapping system utilizing data measured by a GPS-based random sampling method according to an embodiment of the present invention. FIG.
9 is a specific configuration diagram of an artificial neural network model for calculating an algae index in a greenhouse mapping system utilizing data measured by a GPS-based random sampling method according to an embodiment of the present invention.
10 is a flowchart illustrating a greenhouse mapping method using data measured by a GPS-based random sampling method according to an embodiment of the present invention.
11 is a view illustrating a green tide map created by the tide index according to the embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings, which will be readily apparent to those skilled in the art. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. In order to clearly illustrate the present invention, parts not related to the description are omitted, and similar parts are denoted by like reference characters throughout the specification.

Throughout the specification, when an element is referred to as " comprising ", it means that it can include other elements as well, without excluding other elements unless specifically stated otherwise. Also, the term " part " or the like, as described in the specification, means a unit for processing at least one function or operation, and may be implemented by hardware, software, or a combination of hardware and software.

First, the green map that represents the green tide distribution of the river or the reservoir is used for the two-dimensional surface or the three-dimensional stereo analysis, the equality of the measurement points, the acquisition of a plurality of measuring points to be measured, the data consistency for the algae index inference, And the speed and convenience of mapping and mapping of the green tract should be ensured.

[Green Tide Mapping System Utilizing Data Measured by GPS-Based Random Sampling Method (100)]

FIG. 2 is a schematic view for explaining the operation of an autonomous traveling unmanned aerial vehicle and a moving type unmanned vehicle for calculating an aviation index measured by a GPS-based random sampling method according to an embodiment of the present invention.

2, the greenhouse mapping system 100 using the data measured by the GPS-based random sampling method according to the embodiment of the present invention includes an ultra-spectral sensor 120 And the GPS module 111 are mounted on the autonomous flight path 230 and the ultra-spectral sensor 120 photographs the measurement target region 210 using the GPS-based random sampling method and acquires an ultra-spectral image And the water quality measurement sensor 140 moves along the autonomous movement path 240 by mounting the water quality measurement sensor 140 and the GPS module 132 on the portable type unmanned body 130 so that the water quality measurement sensor 140 performs GPS- Water quality such as a chlorophyll-a concentration or a phycocyanin concentration can be measured in a water body such as a river, a lake, a reservoir, a sea area, or the like, that is, Here, reference numeral 220 denotes a ground reference point.

The green tide mapping system 100 that utilizes the data measured by the GPS-based random sampling method according to the embodiment of the present invention is configured to generate a green tide map by using a random sampling method on an autonomous flight path 230, And a green algae prediction model based on the algae factor calculated by the neural network algorithm that takes into consideration the optimum estimated model formula value, the water surface reflectance, and the field weather conditions, The spatial generation area and density characteristic of the green tide of the measurement target area 210 can be calculated based on the measured data.

Meanwhile, FIG. 3 is a block diagram of a green tide mapping system that utilizes data measured by a GPS-based random sampling method according to an embodiment of the present invention.

Referring to FIG. 3, a greenhouse mapping system 100 using data measured by a GPS-based random sampling method according to an embodiment of the present invention includes an autonomous navigation unmanned air vehicle 110, an ultra-spectral sensor 120, A moving image unmanned body 130, a water quality measuring sensor 140, a data collecting and correcting unit 150, a data merging processing unit 160, an algae index calculating unit 170, a correlation analysis and map mapping unit 180, And a creating unit 190.

The autonomous traveling unmanned flying vehicle 110 is equipped with an autonomous navigation device so as to enable autonomous traveling and includes an ultra-spectral sensor 120 and a GPS module 111 mounted thereon, And the ultra-spectroscopic sensor 120 scans the ultrasound image data in a random sampling manner. That is, the autonomous traveling unmanned aerial vehicle 110 records the photographed ultrasound image in a data logger so as to calculate an optimal estimated model expression from the ultrasound image photographed by the ultrasound spectral sensor 120.

The supersonic spectroscopic sensor 120 is installed at the bottom of the autonomous traveling unmanned aerial vehicle 110 to take an ultra-spectroscopic image of the measurement target region 210 by a random sampling method. Here, the ultrasound spectral sensor 120 can measure the visible light region wavelength band of 400 nm to 700 nm and the near-infrared region wavelength region of 700 nm to 1000 nm in several hundreds of bands.

The moving type unmanned body 130 is remotely adjusted and the GPS module 132 and the water quality measurement sensor 140 are installed on the surface of the measurement area 210 along the autonomous movement path 230, And travels in the cross flow direction with the autonomous traveling unmanned air vehicle 110 of the measurement target area 210 by a random sampling method in order to acquire the water quality measurement data for each measurement point by the sampling method. Therefore, since the autonomous traveling unmanned aerial vehicle 110 and the mobile unmanned vehicle 130 have GPS modules 111 and 132 embedded therein, the measurement points of the measurement target area 210, which is determined by the random sampling method, And stored in the logger as geographical coordinates.

The water quality measurement sensor 140 is fixed to the rear end of the movable type unmanned body 130 and measures the water quality change of the measurement target area 210 by a random sampling method. For example, the water quality measurement sensor 140 measures the chlorophyll-a concentration or the pycocyanin concentration on the autonomous movement path of the measurement target region 210.

The data collecting and correcting unit 150 collects data from the autonomous traveling unmanned aerial vehicle 110 and the moving unmanned vehicle 130 and corrects the geometric errors and positions so that the data can be represented by the measurement point geographical coordinates. The data collecting and correcting unit 150 performs geometric correction of the image data and position coordinate correction without physical connection points with the ground in consideration of the mechanical errors of the autonomous traveling unmannurized vehicle 110 and the moving type unmanned body 130 And the correlation between the ultrasound image and the geographical coordinates is analyzed through a ground control point (GCP), and the collected data is collected from the autonomous traveling unmanned vehicle 110 and the mobile unmanned vehicle 130 It is possible to correct the error of the data and measurement point geographical coordinates, thereby eliminating the geometrical error.

The data merge processing unit 160 merges measurement data of the autonomous traveling unmanned aerial vehicle 110 and the moving type unmanned vehicle 130 based on the measurement point geographical coordinates and calculates an optimal estimation model formula. Here, the data merge processing unit 160 collects a plurality of measurement values using a random sampling method on a measurement path that is not a predetermined measurement point, and measures the measurement values in the autonomous traveling unmanned air vehicle 110 and the mobile unmanned body 130 Based on the measurement point geographical coordinates determined through the position correction, and calculates an optimal estimation model expression through correlation analysis with the data measured by the mobile type unattended body 130. [

The algae index calculating unit 170 calculates the Algae Factor by calculating the value calculated from the optimal estimation model formula, the water surface reflectivity, and the field weather condition, and calculates the algae factor using an artificial neural network algorithm. Here, the value calculated according to the optimal estimation model equation is stored for each measurement point in the measurement target area 210, and the value calculated from the optimal estimation model equation is calculated based on the water surface reflectance, the temperature, the humidity, We can calculate the algae index of a water body by implementing an artificial neural network algorithm.

The correlation analysis and map mapping unit 180 performs correlation analysis between the chlorophyll-a concentration or the phycocyanin concentration data processed in the data merge processing unit 160 and the algae index calculated by the algae index calculating unit 170, Map to the map.

The green tide mapping unit 190 calculates the green tide concentration map and the green tide area of the measurement target area 210 to create a point-based green map. At this time, the GPS module 111 mounted on the autonomous traveling unmanned aerial vehicle 110 and the GPS module 132 mounted on the movable unmanned vehicle body 130 are respectively connected to the point-based green tide map GPS coordinates can be converted and displayed as geographical coordinates.

According to the embodiment of the present invention, the supersonic spectral sensor 120 is mounted on the drone, which is the autonomous traveling unmanned aerial vehicle 110, and the water quality sensor 140 is mounted on the movable type unmanned body 130, The water quality of the water bodies such as the sea area can be measured. Specifically, the value calculated by the optimal estimation model equation stored for each measurement point, the water surface reflectance, and the algae index calculated by the artificial neural network algorithm for the field weather conditions are determined based on the chlorophyll-a concentration measured in the water quality measurement sensor 140 The correlation between the concentration of phycocyanin and the concentration of the measurement target area 210 is automatically calculated based on the calculated result and is automatically mapped onto the topographical map together with the geographical coordinate information. And the area of the green watershed area can be calculated based on the chlorophyll-a concentration of 50 쨉 g / L or more as the reference.

Meanwhile, FIG. 4A is a diagram illustrating an autonomous navigation unmanned aerial vehicle in a greenhouse mapping system utilizing data measured by a GPS-based random sampling method according to an embodiment of the present invention, FIG. 4B is a block diagram of an autonomous navigation unmanned aerial vehicle .

Referring to FIGS. 4A and 4B, in the green tide mapping system 100 using data measured by the GPS-based random sampling method according to the embodiment of the present invention, the autonomous traveling unmanned air vehicle 110 includes a GPS module 111 A control unit (MCU) 112, a data logger 113, a flight unit 114, and a battery 115.

As shown in FIG. 4A, the GPS module 111 is mounted on the body of the UAV 110 to determine the position of the UAV 110, and the position thus determined is converted into geographical coordinates Can be displayed. That is, the GPS coordinates received from the satellites are converted into geographical coordinates, and a green map can be created according to the geographical coordinates of the measurement points.

The control unit (MCU) 112 controls the flight unit 114 by the autonomous travel program and controls the driving of the ultrasonic sensor 120. At this time, the ultrasound sensor 120 takes an ultra-spectroscopic image by a random sampling method.

The data logger 113 stores the ultra-spectral image data photographed by the ultra-spectral sensor 120 in a random sampling manner.

The flight unit 114 is provided with an automatic navigation device for allowing the autonomous traveling unmanned aerial vehicle 110 to autonomously travel along an autonomous flight path and an altitude and is driven under the control of the control unit 112, ) Along the autonomous flight path.

The battery 115 supplies power to the GPS module 111, the control unit 112, the data logger 113, the flight unit 114, and the ultra-spectral sensor 120).

Meanwhile, FIG. 5 is a diagram illustrating a moving type unattended body in a green tide mapping system utilizing data measured by a GPS-based random sampling method according to an embodiment of the present invention.

Referring to FIG. 5, in the green tide mapping system 100 using data measured by the GPS-based random sampling method according to the embodiment of the present invention, the mobile unmanned body 130 includes an unmanned body 131, Module 132 and a data logger 133. [

The unmanned body 131 may be a floating structure, for example, a ship, which is operated along the surface on the measurement area 210, and is preferably remotely controlled.

The GPS module 132 is mounted on the unmanned main body 131 so as to enable autonomous travel and remote operation, and grasps the position of the mobile unmanned body 130. The position thus detected is converted into geographical coordinates .

The data logger 133 stores the concentrations of chlorophyll-a or phycocyanin measured by the water quality measurement sensor 140 as water quality measurement data.

FIG. 6 is a diagram illustrating a flight path of a self-driving unmanned aerial vehicle and a moving path of a moving unmanned vehicle in a green tide mapping system utilizing data measured by a GPS-based random sampling method according to an embodiment of the present invention.

6, in the green tide mapping system 100 using the data measured by the GPS-based random sampling method according to the embodiment of the present invention, the autonomous traveling unmanned vehicle 110 includes a GPS-based random sampling method The supersonic spectroscopic sensor 120 photographs the hyperspectral image of the measurement target region 210 along the autonomous flight path and the mobile type unmanned body 130 acquires the water quality along the autonomous movement path in a GPS- The measurement sensor 140 measures the chlorophyll-a concentration or the phycocyanin concentration. For example, when the measurement target area 210 is a river, the mobile unmanned body 130 moves along the both sides of the river from the upstream to the downstream of the river, and the concentration of chlorophyll- And the autonomous traveling unmanned aerial vehicle 110 photographs an ultrasound image by a GPS-based random sampling method while traveling in the longitudinal direction of the river.

Meanwhile, FIG. 7 is a block diagram of a data merge processing unit in a green tide mapping system utilizing data measured by a GPS-based random sampling method according to an embodiment of the present invention.

7, the data merge processing unit 160 in the green tide mapping system 100 using data measured by the GPS-based random sampling method according to the embodiment of the present invention includes an image matching processing unit 161, A projection image creating unit 162 and an optimal estimation model expression calculating unit 163.

The image matching processing unit 161 may be configured to perform a process of superimposing the ultrasound image data obtained from the ultrasound spectral sensor 120 on the ground for the coordinate accuracy of the measurement target area 210, And performs image matching processing according to the reference point.

The ortho-projection-image creating unit 162 creates an ortho-projected image (Ortho Image) from a plurality of hyperspectral images so as to extract accurate feature points between individual images, for example, according to an SITF object recognition algorithm.

The optimal estimation model formula calculator 163 calculates an optimal estimation model equation by analyzing the correlation between the hyperspectral image data and the water quality measurement data of the measurement area 210.

Meanwhile, FIG. 8 is a graph showing the relationship between the geographical coordinates accumulated by random sampling in the green tide mapping system utilizing the data measured by the GPS-based random sampling method according to the embodiment of the present invention, FIG. 9 is a specific configuration diagram of the artificial neural network model for calculating the bird's water index. FIG.

Referring to FIG. 8, in a green tide mapping system utilizing data measured by a GPS-based random sampling method according to an embodiment of the present invention, algae indexes are used to infer the algae index through geographical coordinates and data accumulated by random sampling The exponent calculation unit 170 includes an artificial neural network model 171 and a feedback unit 172. The artificial neural network model 171 includes a data input unit 171-1, an artificial neural network algorithm operation unit 171-2, And an algebraic index output unit 171-3. The feedback unit 172 includes an algebraic index error calculation unit 172-1 and a feedback control unit 172-2. The data merge processing unit 160 , The water surface reflectivity, and the field weather condition are input to the data input unit 171-1.

As described above, the green tide mapping system 100 utilizing the data measured by the GPS-based random sampling method in the embodiment of the present invention includes the autonomous traveling unmanned aerial vehicle 110 having the GPS modules 111 and 132, And the movable type unmanned body 130 are utilized. That is, the measurement point geographical coordinates and the deduced algae factor (AF) are constructed in a database using a random sampling method for the measurement target area using the autonomous traveling unmanned aerial vehicle 110 and the mobile unmanned vehicle 130.

In order to calculate the algae index, the supersonic spectral sensor 120 mounted on the autonomous traveling unmanned aerial vehicle 110 has the following measurement wavelength band. The ultra-spectral sensor 120 can measure the visible light region wavelength band of 400 nm to 700 nm and the near-infrared region wavelength region of 700 nm to 1000 nm in several hundreds of bands,

In addition, the concentration of chlorophyll-a or phycocyanin generated in the water body is a key molecule of photosynthesis and is a dye that absorbs light energy. Since it has a characteristic of absorbing light at a specific wavelength to a maximum, (120), which will be described in more detail as follows.

Specifically, the optimal estimation model equation is defined by the following equations (1) to (6).

For calculating the algae index, an artificial neural network model algorithm such as the following Equations (7) and (8) is calculated in consideration of the field conditions such as the value calculated in the optimal estimation model equation, the reflectivity of water surface, temperature, humidity, The algae factor (AF) of a waterbody can be calculated, and the distribution of green tide of the water body and the area of the green water well can be selected.

는 입력(x₁, x₂, …, x_p)에 대한 가중합을 나타내고,

는 바이어스를 의미하며,

는 출력결과를 나타낸다.

는 i번째 입력과 j번째 뉴런사이의 연결강도를 나타내고,

는 j번째 뉴런의 활성화 함수이다. 특히,

는 비선형 함수로서, 여기서는 시그모이드 함수를 적용하기로 한다. 이것은 비선형 함수인 시그모이드 함수를 적용하여, 결정 영역이 통상의 직선이 아니라 완만한 곡선으로 경계를 형성시킴으로써 행위의 분석이 복잡하지만 미분을 통해 은닉층을 학습할 수 있다. 이때, 도 8에 도시된 바와 같이, 인공신경망 모델(171)은 예측값 및 실제값을 비교하여 오차가 작은 방향으로 노드간의 연결강도(

)를 다음의 수학식 9를 이용하여 조절하고, 출력값을 출력한다.here,

Represents a weighted sum for the input (x ₁ , x ₂ , ..., x _p )

Quot; means a bias,

Represents the output result.

Represents the connection strength between the i-th input and the j-th neuron,

Is the activation function of the jth neuron. Especially,

Is a nonlinear function, and a sigmoid function is applied here. This is a nonlinear function applying a sigmoid function, and the decision domain is not a normal straight line but a gentle curved line to form a boundary, so that the analysis of the behavior is complicated, but the hidden layer can be learned through the derivative. 8, the artificial neural network model 171 compares the predicted value and the actual value, and determines the connection strength between nodes in a direction with a small error

) Using the following equation (9), and outputs an output value.

는 p번째 목표패턴의 j성분이고,

는 p번째 입력패턴에서 신경망 모델이 계산한 출력패턴의 j성분이며,

는 p번째 입력패턴의 i성분이고,

는

로서, 목표패턴과 실제패턴의 차(오차)를 나타내며, 또한,

는 점성계수를 나타낸다.here,

Is the j component of the p-th target pattern,

Is the j component of the output pattern calculated by the neural network model in the p-th input pattern,

Is the i-th component of the p-th input pattern,

The

(Error) between the target pattern and the actual pattern,

Represents the viscosity coefficient.

9, an artificial neural network model 171 applied to an embodiment of the present invention uses a multi-layer perceptron including an algorithm for training data in a multi-layer structure having a hidden layer between an input layer and an output layer It is done through mechanical learning. Here, the parameters of the input layer are obtained by learning the relationship between the optimum estimated model formula, water surface reflectance, temperature, humidity, and the amount of sunshine, and output the standardized algae index to the output layer.

In this artificial neural network model 171, at least one input node can be adjusted, and the hidden layer has a hidden node for computing an artificial neural network model according to the number of input nodes. The output layer includes at least one output Results. In this artificial neural network model 171 implementation, the number of hidden layers is preset. In the input layer, predetermined input conditions are input from the data input section 171-1, and the output layer outputs the classified results to the tidal-coefficient output section 171-3.

When the input condition is input to the input layer, the artificial neural network model 171 becomes the input value of the node at the next level in the operation result performed by the node in the hidden layer. . In this case, it is the connection strength to connect the nodes in each layer.

These connections can not connect nodes in the same layer, but can connect nodes in different layers, and typically one node is connected to all nodes in the next layer. In the artificial neural network model 171, each node can be modeled as an artificial neuron. Each neuron combines the input external stimulus and responds to the result. This can be expressed by Equations (7) and (8) described above. The weighted sum of the inputs together with the bias (Bias) is transferred to the activation function (or transfer function) to output the result.

An algae index model is generated through an artificial neural network operation processing unit 171-2 deduced by the artificial neural network model 171, and an ortho-projection image is generated for each measurement point. A plurality of point-based high-resolution greenery maps can be automatically created by automatically mapping derived algae index models and geo-coordinate information corrected by the above-described process to the topographical map.

In other words, in the artificial neural network algorithm processing unit 171-2 of the artificial neural network model 171 in which the artificial neural network algorithm is implemented, the optimal estimation model equation stored for each measurement point is correlated with the chlorophyll-a or the phycocyanin concentration And the neural network algorithm is operated on the basis of the calculated optimal model formula value, the reflectivity of the water surface, and the field weather condition data, and the algae index output unit 171-3 calculates the algae index AF) can be calculated.

Thereafter, the correlation analysis and map mapping unit 180 analyzes the result of the regression analysis between the algae index and the field chlorophyll-a concentration or the phycocyanin concentration, together with the position-corrected geographical coordinate information, for example, GIS The green map mapping unit 190 is configured to identify the green map concentration distribution of the measurement target area 210 and to select the green mapping area using the map, The area of the green watershed area can be calculated based on the chlorophyll-a concentration of 50 μg / L or more.

In the case of the green tract mapping system 100 that utilizes the data measured by the GPS-based random sampling method according to the embodiment of the present invention, in order to improve the reliability of the conventional method of measuring the area- The accuracy of the green algae distribution inferencing model can be improved by utilizing random sampling data capable of sampling a plurality of unspecific ones. Also, in order to calculate the greenhouse concentration of the measurement target area by using a plurality of measurement values of the previous step, the ultrasound sensor 120 is used to measure the concentration of chlorophyll-a or phycocyanin- And the constructed measurement point geographic coordinates and algae indexes are automatically mapped to the topographic map to automatically generate a number of point-based, high resolution green map maps that are not image based.

(Eg, chlorophyll-a concentration or phycocyanin concentration) and algae factor (AF), which are automatically measured for algae factor (AF) and in-situ random sampling points, Can be inferred as shown in the following Equation (10) and Equation (11).

Here, C represents chlorophyll-a concentration (占 퐂 / L) or phycocyanin concentration (占 퐂 / L), and a and b represent correlation coefficients.

At this time, in order to improve the reliability of the model, the measurement point geographical coordinates and the water quality data, which are automatically measured by the random sampling method, are taken into consideration in consideration of the mechanical errors of the autonomous traveling unmanned aerial vehicle 110 and the moving unmanned vehicle 130 Geometric correction and position correction of the image data are performed without physical connection points with the ground. Correlation between the image and the geographical coordinate is formed through the ground control point (GCP) By correcting the terrain by correcting the terrain by removing the error, the value calculated by the optimal estimation model formula for each coordinate point based on the above ground reference point (GCP), the water surface reflectance, and the data value considering the field weather conditions are deduced from the artificial neural network algorithm The algae index (AF) and water quality data can be matched at the same point and displayed as a topographic map.

Through this, it is possible to quickly grasp the greenhouse watershed area when the greenhouse removal apparatus is operated, and the operating range of the greenhouse removal apparatus can be determined based on this.

Also, since the algae index (AF) varies depending on the measurement condition depending on the reflectance of the surface of the water body according to the weather conditions, the algae index (AF) It is possible to replace the method of operating the autonomous traveling unmanned aerial vehicle 110 and the moving unmanned vehicle 130 to acquire data and perform arithmetic processing. Therefore, when the big data measured by repeating the above-described method for a long time is collected, the greenhouse distribution map, the right heart area, and the area can be calculated using only the autonomous traveling unmanned air vehicle 110 without the operation of the mobile type unmanned body 130.

In addition, the algae prediction algorithm based on the algae index calculated by GPS-based artificial neural network algorithm, which is represented by geographical coordinates by random sampling on autonomous flight paths or autonomous motion paths, The calculated algae index value is calculated by the neural network algorithm that considers the water surface reflectance and the field weather conditions, and it is less troublesome to carry out the field water quality survey every time the air vehicle is operated , The optimal calculated algae index can be used as an absolute value, and it can be utilized as big data according to random sampling.

[How to map green tide using data measured by GPS-based random sampling method]

FIG. 10 is a flowchart illustrating an operation of a greenhouse mapping method using data measured by a GPS-based random sampling method according to an embodiment of the present invention, FIG. 11 is a flowchart illustrating a greenery map generated by an algae index according to an embodiment of the present invention, Fig.

Referring to FIG. 10, a greenhouse mapping method using data measured by a GPS-based random sampling method according to an embodiment of the present invention includes a GPS module 111 and an autonomous The traveling UAV 110 measures the measurement target area 210 in a random sampling manner along the autonomous flight path 230 to acquire superspectral image data at step S110. That is, the supersensor sensor 120 is installed at the lower end of the autonomous traveling unmanned aerial vehicle 110 to photograph a superspectral image of the measurement target region 210 in a random sampling manner, The visible light region wavelength band of 700 nm to 700 nm and the near infrared region wavelength band of 700 nm to 1000 nm can be divided into hundreds of bands.

Next, the portable unmanned body 130 mounted with the GPS module 132, the water quality sensor 140 and the radiometer measures the chlorophyll-a concentration or the phycocyanin concentration by a random sampling method along the autonomous movement path 240 (S120). At this time, the water quality measurement sensor 140 is fixed to the rear end of the movable type unmanned body 130 and measures the chlorophyll-a concentration or the phycocyanin concentration of the measurement target region 210 by a random sampling method. Here, the GPS module 111 mounted on the autonomous traveling unmanned aerial vehicle 110 and the GPS module 132 mounted on the movable unmanned vehicle 130 are configured to generate point-based greenhouse maps, respectively, Conversion display.

Next, the water surface reflectivity of the measurement target area 210 and the field weather conditions are examined (S130).

Next, the data is collected from the autonomous traveling unmanned aerial vehicle 110 and the moving type unmanned vehicle 130, and the geometric errors and positions are corrected so as to be represented by measurement point geographical coordinates (S140). That is, considering the mechanical errors of the autonomous traveling unmanned aerial vehicle 110 and the moving type unmanned body 130, geometric correction and positional coordinate correction of the image data are performed without a physical connection point with the ground, and a ground reference point (GCP) By analyzing the correlation between the spectroscopic image and the geographical coordinates, it is possible to correct the errors of the data and measurement point geographical coordinates collected from the autonomous traveling unmanned aerial vehicle 110 and the moving type unmanned vehicle 130.

Next, the measurement data of the autonomous traveling unmanned aerial vehicle 110 and the moving unmanned vehicle 130 are merged according to the measurement point geographical coordinates (S150). That is, a plurality of measured values are collected by a random sampling method on a measurement path other than a predetermined measurement point, and data measured by the autonomous traveling unmanned aerial vehicle 110 and the movable unmanned vehicle 130 are determined through position correction Based on the geographical coordinates of the measurement point, the optimal estimation model equation for each measurement point is calculated by analyzing the correlation with the water quality data.

Next, an optimum estimation model equation is calculated according to the measured data from the autonomous navigation unmanned air vehicle 110 (S160).

Next, the calculated value of the optimal estimation model equation, the reflectivity value of the water surface, and the field weather conditions are input and an algebraic factor is calculated by an artificial neural network algorithm (S170).

Next, the correlation between the algae index and the concentration of chlorophyll-a or phycocyanin is analyzed (S180).

Next, the topographic map is mapped according to the analyzed correlation (S190).

Next, a point-based greenery map as shown in FIG. 11 is created for the measurement target area 210 (S200). At this time, the greenhouse concentration distribution diagram and the greenhouse rainfall area of the measurement target area 210 can be calculated. In other words, the value calculated from the stored optimal estimation model equation for each measurement point is regression analyzed with chlorophyll-a concentration or phycocyanin concentration, and the calculated result is automatically mapped on the topographical map together with the positionally corrected geographical coordinate information The greenhouse rainforest area is selected by determining the greenhouse concentration distribution diagram of the measurement target area 210, and the greenhouse rainforest area is selected by setting the chlorophyll-a concentration to be greater than or equal to 50 占 퐂 / Can be calculated.

As a result, according to the embodiment of the present invention, the GPS-based optimal estimation model equation and water surface reflectance by random sampling on the autonomous flight path and the autonomous movement path, not the predetermined measurement points of the autonomous navigation unmanned aerial vehicle and the mobile subject, By using the artificial neural network algorithm, the algae index can be deduced to easily calculate the greenhouse distribution concentration and area, and the greenhouse distribution can be analyzed temporally and spatially. In addition, it is possible to grasp the spatial distribution of the green tide and the temporal migration phenomenon, to create the area and concentration distribution map of the green zone, and to quickly evaluate the performance of the green tide processing technology according to the measured results.

According to an embodiment of the present invention, a green algae prediction model algorithm based on a GPS-based artificial neural network algorithm represented by geographical coordinates by random sampling on an autonomous flight path or an autonomous movement path, The calculated algae index values are calculated by neural network algorithm considering water surface reflectivity and field weather conditions. Based on the calculated area and concentration characteristics of green tide, It can reduce the cumbersome hassle and can use the best calculated algae index as an absolute value and can be utilized as big data according to random sampling.

According to an embodiment of the present invention, a green algae prediction model algorithm using an algae index using a GPS-based optimal estimation model expression converted to geographical coordinates by random sampling on an autonomous flight path or an autonomous movement route, , It is possible to improve the convenience and efficiency of the operation method when operating the green alga cancellation device for the measurement target area by utilizing the algae elimination line or the water circulation device by calculating the spatial occurrence area and concentration characteristic of the green algae have.

It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims and their equivalents. There will be. It is therefore to be understood that the above-described embodiments are illustrative in all aspects and not restrictive. For example, each component described as a single entity may be distributed and implemented, and components described as being distributed may also be implemented in a combined form.

The scope of the present invention is defined by the appended claims rather than the detailed description and all changes or modifications derived from the meaning and scope of the claims and their equivalents are to be construed as being included within the scope of the present invention do.

100: Green Tide Mapping System
110: Self-driving unmanned vehicle
120: second spectroscopic sensor
130: movable unmanned body
140: Water quality sensor (chlorophyll-a sensor or phycocyanin sensor)
150: Data collection and correction unit
160: Data merging processor
170: Bird Index Calculation Unit
180: correlation analysis and map mapping unit
190: green tide map maker
110a: body of the unmanned aerial vehicle
111: GPS module
112: Control unit (MCU)
113: Data logger
114: Flying unit
115: Battery
131: Unmanned body
132: GPS module
133: Data logger
161: image matching processing unit
162: ortho projection image creating unit
163: Optimal Estimation Model Expression Unit
171: Artificial Neural Network (ANN) Model
172:
171-1: Data input unit
171-2: Artificial neural network algorithm processing unit
171-3: Algae index output unit
172-1: Bird Index Error Calculation Section
172-2:
210: Area to be measured (river or reservoir)
220: ground reference point
230: autonomous flight path
240: autonomous movement path

Claims (10)

The ultrasonic spectroscopic sensor 120 and the GPS module 111 are mounted so that the autonomous navigation can be carried out and the supersonic spectral sensor 120 is installed on the measurement area 210 along the autonomous flight path 230, An autonomous navigation unmanned vehicle (110) flying to acquire ultrasonic image data by a random sampling method;
And the water quality measurement sensor 140 is controlled by a random sampling method along the autonomous movement path 230 on the surface of the measurement target area 210 by using the water quality measurement sensor 140, the GPS module 132, A portable unmanned body (130) for acquiring water quality measurement data and traveling in a cross flow direction with the autonomous traveling unmanned air vehicle (110);
A data collecting and correcting unit 150 for collecting data from the autonomous traveling unmanned aerial vehicle 110 and the moving type unmanned vehicle 130 and correcting the geometrical errors and position so that the data can be represented by measurement point geographical coordinates;
The measurement data of the autonomous traveling unmanned aerial vehicle 110 and the moving type unmanned vehicle 130 are merged based on the measurement point geographical coordinates and the correlation between the chlorophyll a concentration or the phycocyanin concentration is analyzed, A data merge processing unit 160 for calculating the data merge function;
An algae index calculating unit 170 for calculating an algebraic factor by calculating an optimum estimated model formula value, a water surface reflectance, and a field weather condition, and calculating the algae factor by an artificial neural network algorithm;
A correlation analysis and map mapping unit 180 for mapping the data processed by the data merge processing unit 160 and the values calculated by the correlation analysis between the tide index calculated by the tide index calculating unit 170 and the water quality data to a map; And
A green tide mapping unit 190 for generating a point-based green tide map by calculating a green tide concentration distribution chart and an area of a green tide area in the measurement target area 210,
The data merge processing unit 160 collects a plurality of measurement values using a random sampling method on a measurement path that is not a predetermined measurement point and transmits the measured data to the mobile unmanned body 110 and the mobile non- Based on the geographical coordinates of the measurement point determined through the position correction, and calculating an optimal estimation model expression for each measurement point by analyzing the correlation with the water quality data, thereby utilizing the data measured by the GPS-based random sampling method A green algae mapping system. The method according to claim 1,
The GPS module 111 mounted on the autonomous traveling unmanned aerial vehicle 110 and the GPS module 132 mounted on the mobile unmanned vehicle body 130 are configured such that the greenquality map creating unit 190 creates a point- Wherein the GPS coordinates are converted into geographical coordinates so that the GPS coordinates are displayed on the basis of the GPS-based random sampling method. The method according to claim 1,
The ultrasonic spectroscopic sensor 120 is capable of measuring the visible light region wavelength band of 400 nm to 700 nm and the near infrared region wavelength region of 700 nm to 1000 nm in hundreds of bands to measure the GPS-based random sampling method Green Tide Mapping System Utilizing Measured Data. delete The data merge processing unit (160) according to claim 1,
An image matching processing unit (161) for performing image matching processing according to a ground reference point (220) for coordinate accuracy of the measurement target area (210) with respect to the hyperspectral image collected from the data collection and correction unit (150);
An ortho-projection image creating unit 162 for creating an ortho-projection image from a plurality of hyperspectral images so as to extract accurate feature points between individual images according to an object recognition algorithm; And
Using the data measured by the GPS-based random sampling method including the optimal estimation model expression calculating unit 163 for calculating the optimal estimation model expression using the ultrasound image data and the water quality data of the measurement target region 210 Greenery mapping system. The method according to claim 1,
The correlation analysis and map mapping unit 180 analyzes the result of the regression analysis of the algae index and the concentration of chlorophyll-a in the field or the concentration of the phycocyanin, And the green tide map creating unit 190 grasps the tide concentration distribution chart of the measurement target area 210 and selects the green tide tide area through the map. The green tide area includes chlorophyll-a And calculating an area for the green-marine watershed area by setting a reference value of the concentration of at least 50 占 퐂 / L as a reference, and calculating an area for the green-marine watershed area. a) An autonomous traveling unmanned flying vehicle 110 equipped with a GPS module 111 and an ultra-spectral sensor 120 measures a measurement target area 210 in a random sampling manner along an autonomous flight path 230, Lt; / RTI >
b) measuring the chlorophyll-a concentration or the phycocyanin concentration in a random sampling manner along the autonomous movement path 240 by the GPS module 132 and the water quality measurement sensor 140;
c) examining the water surface reflectivity and the field weather conditions of the measurement target area 210;
d) collecting data from the autonomous traveling unmanned aerial vehicle (110) and the moving type unmanned vehicle (130) and correcting geometrical errors and positions so that they can be represented by measurement point geographical coordinates;
e) merging measurement data of the autonomous traveling unmanned aerial vehicle (110) and the moving unmanned vehicle (130) according to the measurement point geographical coordinates;
f) calculating an optimal estimation model equation according to the measured data from the autonomous traveling unmanned aerial vehicle (110) and the movable type unmanned vehicle (130);
g) calculating an algebra factor by inputting the calculated value of the optimal estimation model equation, the reflectivity value of the water surface, and the field weather condition, and calculating the calculated value by the artificial neural network algorithm;
h) analyzing the correlation between the optimal estimation model equation and the chlorophyll-a concentration or phycocyanin concentration measured from the mobile unattended body 130;
i) mapping the topographic map according to the analyzed correlations; And
j) creating a point-based greenhouse map for the measurement area 210
, ≪ / RTI &
The step e) includes the steps of: collecting a plurality of measurement values by a random sampling method on a measurement path that is not a predetermined measurement point; , And calculates the optimal estimation model equation for each measurement point by analyzing the correlation with the water quality data; And
Wherein the greenhouse concentration map and the greenhouse well zone area of the measurement target area 210 are calculated in the step j), using the GPS-based random sampling method. 8. The method of claim 7,
The hyperspectral sensor 120 is installed at the lower end of the autonomous traveling unmanned aerial vehicle 110 to photograph a hyperspectral image of the measurement target region 210 in a random sampling manner, The concentration of chlorophyll-a or phycocyanin in the measurement target region 210 is measured by a random sampling method fixedly provided at the rear end of the body 130, and the data measured by the GPS-based random sampling method and the water surface reflectance And a method of generating a green algae map using data measured by a GPS-based random sampling method, which uses an algae index obtained by processing a field weather condition with an artificial neural network algorithm. delete 8. The method of claim 7,
The optimal estimation model equation stored for each measurement point is calculated through correlation analysis with chlorophyll-a or phycocyanin concentration. The calculated optimum model expression value, the reflectivity of the water surface, The water quality data and the result calculated through the regression analysis are mapped to the geographical information by using the geographical classification method of the GIS together with the positionally corrected geographical coordinate information, Wherein the area for the green well zone is calculated by setting a reference value of chlorophyll-a concentration of 50 占 퐂 / L or higher as the reference value for the green well zone, by determining the greenhouse concentration distribution diagram, A method for mapping a greenery using data measured by a GPS-based random sampling method.

Images