Hyper spectrum remote sensing imaging technology-based miniaturized drone load

Abstract

To provide a miniaturized drone load based on a hyper spectrum remote sensing imaging technology.SOLUTION: A miniaturized drone load includes: a spectrum collection system that is fixed to a tip of a drone; a non-axial parabolic mirror that collects a zenith reference spectrum and a negative elevation angle ground surface scattering light measurement spectrum array, and transmits them to a spectrometer by an optical fibre; an imaging auxiliary module that images a detection area and recognizes an object material as auxiliary information; and a gyro stabilization module that positions the actuation azimuth of the drone and the spectrum collection system. The spectroscopic system includes: the spectrometer that converts an input optical signal into an electric signal, and combines, by a micro-computer, the auxiliary information and modification information to perform hyper spectrum remote sensing imaging; a semiconductor freezing module that freezes the spectrometer; and a radiator of the semiconductor freezing module. A load is higher in spatial resolving power than a satellite remote sensing instrument and a ground remoter sensing instrument, and can flexibly and quickly monitor a traceability area.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to the field of optical remote sensing, and more particularly to a miniaturized drone load based on hyperspectral remote sensing imaging technology. [Background technology]

[0002] Air quality problems around the world have been a constant threat to people's lives and health. In order to combat environmental problems, in addition to air pollutant prevention and control measures such as optimizing the industrial chain and energy conservation and emission reduction, effective monitoring of pollutant emissions is an important part of environmental protection.

[0003] Currently, high-concentration air pollution emissions are often concentrated in industrial areas such as factories and power plants, where emissions are often concentrated and complex. Current environmental monitoring methods struggle to effectively monitor and trace these emissions. While environmental remote sensing satellites can monitor large areas, their spatial and temporal resolution is severely insufficient. Due to cloud obscuration, an industrial park may only be represented by a single pixel in the satellite image, making it impossible to distinguish between different plants and making traceability impossible. For example, the vehicle-mounted mobile VOCs and odor gas mass spectrometer and pollution source precision locking system (Publication No. CN108415038B) cannot penetrate deep into the plant interior for observation, making it vulnerable to detection by the industrial park, forcing the park to suspend heavy pollutant emissions and making traceability difficult. For example, the remote sensing imaging aerial photography drone disclosed under publication number CN108016628A has a very limited measurement range of the point-type device mounted on the drone, and can only measure the pollutant concentration at the drone's flight altitude (target sampling), making it extremely inefficient when monitoring industrial parks.

[0004] Therefore, to monitor the horizontal distribution of air pollutants in industrial parks with high spatiotemporal resolution, trace the pollutants, and simultaneously prevent equipment from entering the target area, a hyperspectral remote sensing imaging payload mounted on a drone is required. Furthermore, fixed-wing drones and large drones are clearly inappropriate, given that they would not attract the attention of air traffic control or the plant. Therefore, there is currently a strong demand for miniaturized payloads that can be mounted on small drones, in order to quickly fill the technological gap in monitoring and tracing air pollutants in industrial plants. Summary of the Invention [Problem to be solved by the invention]

[0005] In view of the above, the object of the present invention is to provide a miniaturized drone load based on hyperspectral remote sensing imaging technology, which has higher spatial resolution (reaching meter-scale resolution) than satellite remote sensing equipment and ground remote sensing equipment, and can flexibly and quickly monitor areas where traceability is not possible with various remote sensing equipment and point sampling equipment. [Means for solving the problem]

[0006] To achieve the above object of the invention, a miniaturized drone load based on hyperspectral remote sensing imaging technology according to an embodiment of the present invention includes a spectrum collection system and a spectroscopic system that are designed separately.

[0007] The spectrum collection system is fixed to the tip of the drone and has an unobstructed field of view. It includes an off-axis parabolic mirror, an imaging auxiliary module, a gyro stabilization module, and a motor. The motor drives and synchronously adjusts the on-axis parabolic mirror, the imaging auxiliary module, and the gyro stabilization module. The off-axis parabolic mirror collects the zenith reference spectrum and the negative elevation angle ground scattered light measurement spectrum array and transmits them to the spectrometer via optical fiber. The imaging auxiliary module images the detection area and recognizes the object material in the detection area as auxiliary information. The gyro stabilization module positions the operating orientation of the drone and the spectrum collection system as correction information.

[0008] The spectrometer system is fixed below the drone and includes a spectrometer, a circuit control system, a microcomputer, a semiconductor refrigeration module, and a heat sink. The spectrometer converts the input optical signal into an electrical signal, and the microcomputer combines the auxiliary information and correction information to perform hyperspectral remote sensing imaging. The semiconductor refrigeration module refrigerates the spectrometer, and the heat sink dissipates heat from the semiconductor refrigeration module. The circuit control system controls the operation of the motor, spectrometer, microcomputer, gyro stabilization module, and semiconductor refrigeration module.

[0009] Preferably, the off-axis parabolic mirror is used to fold the optical path by 90° to focus the solar scattered light onto the end face of the optical fiber, thereby realizing the collection of the negative elevation angle ground scattered light measurement spectrum array and the zenith reference spectrum, and the focusing axis of the off-axis parabolic mirror is coaxial with the rotation axis of the motor, and the parameters of the parabolic mirror must satisfy the numerical aperture NA of the optical fiber used so as to maximize the density of the light flux entering the optical fiber.

[0010] Preferably, the off-axis parabolic mirror is an off-axis parabolic mirror plated with a high ultraviolet reflectance film.

[0011] Preferably, the imaging auxiliary module includes, in its hardware, a high-definition camera and a small circuit board equipped with an image recognition algorithm and a surface albedo algorithm based on a radiation transmission model. The observation direction of the high-definition camera is aligned with the detection direction of the off-axis parabolic mirror and can autofocus from 100 m to infinity. The captured detection area is used for mapping the late-stage hyperspectral trace gas imaging results and the detection area. An image recognition algorithm is installed to recognize the object material in the detection area image. A surface albedo algorithm based on a radiation transmission model calculates the recognized object material to obtain the surface albedo of different objects in different environments. The detection area image and the surface albedo corresponding to the object material are transmitted to the microcomputer as auxiliary information.

[0012] Preferably, the gyro stabilization module is used to feedback the current rotation angle in real time, and the feedback current rotation angle is used as the basis for angle adjustment. When the spectrum collection system measures a certain elevation angle, the motor can adjust its angle based on the feedback current rotation angle to ensure that the collected certain elevation angle is fixed at a set value.

[0013] The gyro stabilization module is also used to monitor the real-time attitude angle of the hyperspectral drone load in real time, and the real-time attitude angle is used as correction information for geometric correction of the hyperspectral remote sensing imaging, and the monitoring accuracy of the gyro stabilization module is greater than or equal to 0.1°.

[0014] Preferably, the spectrometer system further includes a heat conductor employing a metal layer having high thermal conductivity and light weight to completely encase the outer surface of the spectrometer or to partially encase the outer surface adjacent to the spectrometer CCD, thereby realizing cooling of the spectrometer by increasing the heat exchange area between the spectrometer and the semiconductor refrigeration module.

[0015] Preferably, the spectroscopic system further includes a temperature sensor, which feeds back the temperature of the spectrometer to the circuit control system in real time, and the circuit control system controls the semiconductor refrigeration module to change the temperature in real time, so that the semiconductor refrigeration module can automatically adjust and ensure a constant temperature of the spectrometer.

[0016] Preferably, the heat sink includes a heat dissipation fin array and a heat dissipation fan that blows external air onto the heat dissipation fin array. The air passes through the heat dissipation fin array, and the air flow direction in the heat dissipation fins and the air flow direction generated by the drone propeller can form a vortex. Due to the position of the heat sink, the air heat absorbed by the heat dissipation fins from the semiconductor refrigeration module can be more quickly carried away by the drone propeller. Here, the heat dissipation fins are made of a material that is lightweight and has high thermal conductivity, and the heat dissipation effect of the heat dissipation fan is greater than the heat generation effect of the semiconductor refrigeration module.

[0017] Preferably, the slits between the spectrometer, the heat conductor, the semiconductor refrigerator module, and the heat sink are filled with heat-dissipating silicone grease.

[0018] When assembled, a lightweight insulating material is used to separate the heat conductor and spectrometer from the outside world to prevent heat exchange with the outside world.

[0019] Preferably, the spectroscopic system adopts a skeleton and housing structure, with a microcomputer, spectrometer, circuit control system, semiconductor refrigeration module, heat sink, and WIFI module fixed inside the skeleton, and the outside of the skeleton fixed to the drone. The housing uses multiple thin carbon fiber plates to encase the entire spectroscopic system, and each carbon fiber plate is fixed to the skeleton, and a method of assembling the plates together is adopted. [Effects of the Invention]

[0020] Compared with the prior art, the beneficial effects of the present invention include at least the following: The lightweight design of the drone's overall load structure ensures the normal implementation of hyperspectral technology and allows it to be normally mounted on an electric mini-rotor drone with a load of around 2.8 kg, thereby realizing the ability to mount ultra-hyperspectral equipment on a drone. The design of the equipment's spectrum collection system can normally measure the zenith reference spectrum, ground measurement spectrum, surface albedo, etc. required for hyperspectral technology, and ensure its stable operation. The design of the spectroscopic system allows the entire spectrum processing system to maintain a temperature change within 0.1°C, ensuring the normal operation of the spectrometer and the quality of spectral imaging. [Brief explanation of the drawings]

[0021] In order to more clearly describe the technical solutions in the embodiments of the present invention or the existing technology, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the existing technology. It is obvious that the accompanying drawings in the following description are only some embodiments of the present invention, and those skilled in the art can obtain other accompanying drawings based on these accompanying drawings without any creative efforts. [Figure 1] FIG. 1 is a structural schematic diagram of a miniaturized drone load according to an embodiment. [Figure 2] 1 is a structural schematic diagram of a spectrum collection system according to an embodiment; [Figure 3] 1 is a structural schematic diagram of a spectroscopic system according to an embodiment. [Figure 4] 1A and 1B are diagrams illustrating a fitting relationship between a spectroscopic system and a drone rotor according to an embodiment. [Figure 5] FIG. 1 is a schematic diagram illustrating a lightweight design of a spectroscopic system according to an embodiment. [Figure 6] 10 shows the horizontal imaging results of a plant equipped with a hyperspectral drone load according to an embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0022] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the drawings and examples. It should be understood that the specific embodiments described herein are only used to illustrate the present invention, and do not limit the protection scope of the present invention.

[0023] The technical concept of the present invention solves the key technical problems of existing pollution source monitoring, such as low timeliness, low effectiveness, and low flexibility. In order to do this, an embodiment of the present invention provides a miniaturized drone load based on hyperspectral remote sensing imaging technology, which can perform meter-scale resolution horizontal observation of pollutants in industrial parks, locate the specific location of emission sources, and achieve effective monitoring. It can also complete a single measurement for multiple plants within 20 minutes, eliminating plant reaction time and enabling rapid inspection. It can also monitor different types of plants without drones flying into the factories, allowing for flexible inspection.

[0024] Based on the above technical concept, the miniaturized drone load based on hyperspectral remote sensing imaging technology of the embodiment includes two parts, a spectrum collection system and a spectroscopic system, as shown in Figure 1, and the two parts are designed separately.

[0025] As shown in Figures 1 and 2, the spectrum collection system includes an off-axis parabolic mirror, an imaging auxiliary module, a gyro stabilization module, and a motor. The off-axis parabolic mirror is the core component, and its rotation is driven by the motor. It folds the optical path by 90° to focus the solar scattered light onto the end face of the optical fiber, enabling the collection of the zenith reference spectrum and the negative elevation angle ground scattered light measurement spectrum array. The use of a single off-axis parabolic mirror optical element can reduce the optical signal loss in the optical system. At the same time, compared to lenses, the use of an off-axis parabolic mirror further reduces the optical signal loss in the optical system without introducing spherical aberration, chromatic aberration, phase delay, or absorption loss.

[0026] The off-axis parabolic mirror collects the zenith reference spectrum and ground-scattered light and focuses the signal light into the optical fiber. To significantly reduce the optical system loss and optical aberrations, the present invention employs an off-axis parabolic mirror coated with a high-ultraviolet reflectivity film. When selecting an off-axis parabolic mirror, its mass must mechanically satisfy the moment of the motor. If the focus axis of the off-axis parabolic mirror is coaxial with the rotation axis of the motor, the moment required for the motor will be minimized. The off-axis parabolic mirror must also optically satisfy the numerical aperture (NA) of the optical fiber. That is, if the acceptance angle of the optical fiber is A degrees (e.g., 14°), the image-side aperture angle at which collimated light passes through the off-axis parabolic mirror must be greater than A degrees (e.g., 14°). The arctangent value of the ratio of the radius of the off-axis parabolic mirror to its effective focal length must be greater than A degrees (14°), ultimately maximizing the density of the signal beam entering the optical fiber.

[0027] The imaging auxiliary module is used to capture images of the detection area and recognize the material of objects in the detection area as auxiliary information. Specifically, the hardware includes a high-definition camera and a small circuit board equipped with an image recognition algorithm and a ground albedo algorithm based on a radiation transmission model. The observation direction of the high-definition camera should match the detection direction of the off-axis parabolic mirror, ignoring the distance between them. The detection field of view of the off-axis parabolic mirror should be midway between the observation screen of the high-definition camera. To accommodate the ground imaging needs of hyperspectral drones at different heights, the high-definition camera should enable automatic autofocus with a focus range of at least 100m to infinity. The imaging auxiliary module is used to image the detection area collected by the optical signal, and the detection area is used for mapping the later hyperspectral trace gas imaging results and the detection area captured by the camera. Meanwhile, the detection area image collected by the high-definition camera is passed through an image recognition algorithm to recognize the object material in the image, and a surface albedo algorithm based on a radiation transmission model is used to obtain the surface albedo of different objects in different environments to assist the hyperspectral imaging, making the imaging results more accurate. Here, the detection area image and the surface albedo corresponding to the object material are transmitted to the microcomputer as auxiliary information. When the surface albedo is taken into account, the hyperspectral imaging results are more consistent with reality.

[0028] The gyro stabilization module, on the one hand, feeds back the spectrum collection system attitude signal and cooperates with the motor to maintain the stability of the spectrum collection system in real time; on the other hand, it determines the real-time attitude angle of the hyperspectral drone load, which is convenient for correcting the hyperspectral imaging results in the future, i.e., it is used to position the operating orientation of the drone and the spectrum collection system as correction information. Specifically, it is used to monitor and feed back the current rotation angle in real time. During actual observation, the drone's flight attitude changes relatively significantly, which has a relatively large impact on optical measurement. Therefore, the gyro stabilization module can accurately feed back the current rotation angle in real time, and the fed back current rotation angle serves as the basis for angle adjustment. When the spectrum collection system measures a certain elevation angle, the motor can adjust its angle based on the fed back current rotation angle to ensure that the collected elevation angle is fixed at the set value.

[0029] The gyro stabilization module is also used to monitor the real-time attitude angle of the hyperspectral drone payload in real time, and the real-time attitude angle is used as correction information for geometric correction of the hyperspectral remote sensing imaging. The monitoring accuracy of the gyro stabilization module is greater than 0.1°, and the mass is as small as possible.

[0030] The off-axis parabolic mirror, the imaging assistance module, and the gyro stabilization module are all fixed together, and the mass and moment of inertia of the three components should be smaller than the rated load of the motor. The motor drives and synchronizes the off-axis parabolic mirror, the imaging assistance module, and the gyro stabilization module, and the control accuracy of the motor should be less than 0.05°, so a small (micro) stepping motor is appropriate.

[0031] The spectral collection system is fixed to the tip of the drone by a fixing device and does not block the field of view, i.e., the off-axis parabolic mirror and high-definition camera are extended from the drone, and when collecting the zenith reference spectrum and ground scattered light, the drone body and the hyperspectral drone load are not blocked, allowing the hyperspectral drone load to perform normal remote sensing imaging. The material selection for the fixing device should be based on a simple and lightweight foundation, and should be lightweight, high-strength, and easy to process (e.g., aluminum alloy).

[0032] As shown in Figures 1 and 3, the spectrometer system includes a spectrometer, a circuit control system, a microcomputer, a semiconductor refrigeration module, a heat sink, a heat conductor, and a temperature sensor. The spectrometer converts the input optical signal into an electrical signal, and the microcomputer combines the auxiliary information and correction information to perform hyperspectral remote sensing imaging. The semiconductor refrigeration module refrigerates the spectrometer and controls the overall temperature of the spectrometer system to ensure the quality of the hyperspectral imaging. The heat sink dissipates heat from the semiconductor refrigeration module. The circuit control system controls the operation of the motor, spectrometer, microcomputer, and semiconductor refrigeration module. There is also a Wi-Fi module (not shown in Figure 1) for communication between the ground and the hyperspectral drone load.

[0033] Spectrometers are highly sensitive to temperature, which can affect the spectral nonlinear feedback and alter the size, shape, and intensity of spectral images. In hyperspectral technology, the impact of temperature changes on the spectrometer directly affects the imaging efficiency and the validity of the results. Figure 3 shows two refrigeration-equipped spectrometer systems. The left side is a full-wrapped spectrometer system, in which the entire spectrometer surface is refrigerated by wrapping it in a thermal conductor. The right side is a half-wrapped spectrometer system, in which only a portion of the spectrometer surface is refrigerated by wrapping it in a thermal conductor. Currently, most hyperspectral instruments refrigerate the entire spectrometer, or even the entire instrument. However, given the significant payload limitations of small drones, it is possible to refrigerate only key parts of the spectrometer, such as the area near the CCD.

[0034] The heat conductor is typically a lightweight metal layer with high thermal conductivity, such as an aluminum alloy. Because the spectrometer housing may be irregular, the heat conductor must be designed based on the shape and size of the spectrometer. The function of the heat conductor is to increase the heat exchange area between the semiconductor refrigeration module and the spectrometer, thereby achieving refrigeration of the spectrometer.

[0035] Temperature sensors are used to monitor the temperature of selected locations in the refrigeration system and during operation of the hyperspectral drone. One or more temperature sensors can be used, and they are typically placed near the CCD of the spectrometer, away from the top of the heat conductor (half-wrapped refrigeration system). The temperature sensors provide real-time feedback of the spectrometer temperature to the circuit control system, which then controls the semiconductor refrigeration module to change its temperature in real time, allowing the semiconductor refrigeration module to automatically adjust and maintain a constant temperature for the spectrometer.

[0036] The semiconductor refrigeration module freezes the thermal conductor. Taking into account the quality of the entire device, the semiconductor refrigeration module should be combined with temperature feedback from a temperature sensor to stabilize the temperature change of the spectrometer within 0.1°C. It is fundamentally necessary to ensure that the hyperspectral imaging results do not change significantly due to the operation of the spectrometer.

[0037] The radiator, which includes a heat dissipation fin array and a heat dissipation fan, dissipates heat from the semiconductor refrigeration module. The heat dissipation fan blows external air over the heat dissipation fin array, which then carries away the heat absorbed from the semiconductor refrigeration module. The selection and positioning of the heat dissipation fan and fin array must consider both quality and heat dissipation effectiveness. To reduce mass, the radiator's fins should be made of a lighter material with higher thermal conductivity (e.g., copper or aluminum, depending on the actual load mass and the drone's strength; copper can be used for higher thermal conductivity, while aluminum can be used for lighter weight). Increasing the surface area of the fins improves the radiator's heat dissipation effectiveness. The selection of a fan for the radiator must consider mass and power consumption, while the fan's heat dissipation effect must exceed the heat generated by the semiconductor refrigeration module.

[0038] The slits between the spectrometer, heat conductor, semiconductor refrigeration module, and heat sink are filled with heat-dissipating silicone grease to achieve a higher heat dissipation effect. To achieve a higher refrigeration effect, a lightweight insulating material (e.g., aerogel) must be used to separate the heat conductor and spectrometer from the outside during assembly to prevent heat exchange between the spectrometer and the outside.

[0039] The heat dissipation fan of the heat sink of the present invention and the rotor of the drone have a certain cooperative effect, as shown in Figure 4. According to the Bernoulli principle, under the action of the rotor of the drone, the air flow velocity below the rotor is fast and high, generating upward lift for the drone, while low pressure is generated below the rotor, allowing ambient air to be replenished below the rotor. The air flow direction of the bottom-mounted heat sink passes through the heat dissipation fan and the heat dissipation fin array from below and then leaves the side of the heat sink. Therefore, under the action of the rotor, the air in the heat dissipation fin array absorbs heat and accelerates its extraction to the side of the heat sink, improving the heat dissipation effect of the heat sink on the semiconductor refrigeration module. This further increases the refrigeration power of the semiconductor refrigeration module and improves the refrigeration effect of the equipment.

[0040] The lightweight design of the drone payload in the above example is manifested in two main aspects: mass is an important consideration for all components, while the hyperspectral drone payload adopts a separate design and a skeleton + housing structure.

[0041] The separate design separates the spectral collection system from the spectroscopic system, mounting the spectroscopic system below the drone and placing the spectral collection system at the tip of the drone. Because the spectrometer, heat conductor, and heat sink of the spectroscopic system are relatively heavy, mounting them below the drone makes the flight of the small rotor drone more stable. The spectral collection system is fixed to the tip of the drone, allowing the off-axis parabolic mirror and high-definition camera to extend from the drone. This allows the drone body and hyperspectral drone load to collect zenith reference spectrum and ground scattered light without being blocked, allowing the hyperspectral drone load to perform normal remote sensing imaging.

[0042] The spectroscopic system adopts a skeleton + housing structure, that is, each element is fixed by a lightweight "skeleton", and the entire device is wrapped in a customized thin carbon fiber plate as a housing. The ultimate goal is to significantly reduce the mass of the drone load while maintaining stability.

[0043] As shown in Figure 5, the framework functions to secure the microcomputer, spectrometer, circuit control system, and Wi-Fi module. The microcomputer, circuit control system, and Wi-Fi module are secured to the upper half and are cooled by miniature cooling fans on both sides. The spectrometer is secured to the lower half and is cooled by the semiconductor module and heat sink. The framework also includes a drone connector for connecting the spectrometer system to the rotor drone. The housing is made of thin carbon fiber material. Due to its low plasticity, multiple flat plates are used to encase the entire spectrometer system, and each carbon fiber plate is fixed to the framework, forming an assembly. The multi-plate structure also offers greater scalability. Simply designing a portion of the carbon fiber plate based on different spectrometer models or circuit boards or other devices eliminates the need to redesign the entire housing, making it convenient for securing loads to different miniature drone models.

[0044] Because this invention needs to be mounted on a rotor drone with a load of around 2.8 kg, the requirements for a lightweight design are very strict. The frame is made of aluminum alloy or other lightweight, high-strength, and easy-to-process materials, and carbon fiber, and the various parts of the spectroscopic system are connected by several columns. Ultimately, the mass of the frame and housing structure should be less than 400 g.

[0045] Figure 6 shows the results of observations of a plant using the miniaturized drone payload described above. Strong emissions of NOx and HCHO were observed in area 1 of a certain company, and strong emissions of HCHO were observed in areas 2 and 3. The observation image is generally smooth, with distinct high and low value areas, corresponding to factory buildings that emit air pollutants and offices and farmland that do not, respectively. The imaging results are consistent with the actual situation.

[0046] While conventional hyperspectral target imaging remote sensing can easily image prominent targets such as chimneys, observing the interior of a plant is extremely difficult. The drone remote sensing imaging method using hyperspectral drone loads of the present invention compensates for the shortcomings of target imaging remote sensing, capturing the spatial distribution of pollutants throughout the plant with meter-scale resolution and accurately locating high-value areas within the industrial park.

[0047] The hyperspectral remote sensing imaging technology of the above embodiment is based on a miniaturized drone load, and by designing a spectrum collection system, spectrometer system, and lightweight load, a hyperspectral remote sensing imager that can be mounted on a small rotor drone with a load of approximately 2.8 kg is realized, and meter-scale monitoring of key industrial parks is carried out. The spectrum collection system design enables the device to achieve stable observation of hyperspectral remote sensing imaging, and the refrigeration system design enables the device to ensure normal operation of the spectrometer for long periods of time. The lightweight design of the hyperspectral drone load allows the device to be stably mounted on a small drone and achieve meter-scale resolution remote sensing imaging of air pollutants.

[0048] The specific embodiments described above have further explained the technical solutions and beneficial effects of the present invention in detail, but it should be understood that what has been described above is merely the best example of the present invention and is not intended to limit the present invention, and any modifications, supplements, equivalent replacements, etc. made within the principle scope of the present invention should all be included in the protection scope of the present invention.

Claims

1. A miniaturized drone load based on hyperspectral remote sensing imaging technology, including a separately designed spectrum collection system and a spectroscopic system, The spectrum collection system is fixed to the tip of the drone and has an unobstructed field of view. The spectrum collection system includes an off-axis parabolic mirror, an imaging auxiliary module, a gyro stabilization module, and a motor. The motor drives and synchronizes the on-axis parabolic mirror, the imaging auxiliary module, and the gyro stabilization module. The off-axis parabolic mirror collects the zenith reference spectrum and the negative elevation angle ground scattered light measurement spectrum array and transmits them to the spectrometer via the optical fiber. The imaging auxiliary module images the detection area and recognizes the object material in the detection area as auxiliary information. The gyro stabilization module positions the operating orientation of the drone and the spectrum collection system as correction information. The spectroscopic system is fixed below the drone and includes a spectrometer, a circuit control system, a microcomputer, a semiconductor refrigeration module, and a heat sink, wherein the spectrometer converts input optical signals into electrical signals, and the microcomputer combines the auxiliary information and correction information to perform hyperspectral remote sensing imaging, the semiconductor refrigeration module refrigerates the spectrometer, the heat sink dissipates heat from the semiconductor refrigeration module, and the circuit control system controls the operation of the motor, the spectrometer, the microcomputer, the gyro stabilization module, and the semiconductor refrigeration module.

2. The off-axis parabolic mirror is used to fold the optical path by 90° to focus the solar scattered light onto the end face of the optical fiber, thereby realizing the collection of a negative elevation angle ground scattered light measurement spectrum array and a zenith reference spectrum; the focusing axis of the off-axis parabolic mirror is coaxial with the rotation axis of the motor; and the parameters of the parabolic mirror must meet the numerical aperture NA of the optical fiber used so as to maximize the density of the light flux entering the optical fiber.

3. The miniaturized drone load based on hyperspectral remote sensing imaging technology as described in claim 1, characterized in that the off-axis parabolic mirror is an off-axis parabolic mirror plated with a high ultraviolet reflectivity film.

4. 2. The imaging auxiliary module includes a hardware high-definition camera and a small circuit board equipped with an image recognition algorithm and a surface albedo algorithm based on a radiation transmission model. The observation direction of the high-definition camera is aligned with the detection direction of the off-axis parabolic mirror and can autofocus from 100 m to infinity. The captured detection area is used for later hyperspectral trace gas imaging results and mapping of the detection area. An image recognition algorithm is installed to recognize the object material in the detection area image. A surface albedo algorithm based on a radiation transmission model calculates the recognized object material to obtain the surface albedo of different objects in different environments. The detection area image and the surface albedo corresponding to the object material are transmitted to a microcomputer as auxiliary information.

5. The gyro stabilization module is used to feedback the current rotation angle in real time, and the feedback current rotation angle is used as the basis for angle adjustment. When the spectrum collection system measures a certain elevation angle, the motor can adjust its angle according to the feedback current rotation angle, ensuring that the collected certain elevation angle is fixed at a set value; The miniaturized drone load based on hyperspectral remote sensing imaging technology of claim 1, characterized in that the gyro stabilization module is also used to monitor the real-time attitude angle of the hyperspectral drone load in real time, and the real-time attitude angle is used as correction information for geometric correction of the hyperspectral remote sensing imaging, and the monitoring accuracy of the gyro stabilization module is not less than 0.1°.

6. The miniaturized drone load based on hyperspectral remote sensing imaging technology of claim 1, characterized in that the spectroscopic system further includes a heat conductor that uses a metal layer with high thermal conductivity and light weight to completely encase the outer surface of the spectrometer or partially encase the outer surface adjacent to the spectrometer CCD, thereby achieving cooling of the spectrometer by increasing the heat exchange area between the spectrometer and the semiconductor refrigeration module.

7. The miniaturized drone load based on hyperspectral remote sensing imaging technology of claim 1, characterized in that the spectroscopic system further includes a temperature sensor, which feeds back the temperature of the spectrometer to the circuit control system in real time, and the circuit control system controls the semiconductor refrigeration module to change the temperature in real time, so that the semiconductor refrigeration module can automatically adjust and ensure a constant temperature of the spectrometer.

8. 2. The miniaturized drone load based on hyperspectral remote sensing imaging technology of claim 1, wherein the heat sink includes a heat dissipation fin array and a heat dissipation fan that blows external air onto the heat dissipation fin array. The air passes through the heat dissipation fin array, and the air flow direction within the heat dissipation fins and the air flow direction generated by the drone propeller form a vortex. The position of the heat sink allows the air heat absorbed by the heat dissipation fins from the semiconductor refrigeration module to be more quickly carried away by the drone propeller. The heat dissipation fins are made of a material with light weight and high thermal conductivity, and the heat dissipation effect of the heat dissipation fan is greater than the heat-generating effect of the semiconductor refrigeration module.

9. The slits between the spectrometer, the heat conductor, the semiconductor refrigeration module, and the heat sink are filled with heat-dissipating silicone grease. The miniaturized drone load based on hyperspectral remote sensing imaging technology described in claim 6, characterized in that when assembled, a lightweight insulating material is used to separate the heat conductor and the spectrometer from the outside and prevent heat exchange with the outside.

10. The miniaturized drone load based on hyperspectral remote sensing imaging technology of claim 1, characterized in that the spectroscopic system adopts a skeleton and housing structure, with a microcomputer, a spectrometer, a circuit control system, a semiconductor refrigeration module, a heat sink, and a WiFi module fixed inside the skeleton, and the outside of the skeleton is fixed to the drone. The housing uses multiple thin carbon fiber plates to encase the entire spectroscopic system, and each carbon fiber plate is fixed to the skeleton, using a plate-by-plate assembly method.

Images