KR20220162902A - Marine searching system using drone and the method thereof - Google Patents

Abstract

The present invention relates to a marine rescue target searching system using a drone and a method thereof. A server collects weather information, marine information, sea surface image information photographed by the drone and GPS information matched with the corresponding sea surface image information, learning in a neural network manner is performed with the same as input values to generate sea surface pattern information for each weather situation, for each marine situation and for each position, the sea surface image information photographed by the drone is transmitted in real time and is compared with the sea surface image pattern to detect a rescue target, and the real-time positional information of the rescue target is generated by calculating the current position of the rescue target based on the weather information, the marine information and GPS information at the point of time when the rescue target was photographed by the drone. Therefore, time for searching for a rescue target can be minimized and the protection of lives and property can be maximized.

Description

Marine rescue target search system and method using drone {MARINE SEARCHING SYSTEM USING DRONE AND THE METHOD THEREOF}

The present invention relates to a system and method for searching for a marine rescue object using a drone, and more particularly, to a system and method for searching for a rescue object including a person who is drifting in the sea and a person to be rescued by using a drone. .

Recently, due to the explosive growth of the fishing hobby population in Korea, the number of people who enjoy boat fishing in the ocean is increasing, and accordingly, fishing boat owners are trying to preoccupy more advantageous fishing points in order to arrive at the point faster than competing vessels due to sea conditions such as digging. It is not uncommon for ships to operate dangerously while ignoring the regulations.

Against this background, accidents in which a ship collides with waves, reefs, rival ships, etc., resulting in stranding and overturning are increasing compared to the past. Moreover, due to the impact of national catastrophic maritime accidents with enormous human damage, such as the recent sinking of the Cheonan and the sinking of the Sewol ferry, the need for technology development to respond to maritime accidents and minimize damage from maritime accidents is increasing at the national level. have.

On the other hand, in the ocean, the shipwrecked person cannot stay at the accident point and is drifted by the waves. A shipwrecked person drifting by waves can be found at a point not far from the accident site if rescue operations are carried out quickly, but in situations where rescue operations are not carried out quickly or the waves around the accident location are very strong, etc. Rescue may be difficult as it drifts far away from the accident site.

The difficulty in identifying the survivors is further aggravated when the accident point is not near the shore, but the drifting distance of the castaway is relatively long, or it is at night when the visibility is extremely limited, hindering smooth rescue work and urgent rescue operations. There is a problem that endangers the life of a castaway in a situation.

According to the above situation, the present invention uses a drone to photograph the sea level, and the photographed sea level image information includes meteorological information including at least one or more of information related to snow, rain, and clouds, and wind direction, wind speed, and atmospheric pressure. , Humidity, water temperature, wave height, wave period, and current-related information are compared and analyzed with marine information including at least one of them to create a sea level image pattern for each weather situation and marine situation, and then the sea level photographed by the drone when a distress accident occurs. A new marine rescue object search system and method capable of automatically detecting a person in distress by comparing the image with the sea level image pattern are proposed.

Next, the prior art that exists in the conventional marine search technology field will be briefly described, and then the technical details to be achieved by the present invention to be differentiated from the prior art will be described.

First, Korean Patent Publication No. 10-2019-0059081 A (published on May 30, 2019) relates to drones for marine monitoring, and more specifically, oil spills/red tides/cold water zones/low salinity/ship accidents, etc. Monitoring of various maritime accidents/disasters and diagnosis of defects in ships/marine structures can be performed easily and smoothly even with a small drone flying unmanned. After landing, horizontal movement on the sea surface is possible, expanding the activity area for checking various maritime environments and improving operational adaptability to the maritime environment. A technology related to a marine monitoring drone is described that allows seawater collection/seawater condition measurement to be carried out smoothly and easily while operation (flight, takeoff and landing, attitude control) is stably maintained.

In addition, Korean Patent Laid-open Publication No. 10-2020-0106278 A (published on September 14, 2020) relates to a device for transmitting the location of a person in distress at sea, a receiver thereof, and a drone operation system for search, and more specifically, rescue from a person in distress. Receiving a signal, determining the location of the person in distress based on the distress signal, and moving to the location of the person in distress, wherein if the rescue signal is not received from the person in distress, the previously received rescue signal is received from the monitoring device A technique related to a system for estimating a rescuer's position based on at least one piece of maritime information is described.

In addition, Korean Patent Publication No. 10-2020-0095155 A (2020.08.10. Publication date) relates to a method for monitoring marine structures using drones and a system using the same, and more specifically, a structure monitoring system for design information of marine structures. Design information input step to input to; A current information input step of photographing and analyzing current information of the offshore structure using a drone and inputting the current information to the structure monitoring system; A comparison step of comparing the design information and the current information in the structure monitoring system; and a follow-up processing step of instructing a preset follow-up action by analyzing the difference between the design information and the current information compared in the comparison step in the structure monitoring system.

Conventional technologies including the above prior art documents include technologies for monitoring and searching for marine structures and shipwrecked persons using drones, but due to the environmental characteristics of the ocean having various sea level patterns depending on weather and ocean conditions, the drones take pictures. There was a limit to accurately detecting the person in distress through only the video provided.

Korean Patent Publication No. 10-2019-0059081 A (2019.05.30. Publication date) Korean Patent Publication No. 10-2020-0106278 A (2020.09.14. Publication date) Korean Patent Publication No. 10-2020-0095155 A (2020.08.10. Publication date)

The present invention was created to solve the limitations of the prior art as described above, and the server collects weather information, marine information, sea level image information captured by a drone, and GPS information matched with the corresponding sea level image information, and inputs them. Neural-Newtork method learning is performed using values as sea level image pattern information for each weather condition, ocean condition, and location, and the sea level image information captured by the drone is transmitted in real time, and the sea level image Compared with the pattern information, the drone that detects the rescue target, calculates the current location of the rescue target based on the weather information, marine information, and GPS information at the time the rescue target was captured by the drone to generate real-time rescue target location information. The purpose is to provide a marine rescue target search system and method using the same.

A marine rescue target search system using a drone according to an embodiment of the present invention includes at least one drone including a flight means, a flight control module, a communication module, a GPS module, a battery, and a camera for photographing the sea surface; Based on the search range information for the rescue target including the communication unit communicating with the drone through a wired/wireless communication network, a marine victim and a shipwreck, a flight path allowing the drone to fly and photograph the sea surface within the search range Generates a flight path Flight route creation unit, meteorological information including at least one of information related to snow, rain, and clouds, and marine information including at least one of information related to wind direction, wind speed, air pressure, humidity, water temperature, wave height, wave period, and current information, a marine information collection unit that collects sea level image information captured by the drone and GPS information matched with the corresponding sea level image information, meteorological information, marine information, sea level image information, and GPS information collected by the marine information collection unit A sea level pattern analyzer that generates sea level image pattern information for each weather situation, ocean situation, and location by learning in a neural network (Neural-Newtork) method using input values, sea level image information captured by the drone in real time A sea level image that is received and compared with the sea level image pattern information generated by the sea level pattern analyzer to detect an image pattern different from the sea level image pattern according to the sea level image pattern information within the sea level image captured in real time by the drone An analysis unit sets an image pattern different from the sea level image pattern detected by the sea level image analysis unit as a rescue target, and the weather information and ocean information collected by the marine information collection unit and the time point at which the rescue target is captured by the drone Based on the GPS information of the rescue object position analysis unit for generating real-time rescue object location information by calculating the current location of the rescue object, and the drone flying based on the flight path generated by the flight path generator, It is characterized in that it includes; a server including a controller for transmitting a control signal to the drone through the communication unit so as to capture an image of the sea level based on search range information.

In addition, as an embodiment, the flight route generation unit is capable of photographing the sea level according to the real-time target location information while the drone flies based on the location information of the target rescued in real time calculated by the location analysis unit of the target rescued. and the control unit generates a rescue object tracking flight path that enables the rescue object tracking flight path generated by the flight route generator to generate a rescue object tracking flight path, while at least one or more of the drones fly, at sea level according to the rescue object location information in real time. It is characterized in that a control signal is transmitted to the drone through the communication unit to perform image capture.

A marine information collection unit that collects marine information including at least one of related information, sea level image information captured by the drone and GPS information matched with the corresponding sea level image information, weather information collected by the marine information collector, A sea level pattern analysis unit that generates sea level image pattern information for each weather situation, ocean situation, and location by learning in a neural network (Neural-Newtork) method using marine information, sea level image information, and GPS information as input values; The sea level image information captured by the drone is received in real time, compared with the sea level image pattern information generated by the sea level pattern analysis unit, and the sea level image according to the sea level image pattern information within the sea level image captured in real time by the drone A sea level image analysis unit that detects an image pattern different from the pattern, an image pattern different from the sea level image pattern detected by the sea level image analysis unit is set as a rescue target, and the meteorological and marine information collected by the marine information collection unit Based on the GPS information at the time the rescue object was captured by the drone, the rescue object position analysis unit calculates the current location of the rescue object and generates real-time rescue object location information, and the flight generated by the flight path generator. and a server including a control unit for transmitting a control signal to the drone through the communication unit so that the drone flies based on a path and captures an image of the sea level based on the search range information.

In addition, as an embodiment, the server allows the user to monitor information collected and generated by the server by accessing the server through at least one means of a PC, a smartphone, a tablet PC, and a dedicated terminal It is characterized in that it includes; a monitoring unit providing a user interface.

In addition, as an embodiment, the drone is equipped with at least one or more of lifesaving means including an inflatable life jacket, an inflatable lifesaving tube, an inflatable lifeboat, an underwater warming means, and a swimming pool rain, and an electrical control signal and a life-saving means unit for attaching and detaching the life-saving means from the drone body according to the drone, wherein the control unit is configured so that the drone flying along the flight path tracking the rescue object is rescued at the location of the rescue target according to the real-time rescue target location information. It is characterized in that a control signal is transmitted to the drone through the communication unit to detach from the body.

In addition, as an embodiment, the server includes an information storage unit for storing information including information collected and generated by the flight path generation unit, marine information collection unit, sea level pattern analysis unit, and rescue target location analysis unit; It is characterized by further including.

In addition, as an embodiment, the marine information collection unit collects ship information including ship specification information, location information, and radio frequency information about ships sailing in the ocean, and the control unit includes the rescue target location analysis unit. Based on the real-time rescue target location information generated from and the ship information collected by the marine information collection unit, a rescue assistance radio is transmitted through the communication unit to a ship that is determined to be able to access the current location of the rescue target within a predetermined time. characterized by

In addition, as an embodiment, the sea surface pattern analysis unit sets snow, rain, and cloud-related information in the weather information as weather information variables, and wind direction, wind speed, atmospheric pressure, humidity, water temperature, wave height, and wave period in the ocean information. and setting tide-related information as a marine information variable, setting the GPS information as a location information variable, and comparing and analyzing sea level image information captured by the drone with the weather information, the ocean information, and the GPS information, A method of learning the effects of information variables, each marine information variable, and location information variables on sea level image patterns through a neural network (Neural-Newtork) method to generate sea surface image pattern information by weather conditions, ocean conditions, and locations. characterized by

In addition, according to another embodiment of the present invention, a marine rescue object search method using a drone includes: a) at least one drone flies based on search range information on a rescue object including a marine shipwrecked person and a shipwrecked article; A flight path creation step of generating, in a server, a flight path for photographing the sea surface within the search range; b) meteorological information including at least one of snow, rain, and cloud-related information, and marine information including at least one of wind direction, wind speed, air pressure, humidity, water temperature, wave height, wave period, and current-related information in the server a marine information collection step of collecting information, sea level image information captured by the drone and GPS information matched with the corresponding sea level image information; c) The server proceeds with Neural-Newtork method learning using the weather information, ocean information, sea level image information, and GPS information collected in step b) as input values, and a sea level pattern analysis step of generating sea level image pattern information for each location. d) The server receives the sea level image information captured by the drone in real time and compares it with the sea level image pattern information generated in step c), A sea level image analysis step of detecting an image pattern different from the sea level image pattern according to the information; e) The server sets an image pattern different from the sea level image pattern detected in step d) as a rescue target, and the weather information and ocean information collected in step b) and the rescue target are captured by the drone. Based on the GPS information at the point of view, a rescue object location analysis step of generating real-time rescue object location information by calculating the current location of the rescue object; and f) a control step of transmitting a control signal from the server to the drone so that the drone flies based on the flight path generated in step a) and captures an image of the sea surface based on the search range information. It is characterized by doing.

In addition, as an embodiment, in the step a), based on the real-time location information of the rescue object calculated in the step e), the drone flies and captures the sea level according to the real-time location information of the rescue object. In the server, a flight path for tracking an object to be rescued is generated in the server, and in step f), based on the flight path for tracking an object to be rescued generated in step a), while at least one drone is flying, the location information of the object to be rescued is displayed in real time. It is characterized in that the server transmits a control signal to the drone to perform image capture of the sea surface according to the drone.

In addition, as an embodiment, in the marine rescue target search method using the drone, g) a user accesses the server through at least one of a PC, a smartphone, a tablet PC, and a dedicated terminal, and collects and and a monitoring provision step of providing a user interface for monitoring generated information from the server.

In addition, as an embodiment, in the step f), the drone flying along the rescue target tracking flight path is inflatable, a life jacket, an inflatable life-saving tube, and an inflatable at the location of the rescue target according to the real-time rescue target location information. It is characterized in that the server transmits a control signal to the drone to detach at least one or more of lifesaving means including a food lifeboat, an underwater warming means, and a swimming pool rainwater from the body of the drone.

In addition, as an embodiment, h) the marine rescue object search method using the drone may further include an information storage step of storing information including information collected and generated by the server.

In addition, as an embodiment, in the step b), the server collects ship information including ship specification information, location information, and radio frequency information about ships sailing in the ocean, and the step f) includes the Based on the location information of the rescue object generated in step e) and the ship information collected by the marine information collection unit, the server determines that the current location of the rescue object can be approached within a predetermined time. It is characterized by sending.

In addition, as an embodiment, in the step c), the server sets information related to snow, rain, and clouds in the weather information as weather information variables, and wind direction, wind speed, air pressure, humidity, water temperature, and wave height in the ocean information. , wave period and tide-related information are set as marine information variables, the GPS information is set as location information variables, and sea level image information captured by the drone is compared and analyzed with the weather information, the ocean information, and the GPS information. , sea level image pattern information by weather conditions, ocean conditions, and locations by learning the effect of each weather information variable, each ocean information variable, and location information variable on the sea level image pattern through a neural network (Neural-Newtork) method It is characterized by generating.

In the marine rescue target search system using drones, the present invention proceeds with neural network (Neural-Newtork) method learning to generate sea surface image pattern information for each weather situation, ocean situation, and location, and photographing in real time from the drone. By comparing the sea level image to be retrieved with the sea level image pattern and detecting a pattern other than the normal sea level image pattern, the marine rescue target is automatically detected with high accuracy, thereby minimizing the search time for the rescue target and maximizing the preservation of life and property. There are possible effects.

1 is a block diagram for explaining a marine rescue object search system using a drone according to an embodiment of the present invention.
2 is a flowchart for explaining a method for searching for a marine rescue object using a drone according to an embodiment of the present invention.

In each drawing of the present invention, the size or dimensions of the structures are enlarged or reduced from the actual ones for clarity of the present invention, and the known configurations are omitted so that the characteristic configurations are revealed, so they are not limited to the drawings. .

In describing the principles of preferred embodiments of the present invention in detail, if it is determined that a detailed description of a related known function or configuration may unnecessarily obscure the subject matter of the present invention, the detailed description will be omitted.

In addition, since the embodiments described in this specification and the configurations shown in the drawings are only one of the most preferred embodiments of the present invention and do not represent all of the technical ideas of the present invention, various alternatives may be used at the time of this application. It should be understood that there may be equivalents and variations.

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.

Throughout this specification, when a certain component is said to "include", it means that it may further include other components without excluding other components unless otherwise stated.

The marine rescue target search system and method using a drone according to the present invention generate sea surface image pattern information by weather conditions, marine conditions, and locations through neural-newtork learning, and in real time from a drone. The present invention relates to a method of detecting a sea level image pattern other than a normal sea level image pattern by comparing a captured sea level image with this, so that a marine rescue target can be automatically and quickly detected with high accuracy.

Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram for explaining a marine rescue object search system using a drone according to an embodiment of the present invention.

As shown in FIG. 1, the marine rescue target search system using a drone according to the present invention includes at least one drone and a server.

The drone includes a flight means, a flight control module, a communication module, a GPS module, a battery, and a camera for photographing the sea level. The camera may include at least one, and may include a visible ray camera and an infrared thermal imaging camera.

The server includes a communication unit, a flight route generation unit, a marine information collection unit, a sea level pattern analysis unit, a sea level image analysis unit, a rescue target location analysis unit, a control unit, and an information storage unit.

The communication unit communicates with the drone through a wired or wireless communication network. For reference, the communication network used in the present invention includes a wired and wireless communication network, and the wireless communication network includes a long-distance communication network including 3G, 4G, 5G or WiBro, and a local area network including Bluetooth, ZigBee, beacon, RFID, NFC, or WiFi all inclusive

In addition, the flight path generation unit generates a flight path that allows the drone to fly and photograph the sea level within the search range based on search range information for a person in distress and a rescue object including a marine victim and a shipwrecked article.

The search range information may be input from the outside, and the movement of the search target based on the minimum occurrence location and occurrence time of the input rescue target, weather information and ocean information collected by the marine information collection unit, and elapsed time. After deriving the possible location range, it may be set as the search range for the rescue target.

In addition, the marine information collection unit includes meteorological information including at least one of snow, rain, and cloud related information, and at least one or more of wind direction, wind speed, atmospheric pressure, humidity, water temperature, wave height, wave period, and current related information. It is to collect marine information to be performed, sea level image information captured by the drone, and GPS information matched with the corresponding sea level image information.

The weather information and marine information may be provided from national and private weather guidance systems, or a method of collecting the measured information from these additional measuring facilities may be used in the system of the present invention.

In addition, the sea surface pattern analyzer performs neural-newtork learning using the weather information, ocean information, sea level image information, and GPS information collected by the ocean information collection unit as input values, It is to generate sea level image pattern information for each situation and location.

More specifically, the sea surface pattern analyzer sets snow, rain, and cloud-related information in the weather information as weather information variables, and wind direction, wind speed, atmospheric pressure, humidity, water temperature, wave height, wave period, and current in the ocean information. The related information is set as a marine information variable, the GPS information is set as a location information variable, and the sea level image information captured by the drone is compared and analyzed with the weather information, the ocean information, and the GPS information to analyze each weather information variable. , Sea level image pattern information is generated by meteorological conditions, ocean conditions, and locations by learning the effect of each ocean information variable and location information variable on sea surface image patterns through a neural network (Neural-Newtork) method.

To this end, the sea surface pattern analysis unit includes an inception module-based Convolutional Neural Networks (CNN), a Deep Neural Network (DNN), a Recurrent Neural Network (RNN), and a restricted Boltzmann machine. Neural network algorithms applied with various deep learning techniques, such as a restricted Boltzmann machine, a deep belief network (DBN), and a deep Q-network, may be used.

In addition, the sea level image analysis unit receives the sea level image information captured by the drone in real time, compares it with the sea level image pattern information generated by the sea level pattern analysis unit, It is to detect an image pattern different from the sea level image pattern according to the sea level image pattern information.

In addition, the rescue target location analysis unit sets an image pattern different from the sea level image pattern detected by the sea level image analysis unit as a rescue target, and the rescue target location analysis unit combines the meteorological information and ocean information collected by the marine information collection unit with the drone. Based on the GPS information at the time the object was photographed, the current location of the object to be rescued is calculated to generate real-time location information of the object to be rescued.

That is, the rescue object position analysis unit derives the current location of the rescue object by analyzing the movement of the rescue object from the initial location to the current location in the ocean by applying weather information and detailed variables in the marine information, and thus such a rescue object It is preferable that the current location of is not a specific location but a range location according to probability.

In addition, the controller transmits a control signal to the drone through the communication unit so that the drone flies based on the flight path generated by the flight path generator and captures an image of the sea surface based on the search range information.

In addition, the information storage unit stores and manages information including information collected and generated by the flight path generation unit, marine information collection unit, sea level pattern analysis unit, and salvage target location analysis unit in a database.

The database is a local storage unit for storing all data used in the system of the present invention, and usually means an internal memory or an external memory, but in the present invention, a virtualized storage device of a cloud computing network in the network for data backup ( Briefly referred to as cloud storage) may be further included. For example, the cloud storage device may further include a dedicated cloud storage device provided by a service provider or a free cloud storage device (or network HDD) provided by a portal site (eg, Naver, Daum, etc.). That is, the database is a database connected to the server through an electric or optical cable, a network including cloud computing, or a combination thereof.

In addition, the server as well as the database is implemented in a cloud computing server environment, so there is no need for a service provider to build a separate server as hardware, and the application (app) software according to the present invention is provided as a kind of service (eg: SaaS (Software as a Service) can be used so that users can select the software they need and receive the service.

For reference, in the above embodiment, each component constituting the rescue target search system using a drone is only described for convenience of explanation, and practically all operations can be processed by any one component, and more diverse configurations. It may be subdivided into means or configured to process by combining the constituent means. In addition, configuration means commonly used in providing services to users based on a server are omitted because they are not necessarily necessary parts in explaining the present invention.

On the other hand, in the marine rescue target search system using a drone according to the present invention, when a rescue target is detected, it may be controlled to track and photograph the detected rescue target using a drone.

To this end, the flight path creation unit enables the drone to fly and photograph the sea level according to the location information of the rescue object based on the location information of the rescue object in real time calculated by the location analysis unit of the rescue object. Create a tracking flight path, and the control unit performs image capture of the sea level according to the real-time location information of the rescue object while at least one or more drones fly based on the flight path for tracking the rescue object generated by the flight route generator. A control signal is transmitted to the drone through the communication unit.

At this time, in selecting the drone to track and photograph the rescue object, the possibility of tracking and photographing the rescue object can be maximized by selecting the drones in the order of neighboring drones from the current location of the rescue object derived from the location analysis unit. .

In addition, in the marine rescue target search system using a drone according to the present invention, when the rescue target is a human, the drone tracking and photographing the rescue target provides a means of survival to the rescue target in order to ensure the safety of the rescue target and increase the possibility of survival. You may.

More specifically, the drone is equipped with at least one or more of life-saving means including an inflatable, a life jacket, an inflatable life-saving tube, an inflatable lifeboat, an underwater warming means, and a pool rain, and according to an electrical control signal and a life-saving means for attaching and detaching from the drone body, wherein the control unit causes the drone flying along the rescue target tracking flight path to attach and detach the life-saving means from the drone body at the position of the rescue target according to the real-time rescue target location information. A control signal is transmitted to the drone through the communication unit.

In addition, the marine rescue target search system using a drone according to the present invention is another method of securing the safety of the rescue target and increasing the survivability, and can request rescue support from a ship adjacent to the location of the rescue target.

More specifically, the marine information collection unit collects ship information including ship specification information, location information, and radio frequency information about ships sailing in the ocean, and the control unit collects ship information generated by the rescue target location analysis unit in real time. Based on the location information of the rescue object and the ship information collected by the marine information collection unit, a rescue assistance radio is transmitted through the communication unit to a ship that is determined to be able to access the current location of the rescue object within a predetermined time.

In addition, the monitoring server provides a user interface that allows a user to access the server through at least one of a PC, a smartphone, a tablet PC, and a dedicated terminal and monitor information collected and generated by the server. Wealth may be further included.

Through the user interface provided by the monitoring unit, people involved in the rescue of the rescue target, such as system administrators, coast guards, and 119, can check the search progress of the rescue target, whether the rescue target has been detected, the current location of the rescue target, etc. from a remote location, and also the user interface Through this, it is possible to control the system of the present invention.

2 is a flowchart for explaining a method for searching for a marine rescue object using a drone according to an embodiment of the present invention.

As shown in FIG. 2, the marine rescue target search method using drones includes: a) flight path creation step, b) marine information collection step, c) sea level pattern analysis step, d) sea level image analysis step, e) rescue A target position analysis step and f) a control step.

In the step a), based on the search range information for the person in distress and the rescue object including the shipwrecked goods, a flight path allowing at least one or more drones to fly and photograph the sea surface within the search range is generated in the server.

The step b) may include weather information including at least one of snow, rain, and cloud related information, and at least one or more of wind direction, wind speed, air pressure, humidity, water temperature, wave height, wave period, and current related information in the server. Marine information including , sea level image information captured by the drone, and GPS information matched with the corresponding sea level image information are collected.

In step c), the server proceeds with neural network (Neural-Newtork) learning using the meteorological information, marine information, sea level image information, and GPS information collected in step b) as input values, It creates sea level image pattern information for each marine situation and location.

More specifically, step c) sets snow, rain, and cloud related information in the weather information as weather information variables in the server, and wind direction, wind speed, air pressure, humidity, water temperature, wave height, and wave in the ocean information. Period and current-related information is set as marine information variables, the GPS information is set as location information variables, and sea level image information captured by the drone is compared and analyzed with the weather information, ocean information, and GPS information, Creates sea level image pattern information by weather conditions, ocean conditions, and locations by learning the effects of weather information variables, each ocean information variable, and location information variables on sea level image patterns through a neural network (Neural-Newtork) method do.

Step d) receives the sea level image information captured by the drone in real time from the server, compares it with the sea level image pattern information generated in step c), An image pattern different from the sea level image pattern according to the sea level image pattern information is detected.

In step e), the server sets an image pattern different from the sea surface image pattern detected in step d) as a rescue target, and the weather information and ocean information collected in step b) and the rescue target in the drone Based on the GPS information at the time of the photographing, the current location of the rescue object is calculated to generate real-time location information of the rescue object.

In step f), the server transmits a control signal to the drone so that the drone flies based on the flight path generated in step a) and captures an image of the sea surface based on the search range information.

In addition, in the step f), the drone flying along the flight path tracking the rescue object is inflatable, life jacket, inflatable lifeboat, inflatable lifeboat, and number at the location of the rescue object according to the real-time location information of the rescue object. The server transmits a control signal to the drone to detach at least one or more of the medium heat preservation means and the lifesaving means including the swimming pool rainwater from the body of the drone.

In addition, in the step a), based on the real-time location information of the rescue object calculated in the step e), the drone flies and captures the sea level according to the real-time location information of the rescue object. A path is created in the server, and in step f), based on the flight path for tracking the object to be rescued created in step a), at least one or more of the drones fly while capturing an image of the sea surface according to the location information of the object to be rescued in real time. The server transmits a control signal to the drone to perform

In addition, in the marine rescue target search method using a drone according to the present invention, in step g), the user accesses the server through at least one means of a PC, smartphone, tablet PC, and a dedicated terminal, and collects data from the server. and a monitoring providing step of providing, from the server, a user interface for monitoring generated information.

In addition, the marine rescue target search method using the drone according to the present invention may further include, as step h), an information storage step of storing information including information collected and generated by the server.

In addition, in the step b), the server collects ship information including ship specification information, location information, and radio frequency information about ships sailing in the ocean, and in the step f), the generation in the step e) is performed. Based on the real-time location information of the object to be rescued and the ship information collected by the marine information collection unit, the server transmits a rescue assistance wireless signal to a ship that is determined to be able to access the current location of the object to be rescued within a predetermined time.

On the other hand, in the marine rescue target search system and method using a drone according to the present invention, in order to more quickly photograph the search range when a marine rescue target occurs, a space in which at least one drone is stored and taken off and landed on the sea is provided, A drone station equipped with a charging means capable of charging the battery of the landed drone may be additionally included.

In addition, the drone station may be provided with means for generating power by using at least one of waves, wind, and solar heat, and the generated power may be used to charge the battery of the drone, and a communication module is provided to Communication between the server and the drone may be relayed.

The present invention has been described above with reference to the embodiments shown in the accompanying drawings, but these are only exemplary, and various modifications and other equivalent embodiments can be made by those skilled in the art in the art. will understand that Therefore, the technical protection scope of the present invention should be determined by the claims below.

100: drone
110: means of flight
120: flight control module
130: communication module
140: GPS module
150: battery
160: camera
200: server
210: Ministry of Communication
220: flight path generation unit
230: marine information collection department
240: sea level pattern analysis unit
250: sea level image analysis unit
260: Rescue target location analysis unit
270: control unit
280: information storage unit

Claims (14)

At least one drone including a flight means, a flight control module, a communication module, a GPS module, a battery, and a camera for photographing the sea level;
A communication unit communicating with the drone through a wired/wireless communication network;
A flight path creation unit that generates a flight path that allows the drone to fly and photograph the sea surface within the search range based on search range information for a person in distress and a rescue target including a shipwrecked person in the sea;
Meteorological information including at least one of snow, rain and cloud related information, marine information including at least one of wind direction, wind speed, air pressure, humidity, water temperature, wave height, wave period and current related information, and the drone A marine information collection unit that collects sea level image information captured in and GPS information matched with the corresponding sea level image information;
Using the meteorological information, marine information, sea level image information, and GPS information collected by the marine information collection unit as input values, neural network (Neural-Newtork) learning is performed to obtain sea level images by weather conditions, sea conditions, and locations. A sea level pattern analysis unit for generating pattern information;
The sea level image information captured by the drone is received in real time, compared with the sea level image pattern information generated by the sea level pattern analysis unit, and the sea level according to the sea level image pattern information within the sea level image captured in real time by the drone A sea level image analysis unit that detects an image pattern different from the image pattern;
An image pattern different from the sea level image pattern detected by the sea level image analysis unit is set as a rescue target, weather information and ocean information collected by the marine information collection unit, and GPS information at the time the rescue target was captured by the drone Based on, a rescue object position analysis unit for generating real-time rescue object location information by calculating the current location of the rescue object, and
A server including a controller that transmits a control signal to the drone through the communication unit so that the drone flies based on the flight path generated by the flight path generator and captures an image of the sea level based on the search range information; Characterized in that it comprises a, marine rescue target search system using a drone.
According to claim 1,
The flight path generation unit,
Based on the real-time rescue target location information calculated by the rescue target location analysis unit, a rescue target tracking flight path is created so that the drone can fly and photograph the sea level according to the rescue target location information,
The control unit,
Transmits a control signal to the drone through the communication unit so that at least one or more drones fly based on the flight path for tracking the rescue object generated by the flight path generation unit and capture images of the sea surface according to the location information of the rescue object in real time. Characterized in that, a marine rescue target search system using a drone.
According to claim 2,
The server,
A monitoring unit providing a user interface that allows a user to monitor information collected and generated by the server by accessing the server through at least one of a PC, a smartphone, a tablet PC, and a dedicated terminal; Characterized in that, a marine rescue target search system using a drone.
According to claim 2,
The drone,
At least one of the lifesaving means including an inflatable, life jacket, inflatable life tube, inflatable lifeboat, underwater warming means, and swimming pool rain is mounted, and a lifesaving means part detaching it from the drone body according to an electrical control signal including;
The control unit,
Characterized in that the drone flying along the rescue target tracking flight path transmits a control signal to the drone through the communication unit so that the lifesaving means is detached from the drone body at the location of the rescue target according to the real-time rescue target location information. A marine rescue target search system using drones.
According to claim 1,
The server,
An information storage unit for storing information including information collected and generated by the flight path generation unit, marine information collection unit, sea level pattern analysis unit, and salvage target location analysis unit; characterized in that it further comprises a drone using Marine rescue target search system.
According to claim 1,
The marine information collection unit,
Collect ship information including ship specification information, location information and radio frequency information for ships sailing in the ocean,
The control unit,
Based on the location information of the object to be rescued in real time generated by the location analysis unit of the object to be rescued and the ship information collected by the marine information collection unit, the ship is determined to be able to approach the current location of the object to be rescued within a predetermined time through the communication unit. A marine rescue target search system using a drone, characterized in that it transmits a rescue assistance radio.
According to claim 1,
The sea level pattern analysis unit,
Snow, rain, and cloud-related information in the weather information is set as weather information variables, and wind direction, wind speed, atmospheric pressure, humidity, water temperature, wave height, wave period, and current-related information in the ocean information are set as ocean information variables, GPS information is set as a location information variable, and sea level image information captured by the drone is compared and analyzed with the weather information, ocean information, and GPS information, so that each weather information variable, each ocean information variable, and location information variable is the sea level. A marine rescue target search system using a drone, characterized by generating sea surface image pattern information by weather situation, marine situation, and location by learning the effect on the image pattern through a neural network (Neural-Newtork) method. .
a) Based on the search range information on the person in distress and the rescue target including the shipwrecked goods, at least one drone flies and the flight path that allows the sea surface to be photographed within the search range is generated in the server. A flight path creation step;
b) meteorological information including at least one of snow, rain, and cloud-related information, and marine information including at least one of wind direction, wind speed, air pressure, humidity, water temperature, wave height, wave period, and current-related information in the server a marine information collection step of collecting information, sea level image information captured by the drone and GPS information matched with the corresponding sea level image information;
c) The server proceeds with Neural-Newtork method learning using the weather information, ocean information, sea level image information, and GPS information collected in step b) as input values, and a sea level pattern analysis step of generating sea level image pattern information for each location.
d) The server receives the sea level image information captured by the drone in real time and compares it with the sea level image pattern information generated in step c), A sea level image analysis step of detecting an image pattern different from the sea level image pattern according to the information;
e) The server sets an image pattern different from the sea level image pattern detected in step d) as a rescue target, and the weather information and ocean information collected in step b) and the rescue target are captured by the drone. Based on the GPS information at the point of view, a rescue object location analysis step of generating real-time rescue object location information by calculating the current location of the rescue object; and
f) a control step of transmitting a control signal from the server to the drone so that the drone flies based on the flight path generated in step a) and captures an image of the sea surface based on the search range information; Characterized in that, a marine rescue target search method using a drone.
According to claim 8,
In step a),
Based on the real-time location information of the rescue object calculated in step e), the server creates a flight path for tracking the rescue object so that the drone can fly and photograph the sea level according to the real-time location information of the rescue object,
In step f),
Transmitting a control signal from the server to the drone so that at least one or more of the drones fly based on the rescue target tracking flight path generated in step a) and capture images of the sea surface according to the real-time rescue target location information Characterized in that, a marine rescue target search method using a drone.
According to claim 9,
The marine rescue target search method using the drone,
g) Monitoring that the server provides a user interface that allows a user to monitor information collected and generated by the server by accessing the server through at least one of a PC, a smartphone, a tablet PC, and a dedicated terminal A marine rescue object search method using a drone, characterized in that it comprises a; providing step.
According to claim 9,
In step f),
The drone flying along the rescue target tracking flight path includes an inflatable, a life jacket, an inflatable life tube, an inflatable lifeboat, an underwater warming means, and a swimming pool fee at the location of the rescue target according to the real-time rescue target location information. A marine rescue target search method using a drone, characterized in that the server transmits a control signal to the drone to detach at least one of the lifesaving means from the drone body.
According to claim 8,
The marine rescue target search method using the drone,
h) an information storage step of storing information including information collected and generated by the server;
According to claim 8,
In step b),
The server collects ship information including ship specification information, location information, and radio frequency information about ships sailing in the ocean,
In step f),
Based on the location information of the object to be rescued in real time generated in the step e) and the ship information collected by the marine information collection unit, the ship is determined to be able to access the current location of the object to be rescued within a predetermined time. A marine rescue target search method using a drone, characterized in that for transmitting.
According to claim 8,
In step c),
In the server, information related to snow, rain, and clouds in the weather information is set as weather information variables, and information related to wind direction, wind speed, air pressure, humidity, water temperature, wave height, wave period, and current in the ocean information is set as ocean information variables. and sets the GPS information as a location information variable, and compares and analyzes the sea level image information captured by the drone with the weather information, the ocean information, and the GPS information to analyze each weather information variable, each ocean information variable, and location information. Marine rescue using drones, characterized by generating sea level image pattern information for each weather situation, ocean situation, and location by learning the effect of variables on sea surface image patterns through a neural network (Neural-Newtork) method How to search for a target.

Images