Automatic rescue system for tourist area and control method

Abstract

The invention provides an automatic rescue system for a tourist area and a control method. The automatic rescue system comprises a rescue trolley, a rescue unmanned aerial vehicle, a detection camera and a dangerous case analysis system. The detection camera is arranged on a stand column in a tourism area and connected with the dangerous case analysis system, and the stand column is provided with a weather detector connected with the dangerous case analysis system and used for carrying out image collection and weather collection on tourists and feeding back the image collection and weather collection to the dangerous case analysis system; the dangerous case analysis system carries out tourist dangerous case analysis according to the collected data; the dangerous case analysis system is connected with the rescue trolley and the rescue unmanned aerial vehicle so as to control the rescue trolley and the rescue unmanned aerial vehicle to go to the tourist in distress for rescue. According to the scheme, whether tourists in all areas are in danger or not can be automatically detected through the cameras on the way, an alarm can be given in time when the tourists are in danger, rescue is conducted in time through the rescue trolley and the rescue unmanned aerial vehicle, and the problems that the tourists cannot be found in time and rescue cannot be conducted in time are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of tourist area rescue, and in particular to an automatic rescue system and a control method for tourist areas. Background Art

[0002] With the development of economy and society, travel has become a must-have option for many people during holidays. With the development of tourism, the number of tourists is also increasing rapidly, which will inevitably bring various personal safety issues.

[0003] At present, when dangerous situations occur in tourist areas, rescue is mainly carried out in the following ways:

[0004] Rescue by scenic area rescue team: Rescue is carried out by professional rescue personnel within the scenic area. For example, rescue team members in mountain scenic areas are familiar with the mountain environment and rescue techniques, and can quickly find missing or injured tourists in the mountains and forests; rescue personnel in water scenic areas have professional water rescue skills and can rescue drowning tourists in time.

[0005] Mutual rescue among tourists: Tourists are guided to rescue each other in emergency situations through scenic area broadcasts and signs. For example, in crowded scenic areas, when a tourist suddenly falls ill, nearby tourists can perform simple first aid treatment, such as cardiopulmonary resuscitation, hemostatic bandages, etc., before medical staff arrive.

[0006] Firefighting and rescue: When major disasters such as fire, earthquake, building collapse, etc. occur in the scenic area, the fire brigade will be dispatched quickly, using professional fire-fighting equipment, demolition tools, rescue ropes, etc. to extinguish the fire, search and rescue trapped tourists, and carry out emergency rescue work.

[0007] Water rescue: In water tourist areas, rescuers will use speedboats, rubber boats, water robots and other equipment for rescue. They will also be equipped with divers and underwater detection equipment to deal with situations such as drowning and shipwrecks. They will also provide water safety education and training to tourists to improve their self-rescue capabilities.

[0008] The above methods are effective in case of some emergencies, such as falling into the water in the lifeboat area, being bitten by animals, dizziness caused by low blood sugar in tourists during mountain climbing, hypothermia, and other situations that cause people to lose their ability to move. According to the current rescue method, there will be problems that cannot be discovered in time. For example, when tourists encounter the above conditions during certain times and sections with low traffic in mountain scenic spots, if no one discovers them in time or there is a problem with communication, the best rescue time will be missed and the lives of tourists will be endangered. Another problem is that it is impossible to rescue in time after discovery, because rescue personnel and rescue materials are generally at the rescue point. Generally, tourists are far away from the rescue point. It often takes a long time to prepare materials and then arrive over a long distance, so timely rescue cannot be carried out.

[0009] In view of this, an automatic rescue system and control method for a tourist area are needed. Summary of the invention

[0010] In view of the problem that tourists in danger cannot be discovered and rescued in time in the prior art, the present invention provides an automatic rescue system and control method for tourist areas, which can automatically detect whether tourists in various areas are in danger through cameras along the way, and can promptly issue an alarm and provide timely rescue through rescue vehicles and rescue drones when danger occurs, thus solving the problem of not being able to discover and rescue in time. The specific technical solution is as follows:

[0011] An automatic rescue system for a tourist area comprises a rescue vehicle, a rescue drone, a detection camera and a danger analysis system; the detection camera is arranged on a pillar in the tourist area and connected to the danger analysis system; a meteorological detector connected to the danger analysis system is arranged on the pillar for collecting images and meteorological data of tourists and feeding them back to the danger analysis system; the danger analysis system performs danger analysis on tourists based on the collected data; the danger analysis system is connected to the rescue vehicle and the rescue drone respectively to control the rescue vehicle and the rescue drone to go to the place where the tourists are in distress for rescue.

[0012] Furthermore, the rescue vehicle includes a wheeled mobile structure and a tracked mobile structure.

[0013] Furthermore, the rescue vehicle is provided with an odor releasing device or an ultrasonic device; the odor releasing device and the ultrasonic device are respectively connected to the controller of the rescue vehicle to release odor or ultrasonic waves to drive away animals.

[0014] Furthermore, the rescue drone is provided with a delivery device for delivering rescue items.

[0015] Furthermore, the delivery device includes a medicine delivery device for delivering medicines in a timely manner.

[0016] Furthermore, the launching device includes a net launching device, which is used to provide timely rescue to people or animals who fall into the water.

[0017] A control method for an automatic rescue system in a tourist area, used to control the automatic rescue system in the tourist area described above, comprises the following steps:

[0018] The detection camera collects images of tourists in real time and feeds them back to the danger analysis system;

[0019] The danger analysis system performs danger analysis on the tourist images and determines whether the tourist is in danger;

[0020] If it is determined that the tourist is in distress, the type of distress the tourist is in is further analyzed;

[0021] According to the type of distress the tourists are in, control the corresponding rescue vehicle or rescue drone to rescue the tourists.

[0022] Furthermore, the method further comprises the following steps:

[0023] The detection camera collects images of the water area in the tourist area in real time and feeds them back to the danger analysis system;

[0024] The danger analysis system analyzes the water area image to determine whether there are people or animals falling into the water;

[0025] If it is determined that a person or animal has fallen into the water, the rescue drone carrying the net-dropping device is controlled to drop the net to carry out the rescue.

[0026] Furthermore, the method further comprises the following steps:

[0027] Preset marks are placed on cruise ships in the water area so that the hazard analysis system can exclude the marked cruise ships when conducting hazard analysis.

[0028] Furthermore, when analyzing the water area image to determine whether a person or an animal has fallen into the water, the water splash feature appearing in the water area image is used as a basis for determining whether a person or an animal has fallen into the water.

[0029] Compared with the prior art, the present invention has the following beneficial effects:

[0030] 1. It can automatically detect whether tourists in various areas are in danger through cameras along the way. When danger occurs, it can promptly issue an alarm and provide timely rescue through rescue vehicles and rescue drones, solving the problem of not being able to discover and rescue in time. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the description of the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0032] Figure 1 It is a structural schematic diagram of an automatic rescue system for a tourist area;

[0033] Figure 2 The present invention is a flow chart of a control method for an automatic rescue system in a tourist area. DETAILED DESCRIPTION

[0034] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0035] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0036] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0037] It should be further understood that the term "and / or" used in the present description and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0038] Embodiment 1

[0039] like Figure 1 The figure shows a schematic diagram of the structure of an automatic rescue system for a tourist area, including a rescue vehicle, a rescue drone, a detection camera and a danger analysis system; the detection camera is arranged on a pillar in the tourist area and connected to the danger analysis system, the pillar is provided with a meteorological detector connected to the danger analysis system, which is used to collect images and meteorological data of tourists and feed them back to the danger analysis system, and the danger analysis system performs danger analysis on tourists based on the collected data; the danger analysis system is connected to the rescue vehicle and the rescue drone respectively to plan a navigation route according to the positions of the rescue vehicle and the rescue drone and the positions of tourists, and control the rescue vehicle and the rescue drone to go to the tourists in distress for rescue.

[0040] In a specific implementation, the rescue vehicle includes a wheeled mobile structure and a tracked mobile structure.

[0041] In specific implementation, when the tourist route is a step-free road, a wheeled mobile rescue vehicle is preferred. For example, the Jiushi Intelligent L4-level unmanned vehicle Z5 is equipped with dual Orin Drive kits, has a computing power of more than 500tops, and uses a perception solution of 14 cameras and 4 solid-state laser radars. Its autonomous driving system uses the ZOE 2.0 architecture, which can achieve the connection of sensor expansion, and through the backbone network, uses Transformer to complete the unified detection of dynamic and static obstacles and map elements. With its powerful perception and autonomous driving capabilities, it can detect abnormal situations in security inspections and other scenes and alarm in time, and provide early support for rescue work by matching other specialized rescue components (such as fire extinguishing systems, ultrasonic devices, and odor release devices).

[0042] Another example is the "Bin Xiaoxin" police unmanned patrol car, which uses the most advanced artificial intelligence, 5G communication, high-definition camera and other technologies, and realizes L4 level highly automated driving, intelligent monitoring and rapid response through visual and perception systems, and can cruise and monitor 24 hours a day. It has functions such as illegal capture, road condition monitoring, remote shouting, one-button alarm and safety publicity and defense. The vehicle is equipped with an intelligent recognition system that can identify traffic violations and potential safety hazards, and can provide timely feedback to the command center, and can realize shouting reminders and remote capture of illegal behaviors. In case of emergency, citizens can scan the QR code on the vehicle body to connect to the service background for help, and police officers can obtain their location in real time and view the nearby overview. The storage compartment in the car is equipped with fire extinguishers, rescue ropes, lifebuoys, riot control equipment, fire blankets, first aid equipment, etc., which can be used by police officers and authorized citizens in emergencies.

[0043] In addition to the two rescue vehicles listed above, other vehicles or solutions with the same functions can also be used. At the same time, emergency supply cabinets such as emergency medicines and food, water, clothing, umbrellas, raincoats, etc. can be added to the existing rescue vehicles to provide tourists with various emergency supplies they need.

[0044] Furthermore, on each route in the tourist area, automatic standby rescue vehicles are placed at certain intervals, so that they can arrive at the scene in time when a dangerous situation occurs, solving the problem that rescue personnel cannot be on standby 24 hours a day, and increasing the probability of successful rescue. Especially in thunderstorms and at night, the rescue vehicles can be on standby outdoors at any time, achieving a rescue effect that human power cannot achieve.

[0045] In specific implementation, when the tourist route is a mountain road with stairs, a rescue vehicle with a track-type mobile structure is preferred. The specific structure is as follows:

[0046] Frame

[0047] Material and strength: The frame is made of high-strength alloy steel. This material has excellent torsion and compression resistance, and can withstand various stresses caused by complex road conditions, heavy loads and tilt angles when moving on the mountain. The overall shape of the frame is generally designed as a box-type or frame-type reinforced structure. The box-type frame can provide better protection for internal equipment, and the frame-type reinforced structure is convenient for the installation and maintenance of various components. At the same time, reinforcement treatment is carried out at key locations, such as adding reinforcing ribs, to ensure that the overall structure is stable and reliable. Size and layout: The size of the frame is determined according to actual usage requirements and track specifications. It is necessary to ensure that there is enough space to install related components such as drive, brake, and cargo. In terms of layout, attention should be paid to the reasonable distribution of the center of gravity, and the center of gravity should be kept as low as possible and close to one side of the track to reduce the risk of instability such as rollover when driving on an inclined track.

[0048] Wheel and track system

[0049] Wheel type. Driving wheel: Usually a special steel wheel is used, and the wheel surface has a pattern that increases friction or a special rubber coating to improve the grip on the inclined track. The size of the driving wheel is selected according to the load weight and the slope of the track. Generally, larger wheels have more advantages when climbing. At the same time, the rim part must be precisely processed to ensure a close fit with the track to prevent derailment. Driven wheel: Also made of steel material, its structure matches the driving wheel. It mainly plays the role of auxiliary support and stabilization of the trolley. The number is reasonably arranged according to the length of the trolley and the load requirements. Multiple driven wheels can share the weight of the vehicle body more evenly and ensure stability during driving.

[0050] Track. Track material: High-strength steel rails are mostly used, which have good wear resistance and compression resistance, and can withstand repeated rolling of the car and erosion of the mountain environment for a long time. The cross-sectional shape of the track is designed according to the type of wheel. Common ones are I-shaped, trough-shaped, etc., to ensure that the wheel runs accurately on the track. Track laying: When laying on the mountain, it is necessary to make reasonable planning according to the terrain of the mountain. Through professional measurement and design, the track is laid along the inclination angle of the mountain, and a suitable curve radius is set. The track is fixed with solid anchor bolts, concrete foundations, etc., to firmly fix the track on the mountain to prevent the track from shifting during the operation of the trolley. Track anti-skid and drainage: In order to cope with the possible wet and rainy conditions on the mountain, the surface of the track will be anti-skid, such as adding anti-skid patterns or using special anti-rust and anti-skid coatings. At the same time, drainage slopes and drainage holes are set on the track to drain rainwater in time to avoid water accumulation on the track affecting the safety of the trolley.

[0051] Drive system

[0052] Motor: If there is electricity supply on the mountain, a high-power variable-frequency speed-regulating motor is often used as the power source. Its advantages are easy control, smooth start, and the ability to flexibly adjust the speed and torque according to different slopes and loads to ensure that the trolley can climb the inclined track smoothly. And it can achieve forward and reverse rotation through the control system to meet the needs of the trolley's round-trip operation.

[0053] Internal combustion engine (optional): In some remote mountainous areas where there is no convenient electricity supply, internal combustion engines such as diesel engines are also used. They have strong power output and long endurance, but are relatively noisy and emit exhaust gas, requiring corresponding heat dissipation and exhaust gas treatment systems.

[0054] Transmission mechanism. Rack and pinion transmission: The power of the electric motor (or internal combustion engine) is transmitted to the gear through the reducer, and the gear meshes with the rack laid on one side of the track. This transmission method has high transmission efficiency and high precision, and can accurately control the moving speed and position of the trolley. It is particularly suitable for use in occasions with large inclinations and requiring precise docking. Chain drive (or wire rope drive): If the track is long and the slope is relatively gentle, chain or wire rope drive can also be used. The chain drive structure is relatively simple and low in cost. The power is transmitted to the drive wheel through the sprocket; the wire rope drive needs to be matched with the corresponding pulley set. The advantage is that it can adapt to complex terrain changes and can cross obstacles at a certain distance. However, it is necessary to regularly check the wear of the wire rope to ensure its safety.

[0055] Braking system

[0056] Mechanical braking: Braking is achieved by friction between the brake pad and the wheel or brake disc. The brake pad is generally made of wear-resistant and high-temperature resistant materials. When braking, pressure is applied through hydraulic pressure, air pressure or mechanical connecting rods to make the brake pad hold the wheel or brake disc tightly and achieve rapid stopping. On inclined tracks, in order to prevent the trolley from sliding down, multiple sets of mechanical brakes are usually set up, distributed in different wheel positions, and have two control modes: manual and automatic triggering, to ensure reliable braking in emergency situations and parking conditions. Electrical braking: For trolleys driven by electric motors, the regenerative braking function of the motor is used. When deceleration or parking is required, the power supply mode of the motor is changed to make the motor a generator, and the kinetic energy of the trolley is converted into electrical energy to feed back to the power grid (if conditions permit) or consumed in the brake resistor, which plays the role of auxiliary braking. This braking method can reduce the wear of the brake pads and achieve smooth deceleration.

[0057] Auxiliary braking and safety devices.

[0058] Ratchet and pawl device: A ratchet and pawl mechanism is installed on the driving shaft or wheel shaft of the trolley. When the trolley is moving upward normally, the pawl and the ratchet wheel slide relative to each other without affecting the rotation of the wheel. Once a downward trend occurs, the pawl will immediately jam the ratchet wheel to prevent the wheel from reversing, thus preventing the trolley from losing control and sliding down, adding a safeguard for safe driving on the inclined track. Emergency brake cable: An emergency brake cable is installed along one side of the track or at a suitable position on the vehicle body. Cable handles are provided at key positions along the track and on the trolley. Once an emergency occurs, the operator or staff along the line can pull the cable to trigger the brake system and stop the trolley quickly.

[0059] Steering system

[0060] Turning on a curve: If the track has a curve on the mountain, the trolley's steering system can adopt a bogie structure, which is similar to the steering principle of a train. The bogie is connected to the frame through a special connection device, and the wheels are installed on the bogie. When entering a curve, the trolley can smoothly travel along the curve by relying on the lateral force of the track on the wheel and the rotation flexibility of the bogie itself, ensuring that the wheel and the track always maintain good contact and rolling state, reducing friction and wear between the wheel rim and the track. Active steering control (optional): For some situations that require more precise steering control, such as at complex mountain track intersections or places with special operating requirements, an active steering control system can be equipped. Sensors detect information such as track curvature and trolley speed, and then use electric or hydraulic steering mechanisms to accurately adjust the steering angle of the wheels to achieve intelligent steering operations.

[0061] Cargo and protective structures

[0062] Cargo platform: A solid cargo platform is set above the frame. The surface of the platform is usually paved with anti-slip patterned steel plates to facilitate the placement of goods and prevent the goods from sliding during transportation. A fence of a certain height is set around the platform. The fence can be detachable or foldable to facilitate loading and unloading of goods. At the same time, the strength of the fence should be sufficient to prevent the goods from falling when the trolley tilts and moves and may encounter bumps. Body protection: Anti-collision guardrails are installed around the body of the trolley. They are made of solid materials such as steel pipes to prevent the trolley from colliding with obstacles such as mountain rocks and trees during operation and causing damage. At the same time, the body surface is painted with protective coatings such as rust-proof paint and wear-resistant paint to resist environmental influences such as wind, sun, rain erosion on the mountain. Personnel protection: For trolleys with operators, a cab or riding area that meets safety standards is set up, equipped with safety facilities such as seat belts and guardrails, and the cab adopts a good field of view design to facilitate operators to observe the track and surrounding conditions to ensure the safety of personnel during the movement of the trolley.

[0063] Control System

[0064] Control center: A hazard analysis system (remote control situation) usually installed in a control room on the vehicle or at the foot of a mountain. It is composed of core equipment such as a programmable logic controller (PLC), an industrial computer, and a server. It receives signals from various sensors, such as speed sensors, position sensors, tilt angle sensors, brake pressure sensors, etc., and centrally controls the vehicle's drive, braking, steering and other systems.

[0065] Sensor Applications:

[0066] Speed ​​sensor: monitors the speed of the trolley in real time and feeds the speed signal back to the control system so that the speed can be adjusted according to the set speed range to prevent the trolley from speeding on the inclined track and causing danger.

[0067] Position sensor: By sensing specific marks on the track or using encoders, the position of the trolley on the track can be accurately determined to achieve functions such as precise parking and fixed-point operation. Limit sensors are also set at key positions and end points of the track to prevent the trolley from crossing the boundary and derailing.

[0068] Tilt angle sensor: always detects the tilt angle of the track and the tilt state of the trolley itself. When the angle exceeds the safe range, it triggers an alarm in time and takes corresponding braking or adjustment measures to ensure the driving safety of the trolley.

[0069] Communication system: Equipped with wireless communication modules (such as radios, digital radios, etc.) or wired communication cables (if conditions permit) to realize information exchange between the trolley and the control center at the foot of the mountain and other related equipment, so that operators can remotely monitor the operating status of the trolley, issue operating instructions, and receive fault alarm information.

[0070] This kind of rail trolley structure, through the coordination of various parts, can realize the functions of transporting goods or people on the inclined track on the mountain in a relatively safe and stable manner. The specific design can also be further optimized and adjusted according to the actual mountain environment characteristics, transportation requirements and other factors. Some existing models are as follows:

[0071] Yuantai Machinery Ivy Cargo Train: The running slope can reach 45°, the rated load is 200kg, the rated speed is about 38m / Min, and it consists of a drive device, a clutch device, a brake device, a track anti-rollover and anti-derailment device, a main frame, a trailer and a single track with teeth.

[0072] Shandong Gonow JA-DG: The body size is 1950550300.

[0073] Zhuji Jinhou Amusement Equipment Manufacturing Co., Ltd. jinhou007: Can be customized and used for climbing tracks in scenic spots.

[0074] In a specific implementation, the rescue vehicle is provided with an odor releasing device or an ultrasonic device; the odor releasing device and the ultrasonic device are respectively connected to the controller of the rescue vehicle to release odor or ultrasonic waves to drive away animals.

[0075] For example, when a dog bite is detected, the danger analysis system controls the rescue vehicle to carry an odor release device and an ultrasonic device, and releases the dog's disgusting odor and ultrasound in the direction of the dog to achieve the effect of driving away the dog. Furthermore, odors such as citrus, vinegar, mint, etc. Ultrasonic waves, such as portable ultrasonic dog repellers, interfere with the dog's auditory system by emitting ultrasonic waves of a specific frequency, making it feel uncomfortable, thereby achieving the purpose of driving away. Furthermore, the odor release device can use existing car-mounted or home aromatherapy machines, such as CEENIU car-mounted aromatherapy car perfume in-car smart fragrance machine [AI smart top configuration], etc., and other odor-emitting devices can also be used. These aromatherapy machines can be electrically connected to the controller of the rescue vehicle, and these aromatherapy machines can be controlled. The specific connection method includes wired connection or wireless connection such as Bluetooth.

[0076] Furthermore, the meteorological detector includes a temperature detector, a humidity detector, an air pressure detector, a wind speed and direction detector, and a precipitation detector, which are used to detect the meteorological conditions at various locations in the tourist area.

[0077] In a specific implementation, the rescue drone is provided with a delivery device for delivering rescue items.

[0078] In a specific implementation, the delivery device includes a medicine delivery device, which is used for timely delivery of medicines.

[0079] Furthermore, the structure of the medicine delivery device includes a mounting rack, a medicine storage box, a delivery mechanism, and a control system.

[0080] The mounting bracket is used to firmly mount the delivery device under the drone body to ensure the stability and safety of the device during flight, while ensuring that it matches the center of gravity of the drone and does not affect the flight control performance.

[0081] The medicine storage box is made of lightweight, corrosion-resistant and well-sealed materials, and is used to store medicines to be delivered and is installed on the mounting frame. Its capacity is determined by the load capacity and mission requirements of the drone, generally ranging from a few liters to dozens of liters, and its shape design must conform to aerodynamic principles to reduce flight resistance.

[0082] The delivery mechanism is one of the core components, and the common ones include electromagnetic delivery mechanism, mechanical opening and closing delivery mechanism, etc. The electromagnetic delivery mechanism instantly opens the delivery port at the bottom of the medicine storage box through the action of electromagnetic force, realizing the rapid delivery of medicines; the mechanical opening and closing delivery mechanism uses a motor to drive the mechanical structure, such as gears and connecting rods, to control the opening and closing of the delivery door, and the delivery action is relatively stable.

[0083] The control system is connected to the flight control system of the drone, receives command signals from the ground control station, and accurately controls the time, quantity, location and other parameters of the delivery. It can accurately trigger the delivery mechanism to perform the delivery action when the drone reaches the designated delivery point based on the preset flight path, target area coordinates and drug delivery strategy.

[0084] Similarly, other existing delivery devices that can perform the delivery function and be mounted and controlled by a drone can also be used.

[0085] In a specific implementation, the launching device includes a net launching device, which is used to provide timely rescue to people or animals who fall into the water.

[0086] Because the current rescue methods for falling into water:

[0087] One is that rescue workers throw life buoys and extend bamboo poles on the shore. This method requires being close to the shore. If you are a little far from the shore, you cannot reach the people who fall into the water. Also, because the people who fall into the water are in a panic, they often cannot grab the life buoys or bamboo poles, so this method is not very effective.

[0088] The second is that the rescuers swim over to save the person who falls into the water, but because the person who falls into the water wants to survive, he will grab the rescuers and push them down, so this method is very risky for the rescuers.

[0089] The third method is for rescue workers to use lifeboats for rescue. However, this method requires a long preparation time in nearby waters where there are no lifeboats, and cannot provide timely rescue.

[0090] Therefore, this solution adopts the method of carrying a net-dropping device under the rescue drone, aiming at the person who falls into the water and launching the net so that the net catches the person who falls into the water. Then the rescue drone can lift the person who falls into the water together with the net (this method requires the drone to have a large take-off load; such as the Assault Eagle-2000: can carry 5 passengers or 500 kg of supplies; 2-ton eVTOL M1 series: the designed maximum load reaches 500 kg; GQ-680 unmanned helicopter: the maximum payload reaches 260 kg; GQ-580Z unmanned helicopter: the maximum payload is 250 kg1; GQ-320 unmanned helicopter: the maximum payload is 100 kg). Or the rescue drone can also hand over a cable connected to the net to the people on the shore to pull it to achieve rescue. This method is safer, more effective and faster than the above-mentioned existing methods, especially for some people who have fallen into the water and have not struggled for a long time due to drowning, and the same rescue effect can be achieved. In specific implementation, other existing drone solutions that meet the load requirements can also be used.

[0091] The structure of the net placing device is as follows:

[0092] The drone mounting platform is the part that connects the entire device to the drone. It is made of lightweight and high-strength alloy material to ensure that it can be firmly mounted under the drone. It also has good shock-absorbing performance to reduce the impact of vibrations generated during drone flight on the device, ensuring the stability and safety of the device during flight.

[0093] The net storage and launching mechanism includes a special net cylinder, which stores the folded high-strength net inside and has a launching port at the front end. The net is usually made of high-strength, corrosion-resistant and flexible fiber materials, such as Kevlar fiber, to ensure that it can withstand large impact forces without breaking when capturing targets. The launching mechanism adopts a compressed gas drive or a spring ejection device. The compressed gas can be an inert gas such as high-pressure nitrogen. The ejection force and direction of the net are precisely controlled by controlling the release amount and speed of the gas; the spring ejection device uses a pre-compressed strong spring to quickly eject the net when triggered, and has the characteristics of simple structure and high reliability.

[0094] The target detection and positioning system is composed of a variety of detection equipment such as high-definition cameras, infrared sensors, and radars. High-definition cameras are used to visually identify and track targets during the day or in a well-lit environment, and can provide clear target image information; infrared sensors work at night or in low light conditions, and determine the position and motion state of the target by sensing the infrared radiation emitted by the target; radars can detect targets at long distances, accurately measure parameters such as the distance, speed, and direction of the target, and provide key data support for the precise launch of the net. The data from these detection devices will be transmitted in real time to the flight control system of the drone and the control system of the net delivery device for rapid data processing and decision-making.

[0095] The control system is the core control unit of the entire device, which is mainly composed of a microprocessor, a signal receiver, a power module, etc. The microprocessor is responsible for receiving and processing various data information from the UAV flight control system and the target detection and positioning system. According to the preset program and algorithm, it accurately calculates the best launch time and angle of the net and sends control instructions to the launch mechanism. The signal receiver is used to receive operating instructions from the ground control station or the UAV remote control to achieve remote control of the device. The power module provides a stable power supply for the entire control system. It usually uses a rechargeable lithium battery pack and has power monitoring and low power alarm functions to ensure the normal operation of the device during the mission execution.

[0096] In a specific implementation, the net placing device may also adopt other existing placing devices having the same function of placing a net.

[0097] Embodiment 2

[0098] like Figure 2 The figure shows a flow chart of a control method of an automatic rescue system in a tourist area, which is used to control the automatic rescue system in the tourist area described above, and includes the following steps:

[0099] S1: The detection camera collects the tourist images in real time and feeds them back to the danger analysis system. It includes the following steps:

[0100] Adjust image parameters: Adjust the image parameters of the camera according to the on-site lighting conditions and collection requirements. Including brightness, contrast, saturation, hue, etc. In a dark indoor environment, appropriately increasing the brightness and contrast can make the tourist image clearer; in a colorful outdoor environment, reasonably adjusting the saturation and hue can make the image more vivid. At the same time, you can also set the white balance mode so that the camera can accurately restore the skin color and clothing color of tourists under different color temperatures.

[0101] Set the frame rate and resolution: The frame rate determines the number of image frames collected per second. It is generally set to 25-30 frames per second. Such a frame rate can smoothly record the movements of tourists without obvious lags. The resolution is set according to actual needs and storage capacity, such as the HD or 4K resolution mentioned above. Higher resolutions can provide clearer image details, but they will also take up more storage space and network bandwidth.

[0102] Activate the auto focus and auto exposure functions: The auto focus function allows the camera to automatically adjust the focal length according to the distance of the visitor, ensuring that the visitor's face and body are always clear; the auto exposure function can automatically adjust the aperture size and shutter speed according to the light intensity to ensure uniform brightness of the image and avoid areas that are too bright or too dark. These automatic functions can effectively ensure the quality of image acquisition when visitors enter and exit areas with different light intensities (such as walking from indoors to outdoors with strong sunlight).

[0103] S2: The danger analysis system performs danger analysis on the tourist image and determines whether the tourist is in danger. The specific analysis methods include the following:

[0104] 1. Movement posture analysis.

[0105] Fall Detection:

[0106] Method based on human key points: Using human posture estimation technology, key points of tourists' bodies, such as head, shoulders, elbows, wrists, hips, knees and ankles, are detected. When the position relationship and movement trajectory of these key points suddenly change in a way that is inconsistent with normal walking or standing, for example, the height of the hip drops sharply or the distance between the head and the ground decreases rapidly, it may mean that the tourist has fallen.

[0107] Action template matching: Create a template for falling actions, and judge by comparing the similarity between the tourists' continuous actions and the template. Action features can be extracted from a large number of falling video data, including the tilt angle of the body, the swinging method of the limbs, etc., as a standard template for falling. When the tourist's action matches this template to a high degree, it is judged that he may fall.

[0108] Abnormal action recognition:

[0109] Climbing or crossing behavior: Observe whether visitors climb railings, walls, or cross prohibited areas. This is determined by analyzing the spatial relationship between the visitor's body and obstacles and body movements. For example, if the visitor's hand touches the top of the railing and the leg moves upward, it may be a dangerous climbing behavior.

[0110] Running or collision: In crowded areas, running or collision may cause dangers such as trampling. The speed and direction of movement of tourists are detected to determine whether there is running behavior. If the speed of tourists exceeds a certain threshold of normal walking speed (such as more than 1.5m / s) and the direction is toward the crowd or obstacles, there may be danger.

[0111] 2. Analysis of the correlation between location and environment.

[0112] Dangerous area determination:

[0113] Geographic information marking: Dangerous areas are marked on the scenic area map in advance, such as cliffs, rivers, construction areas, etc. The location of tourists is determined through information such as landmark buildings and road signs in the image. When tourists are in or close to dangerous areas, their actions (such as approaching the edge, trying to enter the construction area) are combined to determine whether they are in danger.

[0114] Image-based identification of dangerous areas: Using computer vision technology, the characteristics of dangerous areas can be identified directly from images. For example, by analyzing the texture and color in the image, water areas can be identified. When tourists are close to water areas and do not have normal water sports equipment, they may be in a dangerous state.

[0115] Abnormal interactions with people around:

[0116] Stampede analysis: In crowded places, calculate the crowd density and crowd flow direction. If the crowd density exceeds a certain safety threshold (for example, more than 5 people per square meter) and the crowd flow is chaotic, such as flowing in opposite directions or local stagnation of people, a stampede may occur. Observe whether tourists are squeezed or pushed down by the crowd to determine whether they are in danger.

[0117] Conflict behavior recognition: Analyze whether there are intense physical movements between tourists and the surrounding people, such as pushing, fighting, etc. By detecting the contact between human bodies, movement amplitude and facial expressions, it is comprehensively judged whether there is conflict and danger.

[0118] 3. Observe facial expressions and body conditions.

[0119] Expression analysis:

[0120] Fear or pain expression recognition: Using facial expression recognition algorithms, we analyze the muscle movements of tourists' faces, such as frowning eyebrows, wide eyes, and open mouths in a state of fear. When these expressions are detected, we combine their movements and positions to determine whether they are caused by fear or pain due to danger.

[0121] Judging by tired or uncomfortable expressions: Observe whether the tourists have tired expressions, such as squinting eyes, drooping heads, etc. In hot weather or after walking for a long time, these expressions may mean that the tourists are unwell and are in danger of heat stroke or physical exhaustion.

[0122] Physical status monitoring:

[0123] Combined with wearable devices: If tourists wear wearable devices such as smart bracelets, the devices can be used to obtain tourists' heart rate, blood pressure and other physiological data. When the heart rate rises abnormally (such as exceeding 20% ​​of the normal range) or the blood pressure fluctuates abnormally, and the tourists' expressions and movements also show discomfort, it is judged that they may be in danger.

[0124] Thermal imaging technology assistance: Use thermal imaging cameras to observe the heat distribution of tourists' bodies. If the temperature of a certain part of the body is found to be too high (such as local fever, which may be caused by injury or inflammation) or too low (which may be caused by hypothermia), other information will be combined to determine whether it is in a dangerous state.

[0125] 4. Object interaction analysis.

[0126] Hazardous Object Contact:

[0127] Sharp or dangerous tool identification: Identify from the image whether the tourist is touching or holding dangerous objects, such as knives, sharp tools, etc. Object recognition technology is used to detect the shape, color and other features of objects in or around the tourist's hands to make judgments. When a tourist holds an object that may cause harm and moves abnormally, there may be danger.

[0128] Exposure to hazardous substances: Identify whether tourists have come into contact with potentially harmful substances, such as chemical leaks, equipment with electrical leakage, etc. By observing the tourists' contact with specific hazardous areas or objects, combined with environmental signs and actions, determine whether they are in danger.

[0129] Hazards from falling or lost items:

[0130] Loss of important items affects judgment: If tourists lose important items, such as medicines (for tourists with chronic diseases), communication equipment, etc., and behave in a panic during the search process, they may ignore the dangers of the surrounding environment and increase the probability of getting into distress.

[0131] Danger caused by dropped items: When items on a tourist's body fall, such as a backpack falling off while walking on a mountain, it may cause the tourist to lose balance and fall, or enter a dangerous area when picking up the items. Determine whether it will cause danger by observing the movement and position relationship between the tourist and the dropped items.

[0132] S3: If it is determined that the tourist is in distress, further analyze the type of distress the tourist is in. Further, the types of distress include:

[0133] 1. Types of distress related to the natural environment.

[0134] Falls and falls: In mountainous areas, stairs, slopes and other areas with complex terrain, tourists may fall due to slippery, rough roads or lack of concentration. For example, during mountain climbing, tourists may lose their balance and fall due to stepping on loose rocks; on the edge of a high place such as an observation deck, tourists may fall due to being too close to the edge or being crowded by the crowd.

[0135] Getting lost and trapped: In natural environments such as forests, mountains, and caves, tourists may get lost due to lack of correct route guidance or deviation from the established route. Once lost, they may be trapped in these areas and face risks such as lack of food and water, and being attacked by wild animals. For example, in the deep mountains without obvious road signs, tourists can easily lose their way and may get into trouble as time goes by.

[0136] Heat stroke: In hot weather, tourists who are exposed to the sun for a long time are prone to heat stroke symptoms. Especially in open scenic areas without shade facilities, such as deserts and beaches, tourists may feel dizzy, nauseous, weak, or even coma due to dehydration and electrolyte imbalance.

[0137] Low temperature frostbite: In low temperature environments such as high altitude mountainous areas and cold winter scenic spots, if tourists do not have adequate warming measures, exposed parts of the body such as fingers, toes, ears, etc. are easily frostbitten. Being in a low temperature environment for a long time may also lead to hypothermia, a serious condition that can cause the core temperature of the human body to drop and endanger life.

[0138] Heavy rain and floods: In heavy rain, low-lying scenic areas may be flooded. If tourists are not evacuated in time, they may be washed away by the flood or trapped in flooded buildings. In addition, heavy rain may also trigger geological disasters such as landslides and mud-rock flows, which may put tourists at risk of being buried.

[0139] Lightning strikes: In open fields or on mountaintops, tourists are easily targeted by lightning strikes during thunderstorms. Lightning may directly strike tourists, causing burns, cardiac arrest and other serious consequences; even if it does not directly strike, the induced current generated by lightning may also cause harm to tourists.

[0140] Fire: For example, there is a fire near tourists.

[0141] Types of animal damage.

[0142] For example, bites from large dogs, snake bites, etc.

[0143] 3. Types of distress caused by human factors.

[0144] Crowd stampede: During peak tourist seasons or large-scale events, stampede incidents may occur in crowded places within scenic areas, such as entrances, popular attractions, and performance venues. When the flow of the crowd is out of control, the people in front are pushed down, and the people behind continue to crowd forward, tourists may be trampled and injured, or even their lives may be in danger.

[0145] Conflicts and violence: Conflicts may occur between tourists and other tourists, scenic area staff or local residents for various reasons. For example, an argument over queue-jumping may escalate into a physical conflict, or some behavior of tourists may cause dissatisfaction among local residents, leading to violence and causing injuries to tourists.

[0146] Theft and robbery: Theft or robbery may occur in some scenic spots with poor security or when tourists are careless. Tourists may have their belongings stolen and may suffer physical injuries during resistance. For example, tourists may be targeted and robbed by criminals in remote corners of scenic spots.

[0147] 4. Types of distress related to the tourist’s own health.

[0148] Sudden illness: Tourists may suffer from some chronic diseases, such as heart disease, diabetes, asthma, etc., which may be triggered by fatigue, environmental changes and other factors during travel. For example, tourists with heart disease may suffer from sudden heart attacks due to excessive physical exertion during mountain climbing; asthma patients may suffer from asthma attacks due to allergies in scenic spots with more pollen.

[0149] Physical injuries: When tourists participate in some dangerous activities, such as rock climbing, horse riding, boating, etc., they may suffer physical injuries due to accidents. For example, they may fall off the rock without holding on to it during rock climbing, or they may fall off the horse and suffer a fracture.

[0150] S4: Control the corresponding rescue vehicle or rescue drone to rescue the tourists according to the type of distress they are in. Specifically, the following steps are included:

[0151] Falling and Stumbling:

[0152] Rescue cart: If a tourist falls or trips in relatively accessible terrain (such as near a scenic trail), the rescue cart can quickly drive along the road to the accident site. The car is equipped with a stretcher, a first aid kit (including hemostatic drugs, bandages, etc.), communication equipment, etc. When arriving at the scene, rescuers can use a stretcher to lift the injured tourist onto the cart and quickly send him to a medical point. For example, on a winding mountain road in a mountainous area, a rescue cart can quickly reach the location of the injured tourist.

[0153] Rescue drones: Rescue drones can be useful in areas that are difficult to reach by vehicle (such as the bottom of a valley, steep slopes). Drones can carry first aid kits (including simple dressing supplies and medicines) and positioning beacons. They first fly above the tourists and drop first aid kits. At the same time, they use positioning beacons to determine the location of the tourists and guide ground rescuers to the area. In some cliffs, rescue drones can quickly reach the location of injured tourists, buying time for subsequent rescue.

[0154] Lost and Trapped:

[0155] Rescue vehicle: At the edge of the forest or in the scenic area where there are roads connecting lost areas, the rescue vehicle can follow the road as close as possible to where tourists may appear. The vehicle is equipped with a high-power loudspeaker to call out to the surrounding area to find tourists; it can also carry supplies such as food, water, and warm clothing. By combining with the positioning of tourists' mobile phones or the positioning system in the scenic area, the search range can be narrowed and supplies can be provided to lost tourists.

[0156] Rescue drones: For lost tourists deep in the forest or mountains, rescue drones can use thermal imaging technology to search over large areas. Once a tourist is found, rescue packages with positioning devices and simple survival supplies (such as signal sticks, waterproof cloth, and high-calorie food) can be dropped. The drone's communication system can also send the tourist's location and rescue information to calm the tourist's emotions.

[0157] Heat stroke:

[0158] Rescue vehicles: In scenic areas with high temperatures, rescue vehicles should be equipped with sunshade facilities, air conditioning systems, and plenty of drinking water, electrolyte drinks, and anti-heat medicines (such as Huoxiang Zhengqi water). Upon receiving a call for help from a tourist suffering from heat stroke, the rescue vehicle can quickly rush to the scene, transfer the tourist to a cool place in the vehicle, replenish water and medicine for the tourist, and then send the tourist to a medical point for further observation and treatment.

[0159] Rescue drones: Rescue drones can carry small heatstroke prevention and cooling supplies, such as ice packs and small spray cooling devices. After discovering tourists suffering from heatstroke, these supplies will be dropped near the tourists, and tourists will be informed of how to use them through communication equipment, waiting for further rescue by rescue vehicles.

[0160] Low temperature frostbite:

[0161] Rescue vehicle: The rescue vehicle should be equipped with sufficient warming supplies, such as thick blankets, warm clothes, hand warmers, etc. When a tourist with frostbite is found, the rescuers can wrap the tourist with a warm blanket, place the hand warmer on the key parts of the tourist's body (such as the armpits and groin), and send the tourist to a warm indoor environment for treatment as soon as possible.

[0162] Rescue drones: In some cold areas with inconvenient transportation, rescue drones can carry lightweight warm items, such as emergency thermal blankets. After the drone flies over the tourists, it will drop the thermal blankets and notify the ground rescuers of the tourists' location, allowing them to transfer the tourists to a safe and warm place as soon as possible.

[0163] Heavy rain and flooding:

[0164] Rescue vehicle: In the vicinity of scenic spots where floods may occur, the rescue vehicle should have a high chassis and good waterproof performance. The vehicle should be equipped with rescue equipment such as life jackets, lifebuoys, ropes, etc. When tourists are found trapped by floods, the rescue vehicle can stop in a safe place and throw life-saving equipment or use ropes to pull tourists onto the vehicle.

[0165] Rescue drones: Rescue drones can conduct reconnaissance above flooded areas and drop life jackets and positioning devices to trapped tourists. They can also use their communication capabilities to keep in touch with tourists, guide them to take correct self-rescue measures, and provide accurate location information for ground rescue teams.

[0166] Lightning strike:

[0167] Rescue vehicle: In thunderstorm weather, the rescue vehicle must ensure its own lightning protection safety. The vehicle is equipped with first aid equipment such as an insulating stretcher and a defibrillator. When a tourist struck by lightning is found, the rescue personnel must wear insulating equipment, use a stretcher to quickly transfer the tourist to the vehicle, perform cardiopulmonary resuscitation and other first aid measures, and then send the tourist to the hospital.

[0168] Rescue drone: After the thunder and lightning weather eases a bit, the rescue drone can first fly to the location of the tourists struck by lightning to check, drop first aid kits with insulation marks, and guide subsequent rescue vehicles to the location.

[0169] Crowds and stampedes:

[0170] Rescue vehicle: In crowded scenic spots, the rescue vehicle should be able to quickly shuttle through the crowd. The vehicle is equipped with loudspeakers, first aid kits, stretchers and other equipment. When a stampede occurs, rescuers can use the loudspeaker to direct the crowd to evacuate, and use stretchers to carry injured tourists onto the rescue vehicle and send them to medical centers for treatment.

[0171] Rescue drones: Rescue drones can fly above the crowd and use their cameras to observe the flow of the crowd and the location of injured tourists. They can provide information to ground rescuers through communication equipment and guide them to quickly reach the injured tourists. Drones can also drop some small first aid supplies, such as hemostatic bandages, disinfectants, etc., to provide preliminary treatment for injured tourists.

[0172] Conflict and violence:

[0173] Rescue vehicle: When receiving a call for conflict and violence, the rescue vehicle will rush to the scene. The vehicle is equipped with protective equipment (such as riot shields, helmets) and restraint tools (such as handcuffs). After the rescuers put on their protective equipment, they will stop the conflicting parties, give the injured tourists a simple bandage, and then take the injured tourists and related personnel back to a safe area for further treatment.

[0174] Rescue drones: Rescue drones can first arrive above the conflict scene and use their cameras to record the conflict process to provide evidence for subsequent handling. At the same time, drones can use communication equipment to send warnings to both parties in the conflict, reminding them to stop violent behavior and wait for the arrival of rescuers.

[0175] Theft and robbery:

[0176] Rescue vehicle: After receiving a theft or robbery alarm, the rescue vehicle will rush to the scene. The vehicle is equipped with communication equipment, tracking equipment and first aid kit. Rescuers can use the tracking equipment to find the target object, provide first aid to the injured tourists, and send them to a safe area.

[0177] Rescue drones: Rescue drones can use their aerial advantage to quickly search for the direction in which the target object escapes and provide tracking clues for the rescue vehicle. Drones can also carry lighting equipment to help rescuers do their work better at night or in dimly lit areas.

[0178] Sudden illness:

[0179] Rescue vehicle: The rescue vehicle should be stocked with medicines and equipment for common sudden illnesses, such as emergency heart disease medicine, asthma spray, blood glucose meter, etc. When receiving a tourist's emergency message, the rescue vehicle can quickly arrive at the scene, provide the corresponding first aid medicines and equipment according to the symptoms of the tourist, and then send the tourist to the hospital.

[0180] Rescue drones: Rescue drones can carry some simple first aid medicines and positioning devices. When a tourist with a sudden illness is found, the rescue drones will first deliver medicines and use the positioning device to guide the rescue vehicle to come. During the waiting process, the rescue drones will use communication equipment to guide the tourists or their companions to use the medicines correctly.

[0181] Physical Injury:

[0182] Rescue vehicles: Rescue vehicles should be on standby at all times near scenic spots and activity areas where physical injuries are likely to occur (such as ski resorts and rock climbing sites). The vehicles are equipped with first aid equipment such as stretchers, splints, tourniquets, etc. of various specifications. When tourists are injured, the rescue vehicles can quickly arrive at the scene, bandage the wounds, fix the fractures, and then send the tourists to the medical point.

[0183] Rescue drones: Rescue drones can carry lightweight first aid supplies, such as small bandages, disinfectant wipes, etc. After discovering injured tourists, they will drop first aid supplies and inform tourists through communication equipment how to perform simple self-treatment and wait for further rescue by the rescue vehicle.

[0184] The specific implementation also includes the following steps:

[0185] S5: The detection camera collects images of the water area in the tourist area in real time and feeds them back to the danger analysis system.

[0186] S6: The danger analysis system analyzes the water area image to determine whether there are people or animals falling into the water.

[0187] In a specific implementation, when analyzing the water area image to determine whether a person or animal has fallen into the water, the water splash feature appearing in the water area image is used as a basis for determining whether a person or animal has fallen into the water. Further, the water splash feature detection in the water area image includes the following methods:

[0188] 1. Analysis method based on pixel changes

[0189] Inter-frame difference method:

[0190] Principle: This method detects moving targets by comparing the pixel differences between consecutive video frames (or consecutive images). For water images, when water splashes appear, the pixel values ​​in the water splash area will change significantly between adjacent frames. For example, assuming that the current frame is F(t) and the previous frame is F(t-1), by calculating the difference between the corresponding pixel positions of the two frames (such as the absolute value difference), the differential image D(t) = |F(t)-F(t-1)| is obtained. If the pixel value of a certain area in the differential image exceeds the set threshold, it may be the area where the water splash appears.

[0191] Threshold determination: Determination of the threshold is the key. An initial threshold can be determined by statistically analyzing normal water images without splashes. For example, a video frame of a calm water area is differentially calculated, the distribution of pixel differences is statistically analyzed, and a suitable threshold is set based on parameters such as the peak value and standard deviation of the distribution. Generally speaking, if the pixel difference is greater than the average difference plus several times the standard deviation (such as 3 times), it can be considered a possible splash area.

[0192] Background subtraction algorithm:

[0193] Background modeling: First, a water background model needs to be established. For relatively static water scenes, the background image B can be obtained by averaging or median filtering the images without water splashes for a period of time. When a new image frame F arrives, the difference image D = |FB| is calculated. This difference will be larger where water splashes appear.

[0194] Update the background model: The background of the water area may change due to factors such as changes in lighting and slow changes in water flow. Therefore, the background model needs to be updated regularly. Adaptive methods can be used, such as updating based on the frequency of pixel changes. For pixel positions that change frequently (which may be due to normal conditions such as slight fluctuations in the water surface), the corresponding pixel values ​​in the background model can be updated more quickly.

[0195] 2. Use image texture feature analysis

[0196] Texture feature extraction method:

[0197] Gray Level Co-occurrence Matrix (GLCM): describes texture by calculating the joint distribution of gray levels of pixel pairs at a certain distance and direction in the image. For water images, the texture of calm water is relatively uniform, while the texture of the splash area is more complex. The characteristic parameters of GLCM such as contrast, entropy, and correlation can be calculated. For example, when splashes appear, the contrast and entropy usually increase because the gray level changes in the splash area are more drastic and the correlation between pixels decreases.

[0198] Local Binary Pattern (LBP): LBP is an effective method to describe local texture. It compares the central pixel with its surrounding pixels and generates a binary code based on the comparison result to represent the local texture. In water images, the LBP code of calm water surface is relatively simple, while the splash area will have more diverse coding patterns. The frequency of occurrence of different LBP codes can be counted to determine whether there is splash.

[0199] Classifier identifies water splash texture:

[0200] Training classifiers: Collect water image samples with and without splashes, extract the above texture features as training data, and use classifiers such as support vector machines (SVMs) and artificial neural networks (such as simple multi-layer perceptrons) for training. For example, take the texture feature vector as input, mark samples with splashes as positive classes, and samples without splashes as negative classes, and let the classifier learn how to distinguish these two types of samples through training.

[0201] Classification judgment: For new water images, extract texture features and input them into the trained classifier. The classifier outputs a probability value or category label to determine whether there is water splash. If the probability value exceeds a certain threshold (such as 0.5) or the category label is water splash, it can be determined that water splash appears in the image.

[0202] 3. Methods based on optical properties and shape features

[0203] Analysis of light reflection and refraction:

[0204] Water splashes reflect light: Water splashes produce a unique reflective effect when exposed to sunlight or other light sources. Compared with calm water, the reflection in the water splash area is brighter and unevenly distributed. Water splashes can be detected by analyzing the brightness peaks and distribution in the image. For example, using the brightness histogram of the image, when water splashes appear, a new peak will appear at the high end (brighter part) of the histogram or the distribution will become wider.

[0205] Distortion caused by refraction: The formation of splashes causes light to refract at the water surface, causing the visual shape of underwater objects to change. Splashes can be detected by observing the shape distortion of underwater objects. For example, if you see irregular distortions in the outlines of underwater rocks or plants in clear water, it may be because of splashes on the water surface.

[0206] Shape Recognition:

[0207] Connected Region Analysis: After binarizing the water image (for example, converting the image to only black and white pixels, representing water and non-water areas, based on brightness or other features), a connected region analysis algorithm is used to detect possible splash shapes. Splashes usually have irregular shapes and are connected by multiple small areas. The geometric features of the connected regions, such as area, perimeter, and circularity, can be calculated. Compared with the large connected regions of calm water, the connected regions of splashes are smaller in area, relatively longer in perimeter, and less circular.

[0208] Model-based shape matching: Build a splash shape model, such as a combination of basic shapes such as ellipses and irregular polygons. Use template matching or feature point-based shape matching methods to match the model with suspicious areas in the image. If the matching degree exceeds a certain level, it is determined that a splash has appeared. For example, use Hough transform to detect possible circular splash parts, and then combine other shape features for comprehensive judgment.

[0209] S7: If it is determined that a person or an animal has fallen into the water, the rescue drone is controlled to carry a fishing net dropping device to drop a fishing net for rescue.

[0210] In a specific implementation, the control of the rescue drone carrying the net-casting device to cast a net for fishing for rescue includes the following steps:

[0211] Target search and positioning: After the drone takes off with the net-dropping device, it flies to the designated area according to the mission plan. During the flight, the target detection and positioning system continues to work, using cameras, infrared sensors, radars and other equipment to scan the surrounding environment in all directions to search for target objects. Once a splash area is found, the system will immediately lock the target and track the target's position, speed, direction of movement and other information in real time, and transmit this data to the control system.

[0212] Data processing and decision-making: After receiving the relevant data of the target, the control system quickly processes and analyzes the data. Combining the current position, flight attitude, speed and dynamic information of the drone, the built-in algorithm calculates the optimal launch parameters of the net, including the launch angle, jet force and launch timing. At the same time, the control system will also communicate with the drone's flight control system in real time to coordinate the drone's flight status and ensure that the drone can maintain a stable flight attitude when launching the net to increase the capture success rate.

[0213] Net launching and capturing: When the splash area enters the effective capture range of the net, the control system sends a launch command to the net storage and launching mechanism according to the pre-calculated launch parameters. If it is a launch mechanism driven by compressed gas, the high-pressure gas is released quickly, pushing the net out of the launch port at the front end of the net tube at high speed; if it is a spring ejection device, the spring is triggered to release elastic potential energy, and the net is ejected forcefully. The net will quickly unfold during flight to form a larger capture area, using its own weight and the tension of the net, as well as the kinetic energy generated by the ejection, to entangle and capture people who fall into the water.

[0214] Mission completion and recovery. If the capture is successful, the drone can rise according to the command to pick up the person who fell into the water and return to the designated location to land, and the ground personnel will carry out follow-up rescue operations.

[0215] The specific implementation also includes the following steps:

[0216] Preset marks are placed on the cruise ships in the water area so that the hazard analysis system can exclude the marked cruise ships when conducting hazard analysis to avoid mistaking the splashes caused by the cruise ships for people or animals falling into the water.

[0217] In specific implementation, the danger analysis system collects detection data from meteorological detectors and analyzes the meteorological conditions at each location in the tourist area. According to the meteorological conditions and geographical location conditions, the rescue vehicles and rescue drones are equipped with corresponding rescue items to improve the utilization rate of items and the rescue effect. The rescue items include the rescue items required for the above-mentioned various types of dangers, such as first aid kits, positioning beacons, ice packs, blankets, warm clothes, hand warmers, umbrellas, raincoats, medicine bottles, dog repellents, mosquito repellents, etc.

[0218] The present application provides an automatic rescue system and control method for a tourist area, including a rescue vehicle, a rescue drone, a detection camera and a danger analysis system; the detection camera is arranged on a pillar in the tourist area and connected to the danger analysis system, and a meteorological detector connected to the danger analysis system is arranged on the pillar, which is used to collect images and weather data of tourists and feed them back to the danger analysis system, and the danger analysis system analyzes the danger of tourists based on the collected data; the danger analysis system is connected to the rescue vehicle and the rescue drone respectively, so as to plan the navigation route according to the location of the rescue vehicle, the rescue drone and the location of tourists, and control the rescue vehicle and the rescue drone to rescue tourists in distress. This solution can automatically detect whether tourists in various areas are in danger through cameras along the way, and can timely warn when danger occurs and timely rescue through the rescue vehicle and the rescue drone, solving the problem of not being able to find and rescue in time.

[0219] Those of ordinary skill in the art will appreciate that the units of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition of each example has been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0220] In the embodiments provided by the present invention, it should be understood that the division of units is only a logical function division, and there may be other division methods in actual implementation, for example, multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored, etc.

[0221] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0222] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-0nlyMemory), random access memory (RAM, RandomAccessMemory), mobile hard disk, magnetic disk or optical disk, etc., which can store program code.

[0223] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. An automatic rescue system for tourist areas, characterized in that: It includes a rescue vehicle, a rescue drone, a detection camera and a danger analysis system; the detection camera is arranged on a pillar in the tourist area and connected to the danger analysis system; a meteorological detector connected to the danger analysis system is arranged on the pillar for collecting images and meteorological data of tourists and feeding them back to the danger analysis system; the danger analysis system performs danger analysis on tourists according to the collected data; the danger analysis system is connected to the rescue vehicle and the rescue drone respectively to control the rescue vehicle and the rescue drone to go to the tourists in distress for rescue.

2. The automatic rescue system for tourist areas according to claim 1, characterized in that: The rescue vehicle comprises a wheeled moving structure and a track-type moving structure.

3. The automatic rescue system for tourist areas according to claim 1, characterized in that: The rescue vehicle is provided with an odor releasing device or an ultrasonic device; the odor releasing device and the ultrasonic device are respectively connected to a controller of the rescue vehicle to release odor or ultrasonic waves to drive away animals.

4. The automatic rescue system for tourist areas according to claim 1, characterized in that: The rescue drone is provided with a delivery device for delivering rescue items.

5. The automatic rescue system for tourist areas according to claim 4, characterized in that: The delivery device includes a medicine delivery device, which is used for delivering medicines in a timely manner.

6. The automatic rescue system for tourist areas according to claim 4, characterized in that: The launching device comprises a net launching device, which is used for timely rescue of people or animals falling into water.

7. A control method for an automatic rescue system in a tourist area, characterized in that: An automatic rescue system for controlling a tourist area according to any one of claims 1 to 6, comprising the following steps: The detection camera collects images of tourists in real time and feeds them back to the danger analysis system; The danger analysis system performs danger analysis on the tourist images and determines whether the tourist is in danger; If it is determined that the tourist is in distress, the type of distress the tourist is in is further analyzed; According to the type of distress the tourists are in, control the corresponding rescue vehicle or rescue drone to rescue the tourists.

8. The control method of the automatic rescue system in a tourist area according to claim 7, characterized in that: The following steps are also included: The detection camera collects images of the water area in the tourist area in real time and feeds them back to the danger analysis system; The danger analysis system analyzes the water area image to determine whether there are people or animals falling into the water; If it is determined that a person or animal has fallen into the water, the rescue drone carrying the net-dropping device is controlled to drop a net to catch fish for rescue.

9. The control method of the automatic rescue system in a tourist area according to claim 8, characterized in that: The following steps are also included: Preset marks are placed on cruise ships in the water area so that the hazard analysis system can exclude the marked cruise ships when conducting hazard analysis.

10. The control method of the automatic rescue system in a tourist area according to claim 8, characterized in that: When analyzing the water area image to determine whether a person or an animal has fallen into the water, the water splash feature appearing in the water area image is used as a basis for determining whether a person or an animal has fallen into the water.

Images