Double-light personnel search and rescue positioning visual system on unmanned aerial vehicle

Abstract

The invention relates to the technical field of unmanned aerial vehicles, in particular to a dual-light personnel search and rescue positioning visual system on an unmanned aerial vehicle, which comprises an unmanned aerial vehicle body, a shooting camera, a lifting device, a lifting seat, a rotating motor, a mounting seat, an infrared thermal imaging camera, a storage battery, two driving motors, two driving wheels, a plurality of driven wheels and two tracks, the shooting camera shoots specific visual information through visible light to guide rescue; the lifting device is used for downwards putting out the lifting base, the rotating motor is used for adjusting the angle of the infrared thermal imaging camera, the infrared thermal imaging camera assists in searching for a person to be rescued, and the two driving motors are used for driving the crawler belt to rotate and providing the moving capacity; when the unmanned aerial vehicle moves to the terrain with many obstacles, the infrared thermal imaging camera is driven by the driving motor to move, so that the shooting blind area is effectively reduced, the detection range is improved, and the problems that an existing unmanned aerial vehicle is high in cost and poor in detection range increasing effect are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicles, and in particular to a dual-light personnel search and rescue positioning vision system on an unmanned aerial vehicle. Background Art

[0002] Drones play a vital role in personnel search and rescue, and can improve rescue efficiency and success rate. Modern drone technology not only makes search and rescue operations more convenient, accurate and safe, but also demonstrates its unique advantages in many aspects. It is these advantages that make drones an indispensable part of emergency rescue; existing drones usually only have one infrared thermal imaging camera with a limited detection range. To increase the detection range, it is usually necessary to use infrared thermal imaging cameras with higher technical indicators such as imaging accuracy, which has a high development cost and is limited by the degree of technological development.

[0003] The existing patent discloses an unmanned aerial vehicle (CN212766740U) with a personnel search and rescue positioning device, which includes a body and a personnel search and rescue positioning device. The personnel search and rescue positioning device is fixed on the body. The personnel search and rescue positioning device includes five infrared thermal imaging cameras, and the five infrared thermal imaging cameras are all arranged at the bottom of the body. The unmanned aerial vehicle increases the detection range and improves the detection accuracy by arranging five infrared thermal imaging cameras.

[0004] However, this type of drone only expands its detection range by stacking infrared thermal imaging cameras. The unit price of infrared thermal imaging cameras is relatively high, and increasing the number will still significantly increase the cost. Moreover, simply increasing the number will have a poor effect on increasing the detection range. Summary of the invention

[0005] The purpose of the present invention is to provide a dual-light personnel search and rescue positioning vision system on an unmanned aerial vehicle, aiming to solve the problems of high cost and poor effect of increasing the detection range of existing unmanned aerial vehicles.

[0006] To achieve the above-mentioned purpose, the present invention provides a dual-light personnel search and rescue positioning visual system on an unmanned aerial vehicle, comprising an unmanned aerial vehicle body, a shooting camera, a lifting device, a lifting seat, a rotating motor, a mounting seat, an infrared thermal imaging camera, a battery, two driving motors, two driving wheels, a plurality of driven wheels and two tracks, wherein the shooting camera is arranged at the front end of the unmanned aerial vehicle body; the lifting device is arranged below the unmanned aerial vehicle body; the lifting seat is arranged below the lifting device; the rotating motor is fixed in the lifting seat; the mounting seat is fixed at the output end of the rotating motor; the infrared thermal imaging camera is arranged at the front end of the mounting seat; the battery is arranged in the lifting seat; the two driving motors are arranged on both sides of the mounting seat; the two driving wheels are fixed at the output ends of the two driving motors; the plurality of driven wheels are rotatably arranged on both sides of the mounting seat; the two tracks are respectively matched with the two driving wheels and matched with the plurality of driven wheels.

[0007] Among them, the lifting device includes a winding motor, a winding roller and a suspension rope, the winding motor is fixed under the drone body; the winding roller is fixed to the output end of the winding motor; the suspension rope is fixedly connected to the winding roller and fixedly connected to the lifting seat.

[0008] Among them, the dual-light personnel search and rescue positioning visual system on the drone also includes multiple magnetic blocks and multiple electromagnets, and the multiple magnetic blocks are fixed on the lifting seat; and the multiple electromagnets are arranged below the drone body.

[0009] Among them, the dual-light personnel search and rescue positioning vision system on the drone also includes an electric push rod and a walking camera, the lifting seat has a storage groove; the electric push rod is arranged in the storage groove of the lifting seat; the walking camera is arranged on the electric push rod.

[0010] Among them, the dual-light personnel search and rescue positioning vision system on the drone also includes a rotating motor, and the rotating motor is fixedly connected to the output end of the electric push rod and is fixedly connected to the walking camera.

[0011] Among them, the dual-light personnel search and rescue positioning visual system on the drone also includes a visual processing module, which is used to identify and locate the target based on the fusion of the visible light image obtained by the shooting camera and the infrared image obtained by the infrared thermal imaging camera.

[0012] Wherein, the visual processing module includes an image collection unit, an image preprocessing unit, a model training unit and a target detection unit;

[0013] The image collection unit is used to obtain a visible light image and an infrared image containing a target object;

[0014] The image processing unit is used to fuse the visible light image and the infrared image into one image using an image fusion method;

[0015] The model training unit is used to train the YOLO model using the labeled fused image dataset;

[0016] The target detection unit is used to perform target detection based on the acquired image using the trained YOLO model.

[0017] Wherein, the image processing unit includes an image registration subunit, an image decomposition subunit, an image fusion subunit and an image reconstruction subunit;

[0018] The image registration subunit is used to perform image registration and alignment on the visible light image and the infrared image by using feature point matching;

[0019] The image decomposition subunit is used to decompose the visible light image and the infrared image into sub-bands of multiple scales;

[0020] The image fusion subunit is used to fuse the low-frequency information and high-frequency information of the visible light image and the infrared image;

[0021] The image reconstruction subunit is used to recombine the fused low-frequency and high-frequency information and perform inverse transformation to obtain a final fused image.

[0022] The present invention discloses a dual-light personnel search and rescue positioning visual system on an unmanned aerial vehicle. The unmanned aerial vehicle body adopts an existing common unmanned aerial vehicle. The shooting camera shoots specific visual information through visible light to guide the rescue. The lifting device is arranged in the unmanned aerial vehicle body and is used to release the lifting seat downward. The rotating motor is arranged below the lifting seat. The rotating motor is used to drive the mounting seat to rotate. The infrared thermal imaging camera is arranged below the mounting seat. The rotating motor is used to adjust the angle of the infrared thermal imaging camera. The infrared thermal imaging camera assists in finding the rescued personnel by means of infrared light. The storage battery is used to power the infrared thermal imaging camera and the drive The two driving motors are powered by a driving motor, and the two driving motors are arranged on both sides of the mounting seat, and are respectively used to drive the driving wheel to rotate, and with the cooperation of the driven wheel, drive the crawler to rotate, thereby providing mobility; the lifting seat moves with the UAV body. When it moves to a terrain with many obstacles such as floors and woods, the lifting seat can be lowered by the lifting device, and then the driving motor provides moving power to drive the infrared thermal imaging camera to move and enter the place that is difficult for the UAV body to reach, which can effectively reduce the shooting blind area, increase the detection range, and has a low cost, thereby solving the problem of high cost of existing UAVs and poor effect of increasing the detection range. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art are briefly introduced below.

[0024] Figure 1 It is a structural schematic diagram of a dual-light personnel search and rescue positioning vision system on an unmanned aerial vehicle according to the first embodiment of the present invention.

[0025] Figure 2 It is a cross-sectional view of a dual-light personnel search and rescue positioning vision system on an unmanned aerial vehicle according to the first embodiment of the present invention.

[0026] Figure 3 The present invention is a dual-light personnel search and rescue positioning visual system for unmanned aerial vehicles in the first embodiment of the present invention. Figure 2 A partial enlargement of detail A.

[0027] Figure 4 It is a structural schematic diagram of a dual-light personnel search and rescue positioning vision system on an unmanned aerial vehicle according to the second embodiment of the present invention.

[0028] Figure 5 It is a structural schematic diagram of a dual-light personnel search and rescue positioning vision system on an unmanned aerial vehicle according to the third embodiment of the present invention.

[0029] 101-UAV body, 102-shooting camera, 103-lifting seat, 104-rotating motor, 105-mounting seat, 106-infrared thermal imaging camera, 107-battery, 108-driving motor, 109-driving wheel, 110-driven wheel, 111-track, 112-winding motor, 113-winding roller, 114-suspension rope, 115-magnetic block, 116-electromagnet, 117-electric push rod, 118-walking camera, 119-rotating motor, 120-storage slot, 201-alarm light, 202-buzzer, 203-signal detector, 301-connecting block, 302-electric lock, 303-broadcaster, 304-lighting lamp. DETAILED DESCRIPTION

[0030] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limiting the present invention.

[0031] For the first embodiment, please refer to Figure 1 to Figure 3 , Figure 1 2 is a schematic structural diagram of a dual-light personnel search and rescue positioning visual system on an unmanned aerial vehicle according to a first embodiment of the present invention; Figure 2 is a cross-sectional view of a dual-light personnel search and rescue positioning visual system on an unmanned aerial vehicle according to a first embodiment of the present invention; Figure 3 The present invention is a dual-light personnel search and rescue positioning visual system for unmanned aerial vehicles in the first embodiment of the present invention. Figure 2 A partial enlargement of detail A.

[0032] The present invention provides a dual-light personnel search and rescue positioning visual system on an unmanned aerial vehicle, comprising an unmanned aerial vehicle body 101, a shooting camera 102, a lifting device, a lifting seat 103, a rotating motor 104, a mounting seat 105, an infrared thermal imaging camera 106, a battery 107, two driving motors 108, two active wheels 109, a plurality of driven wheels 110, two crawlers 111, a plurality of magnetic blocks 115, a plurality of electromagnets 116, an electric push rod 117, a walking camera 118 and a rotating motor 119, the lifting device comprising a winding motor 112, a winding roller 113 and a suspension rope 114, and the lifting seat 103 having a storage groove 120.

[0033] The above-mentioned solution solves the problems of high cost and poor effect of increasing the detection range of existing drones. It can be understood that the above-mentioned solution can drive the infrared thermal imaging camera 106 to move and enter places that are difficult for the drone body 101 to reach, which can effectively reduce shooting blind spots, increase the detection range, and has a low cost.

[0034] In this embodiment, the shooting camera 102 is arranged at the front end of the drone body 101; the lifting device is arranged below the drone body 101; the lifting seat 103 is arranged below the lifting device; the rotating motor 104 is fixed in the lifting seat 103; the mounting seat 105 is fixed to the output end of the rotating motor 104; the infrared thermal imaging camera 106 is arranged at the front end of the mounting seat 105; the storage battery 107 is arranged in the lifting seat 103; the two driving motors 108 are arranged on both sides of the mounting seat 105; the two The driving wheel 109 is fixed to the output end of the two driving motors 108; the plurality of driven wheels 110 are rotatably arranged on both sides of the mounting seat 105; the two crawlers 111 are respectively matched with the two driving wheels 109 and the plurality of driven wheels 110; the drone body 101 adopts an existing common drone, and the shooting camera 102 shoots specific visual information through visible light to guide the rescue; the lifting device is arranged in the drone body 101, and is used to release the lifting seat 103 downward, and the rotating motor 104 is arranged below the lifting seat 103 The rotating motor 104 is used to drive the mounting base 105 to rotate. The infrared thermal imaging camera 106 is arranged below the mounting base 105. The rotating motor 104 is used to adjust the angle of the infrared thermal imaging camera 106. The infrared thermal imaging camera 106 assists in finding the rescued person by infrared light. The storage battery 107 is used to power the infrared thermal imaging camera 106 and the driving motor 108. The two driving motors 108 are arranged on both sides of the mounting base 105, and are respectively used to drive the driving wheel 109 to rotate and rotate the driven wheel 110. With the cooperation of the above-mentioned crawler 111, the above-mentioned lifting seat 103 is driven to rotate, thereby providing mobility; the above-mentioned lifting seat 103 moves with the above-mentioned UAV body 101. When it moves to a terrain with many obstacles such as floors and woods, the above-mentioned lifting seat 103 can be lowered by the above-mentioned lifting device, and then the above-mentioned driving motor 108 provides the moving power to drive the above-mentioned infrared thermal imaging camera 106 to move and enter the places that are difficult for the above-mentioned UAV body 101 to reach, which can effectively reduce the shooting blind spots, increase the detection range, and have a low cost, thereby solving the problem that the existing UAV has a high cost and a poor effect on increasing the detection range.

[0035] Among them, the winding motor 112 is fixed below the drone body 101; the winding roller 113 is fixed to the output end of the winding motor 112; the suspension rope 114 is fixedly connected to the winding roller 113, and is also fixedly connected to the lifting seat 103; the winding motor 112 is used to drive the winding roller 113 to rotate, release or wind up the suspension rope 114, and connect the lifting seat 103 through the suspension rope 114 to release it downward. The suspension rope 114 has high flexibility and can move with the lifting seat 103 to enter the area that is difficult for the drone body 101 to enter.

[0036] Secondly, multiple magnetic blocks 115 are fixed on the lifting seat 103; multiple electromagnets 116 are arranged below the drone body 101; when the electromagnet 116 is energized, the magnetic blocks 115 can be adsorbed below the drone body 101, and then the lifting seat 103 can be adsorbed below the drone body 101, so as to facilitate its movement. When it is necessary to expand the shooting range, the electromagnet 116 can be released by disconnecting the power.

[0037] Again, the electric push rod 117 is arranged in the storage groove 120 of the lifting seat 103; the walking camera 118 is arranged on the electric push rod 117; the electric push rod 117 is used to push the walking camera 118 upward, and the walking camera 118 is only used to facilitate the operator to control the movement of the infrared thermal imaging camera 106, and the image quality requirement is not high, and the size is small. When not in use, it can be completely stored in the storage groove 120 of the lifting seat 103, and extended through the electric push rod 117 when moving.

[0038] Finally, the rotating motor 119 is fixedly connected to the output end of the electric push rod 117, and is also fixedly connected to the walking camera 118; the rotating motor 119 is used to drive the walking camera 118 to rotate, thereby increasing its shooting range, making it easier for operators to know the surrounding conditions.

[0039] During use, the drone body 101 adopts an existing common drone, and the shooting camera 102 captures specific visual information through visible light to guide the rescue; the lifting device is arranged in the drone body 101, and is used to release the lifting seat 103 downward, and the rotating motor 104 is arranged below the lifting seat 103, and the rotating motor 104 is used to drive the mounting seat 105 to rotate, and the infrared thermal imaging camera 106 is arranged below the mounting seat 105, and the rotating motor 104 is used to adjust the angle of the infrared thermal imaging camera 106, and the infrared thermal imaging camera 106 assists in finding the rescued persons by infrared light. The storage battery 107 is used to supply power to the infrared thermal imaging camera 106 and the drive motor 108. The two drive motors 108 are arranged on both sides of the mounting seat 105, and are respectively used to drive the driving wheel 109 to rotate, and with the cooperation of the driven wheel 110, drive the crawler 111 to rotate, thereby providing mobility; the lifting seat 103 moves with the UAV body 101. When it moves to a terrain with many obstacles such as floors and woods, the lifting seat 103 can be lowered by the lifting device, and then the driving motor 108 provides the moving power to drive the infrared thermal imaging camera 106 to move and enter the UAV body. 101 is difficult to reach, can effectively reduce the shooting blind area, improve the detection range, and the cost is low, the winding motor 112 is used to drive the winding roller 113 to rotate, release or wind the suspension rope 114, and connect the suspension rope 114 and the lifting seat 103 to release it downward, the suspension rope 114 is highly flexible, can move with the lifting seat 103, and enter the area that is difficult for the drone body 101 to enter; when the electromagnet 116 is energized, the magnetic block 115 can be adsorbed under the drone body 101, and then the lifting seat 103 can be adsorbed under the drone body 101, so as to drive it to move conveniently When it is necessary to expand the shooting range, the electromagnet 116 can be released by powering off; the electric push rod 117 is used to push the walking camera 118 upwards. The walking camera 118 is only used to facilitate the operator to control the movement of the infrared thermal imaging camera 106. The image quality requirement is not high and the size is small. When not in use, it can be completely stored in the storage slot 120 of the lifting seat 103, and extended through the electric push rod 117 when moving; the rotating motor 119 is used to drive the walking camera 118 to rotate, thereby increasing its shooting range and making it convenient for the operator to know the surrounding situation; thereby solving the problem that the existing drones are high in cost and have poor effect in increasing the detection range.

[0040] For the second embodiment, please refer to Figure 4 , Figure 4It is a structural schematic diagram of a dual-light personnel search and rescue positioning vision system on an unmanned aerial vehicle according to the second embodiment of the present invention.

[0041] On the basis of the first embodiment, a dual-light personnel search and rescue positioning visual system on a drone of the present invention further includes an alarm light 201 , a buzzer 202 and a signal detector 203 .

[0042] The alarm light 201 is fixed on the top of the drone body 101; the buzzer 202 is arranged on the top of the drone body 101; the alarm light 201 is used to stay on after the rescued person is found, so that the position of the rescued person is easier to be found, and the buzzer 202 provides a sound alarm, which is easier to be found by the rescued person and the operator.

[0043] The signal detector 203 is disposed on the drone body 101 ; the signal detector 203 is used to detect mobile phone signals, so as to facilitate and quickly find the person to be rescued.

[0044] For the third embodiment, please refer to Figure 5 , Figure 5 It is a structural schematic diagram of a dual-light personnel search and rescue positioning vision system on an unmanned aerial vehicle according to the third embodiment of the present invention.

[0045] On the basis of the second embodiment, a dual-light personnel search and rescue positioning vision system on a drone of the present invention further includes a connecting block 301 , an electric lock 302 , a loudspeaker 303 and a lighting lamp 304 .

[0046] The connecting block 301 is arranged on the lifting seat 103 and is fixedly connected to the suspension rope 114; the electric lock buckle 302 is arranged above the connecting block 301; the electric lock buckle 302 is used to limit and fix the connecting block 301 so that the lifting seat 103 and the suspension rope 114 remain connected; when the suspension rope 114 is not long enough, it can also be quickly unlocked by the electric lock buckle 302 to make it detached, so that the infrared thermal imaging camera 106 can be moved to a farther place to increase the search range.

[0047] The loudspeaker 303 is arranged on the drone body 101; the lighting lamp 304 is arranged on the mounting base 105; the loudspeaker 303 is used to enable the operator to communicate with the rescued persons to improve the rescue efficiency; the lighting lamp 304 is used to provide brightness to the walking camera 118 to facilitate entering small and dark places for search and rescue.

[0048] The dual-light personnel search and rescue positioning visual system on the UAV also includes a visual processing module, which is used to identify and locate the target based on the fusion of the visible light image obtained by the shooting camera and the infrared image obtained by the infrared thermal imaging camera.

[0049] The visual processing module includes an image collection unit, an image preprocessing unit, a model training unit and a target detection unit; the image collection unit is used to obtain visible light images and infrared images containing target objects; the image processing unit is used to fuse the visible light images and infrared images into one image using an image fusion method; the model training unit is used to train the YOLO model using a labeled fused image data set; and the target detection unit is used to perform target detection based on the acquired images using the trained YOLO model.

[0050] The image collection unit is the basis of the entire system, which collects image data containing the target object. These image data include two types: visible light images and infrared images. Visible light images are taken under normal lighting conditions, while infrared images are formed in no light or low light environments through infrared radiation emitted or reflected by objects. This dual-modal data collection method can ensure that high-quality target information can be obtained under different environmental conditions.

[0051] The image processing unit processes the collected visible light images and infrared images. Specifically, it combines these two types of images into one image through image fusion technology. Image fusion technology can combine the details in the visible light image with the thermal distribution characteristics in the infrared image, thereby obtaining a new image that contains both rich texture information and temperature distribution characteristics. Such an image is of great significance for subsequent target recognition and classification.

[0052] Then, after the image fusion is completed, the workflow of the model training unit begins. This unit uses the labeled fused image dataset to train the deep learning model. The model used in this case is YOLO (You Only Look Once), a real-time object detection framework known for its fast speed and high accuracy. By training the YOLO model with a large amount of labeled data, the model can learn how to identify specific target objects from complex image backgrounds.

[0053] After the model training is completed, the target detection unit uses the trained YOLO model to detect targets in real-time or newly acquired images. The trained model can quickly and accurately identify targets in the image and provide the target's location information. The entire process is highly automated and can provide efficient and reliable target detection services in various application scenarios.

[0054] The image processing unit includes an image registration subunit, an image decomposition subunit, an image fusion subunit and an image reconstruction subunit; the image registration subunit is used to perform image registration and alignment on the visible light image and the infrared image by using feature point matching; the image decomposition subunit is used to decompose the visible light image and the infrared image into sub-bands of multiple scales; the image fusion subunit is used to fuse the low-frequency information and high-frequency information of the visible light image and the infrared image; the image reconstruction subunit is used to recombine the fused low-frequency and high-frequency information and perform inverse transformation to obtain a final fused image.

[0055] The image processing unit is a key component in the visual processing module, which is further divided into four subunits: image registration subunit, image decomposition subunit, image fusion subunit and image reconstruction subunit. Each subunit undertakes an indispensable task in the image processing process, and works together to achieve high-quality image fusion.

[0056] The first is the image registration subunit, which is used to accurately align the two images by matching the feature points between the visible light image and the infrared image. Since the visible light image and the infrared image may be displaced due to differences in shooting angle, focal length, etc., image registration is a prerequisite for ensuring good subsequent fusion effects. By finding the corresponding feature points in the two images and adjusting the positions of the images so that these feature points overlap in the two images, the spatial alignment of the images is achieved.

[0057] Next is the image decomposition subunit, whose task is to decompose the aligned visible light image and infrared image into multiple different frequency sub-bands. This decomposition is usually achieved through multi-resolution analysis techniques, such as wavelet transform. The decomposed image can be regarded as a hierarchical structure composed of different frequency components, and each layer represents the information within a specific scale range of the image.

[0058] Then it enters the image fusion subunit, whose main task is to fuse the low-frequency information and high-frequency information of the decomposed visible light image and infrared image. Low-frequency information usually represents the basic outline and structure of the image, while high-frequency information reflects the details and edges of the image. Through specific fusion rules (such as pixel value-based, region-based or structure-based fusion strategies), the low-frequency and high-frequency information from the visible light and infrared images are selectively merged to generate new fused image features.

[0059] Finally, the image reconstruction subunit is responsible for recombining the fused low-frequency and high-frequency information. This step usually involves an inverse transformation process, which is to reconstruct the fused subbands of multiple scales into a complete image. Through this series of processing, the final fused image not only retains the rich details in the visible light image, but also integrates the thermal radiation information in the infrared image, thus forming a high-quality image with more comprehensive information and conducive to subsequent target detection and recognition.

[0060] What is disclosed above is only a preferred embodiment of the present invention, and it certainly cannot be used to limit the scope of rights of the present invention. Ordinary technicians in this field can understand that all or part of the processes of the above embodiment and equivalent changes made according to the claims of the present invention still fall within the scope of the invention.

Claims

1. A dual-light personnel search and rescue positioning visual system on an unmanned aerial vehicle, characterized in that: The invention comprises a drone body, a shooting camera, a lifting device, a lifting seat, a rotating motor, a mounting seat, an infrared thermal imaging camera, a battery, two driving motors, two driving wheels, a plurality of driven wheels and two crawlers, wherein the shooting camera is arranged at the front end of the drone body; the lifting device is arranged below the drone body; the lifting seat is arranged below the lifting device; the rotating motor is fixed in the lifting seat; the mounting seat is fixed at the output end of the rotating motor; and the infrared thermal imaging camera is arranged at the front end of the mounting seat; The battery is arranged in the lifting seat; the two driving motors are arranged on both sides of the mounting seat; the two driving wheels are fixed to the output ends of the two driving motors; the multiple driven wheels are rotatably arranged on both sides of the mounting seat; the two crawlers are respectively matched with the two driving wheels and matched with the multiple driven wheels.

2. A dual-light personnel search and rescue positioning visual system on an unmanned aerial vehicle as claimed in claim 1, characterized in that: The lifting device includes a winding motor, a winding roller and a suspension rope, wherein the winding motor is fixed below the drone body; the winding roller is fixed at the output end of the winding motor; the suspension rope is fixedly connected to the winding roller and to the lifting seat.

3. A dual-light personnel search and rescue positioning visual system on an unmanned aerial vehicle as claimed in claim 2, characterized in that: The dual-light personnel search and rescue positioning visual system on the drone also includes a plurality of magnetic blocks and a plurality of electromagnets, wherein the plurality of magnetic blocks are fixed on the lifting seat; and the plurality of electromagnets are arranged below the drone body.

4. A dual-light personnel search and rescue positioning visual system on an unmanned aerial vehicle as claimed in claim 3, characterized in that: The dual-light personnel search and rescue positioning visual system on the UAV also includes an electric push rod and a walking camera. The lifting seat has a storage groove; the electric push rod is arranged in the storage groove of the lifting seat; and the walking camera is arranged on the electric push rod.

5. A dual-light personnel search and rescue positioning visual system on an unmanned aerial vehicle as claimed in claim 4, characterized in that: The dual-light personnel search and rescue positioning visual system on the unmanned aerial vehicle also includes a rotating motor, which is fixedly connected to the output end of the electric push rod and fixedly connected to the walking camera.

6. A dual-light personnel search and rescue positioning visual system on an unmanned aerial vehicle as claimed in claim 4, characterized in that: The dual-light personnel search and rescue positioning visual system on the UAV also includes a visual processing module, which is used to identify and locate the target based on the fusion of the visible light image obtained by the shooting camera and the infrared image obtained by the infrared thermal imaging camera.

7. A dual-light personnel search and rescue positioning visual system on an unmanned aerial vehicle as claimed in claim 6, characterized in that: The visual processing module includes an image collection unit, an image preprocessing unit, a model training unit and a target detection unit; The image collection unit is used to obtain a visible light image and an infrared image containing a target object; The image processing unit is used to fuse the visible light image and the infrared image into one image using an image fusion method; The model training unit is used to train the YOLO model using the labeled fused image dataset; The target detection unit is used to perform target detection based on the acquired image using the trained YOLO model.

8. A dual-light personnel search and rescue positioning visual system on an unmanned aerial vehicle as claimed in claim 7, characterized in that: The image processing unit includes an image registration subunit, an image decomposition subunit, an image fusion subunit and an image reconstruction subunit; The image registration subunit is used to perform image registration and alignment on the visible light image and the infrared image by using feature point matching; The image decomposition subunit is used to decompose the visible light image and the infrared image into sub-bands of multiple scales; The image fusion subunit is used to fuse the low-frequency information and high-frequency information of the visible light image and the infrared image; The image reconstruction subunit is used to recombine the fused low-frequency and high-frequency information and perform inverse transformation to obtain a final fused image.

Images