A linear array remote sensing system integrating sensing, storage, and computing, and its data processing method.

Abstract

This invention discloses an integrated sensing, storage, and computing linear array remote sensing system and its data processing method. The integrated sensing, storage, and computing linear array remote sensing system includes a host system, a slave system, a housing, and a standard reflector. The host system includes a central processing unit, a data storage device, a high-precision pose module, and an incident light calibration module. The high-precision pose module is used to acquire real-time geographical location and attitude information of the linear array remote sensing system during data acquisition while the external carrier is moving. The host system is installed on the back of the housing. The front of the housing includes several compartments. The slave system includes a linear array imaging sensor and a coprocessor, which are located in the compartments and connected to the host system via an electrical bus within the housing. The host system is used to control the slave system to acquire and process linear array remote sensing data. The standard reflector is used for the calibration of the linear array imaging sensor. The data processing method is used for real-time calibration and mapping of the linear array remote sensing data.

Description

Technical Field

[0001] This invention relates to the technical field of crop remote sensing monitoring, and more specifically, to a linear array remote sensing system integrating sensing, storage, and computing, and its data processing method. Background Technology

[0002] Rapid, non-destructive, and accurate monitoring of crop growth, and a precise understanding of crop growth status, are the prerequisites and foundations for carrying out precision agriculture. This helps to improve water, fertilizer, and pesticide management strategies, enhance management precision, and is of great significance for preventing farmland environmental pollution caused by excessive use of fertilizers and pesticides, as well as for conducting in-depth research and application of precision agriculture.

[0003] Traditional aerospace and airborne remote sensing technologies suffer from drawbacks such as numerous meteorological influencing factors, long processing times, and high costs. Near-ground remote sensing of crop growth information using ground tractors, elevated vehicles, and drones can overcome these shortcomings. Drone technology has developed rapidly in recent years, offering improved operability, lower costs, and greater coverage per image compared to ground-based machinery, while also being more efficient. Therefore, the application of low-altitude drone remote sensing in agriculture is rapidly expanding, making it one of the main methods for remote sensing crop information and a crucial direction for the development of precision agriculture.

[0004] Crop growth information remote sensing is mainly divided into three categories according to the arrangement of the data points in a single observation: points, lines, and areas.

[0005] Currently, low-altitude remote sensing monitoring using UAVs primarily employs area array imaging sensors. Representative models include the Rededge-M series multispectral cameras from Micasense, the ADC series multispectral cameras from Tetracam, the MS600 multispectral camera from Changguang Yuchen, the Phantom 4 multispectral version from DJI, the 9-channel multispectral camera from SILIOS, and the SURVEY series multispectral cameras from MAPIR. However, remote sensing UAVs using area array imaging sensors experience high resource consumption and time costs in data transmission, storage, and processing due to the large number of pixels. According to reports, typical low-altitude remote sensing using area array imaging UAVs, with a flight altitude of 200 meters and a 75% overlap in both lateral and forward directions, requires approximately 5,000 images to be acquired before stitching for remote sensing monitoring of 1,000 mu (approximately 67 hectares). If each image has a resolution of 1280×960, it would require approximately 12GB of storage space. Ordinary graphics workstations would need more than 5 hours to complete post-processing tasks such as calibration, stitching, and analysis, with a data analysis speed of only about 3.3 mu / minute. This is insufficient to meet the needs of modern agricultural production, which requires timely mapping and high-frequency use of remote sensing data.

[0006] Point sensors acquire a comprehensive information of the reflectance spectrum of ground objects within the field of view in a single transaction. Their advantages are that they have a small number of pixels, require less resources for post-processing and data utilization, and have high efficiency in acquiring information. They are often used for close-range canopy-scale nitrogen measurement. However, their disadvantage is that it is difficult to directly acquire high-resolution two-dimensional images.

[0007] Compared to point sensors, linear array sensors have a larger number of one-dimensional pixels, and a smaller total number of pixels compared to area arrays. They also have a higher frame rate and higher processing efficiency, allowing for the acquisition of high-resolution two-dimensional images using pushbroom imaging. Currently, they are mainly used for scanning static objects (such as images and text), transient detection of moving objects, and quality inspection. In the field of remote sensing, linear array sensors are widely used on satellite remote sensing platforms, such as IKONOS, QuickBird, SPOT1-4, and the HR camera on my country's CBERS-02B satellite and the panchromatic camera on the ZY-3 satellite. However, research and application of remote sensing technology based on linear array sensors in unmanned aerial vehicle (UAV) platforms are still relatively limited. This is mainly because factors such as UAV flight stability, fuselage vibration, rapid movement at extremely low altitudes, and instantaneous changes in illumination conditions adversely affect the image quality, projection registration, and data accuracy of linear array sensors. Summary of the Invention

[0008] The present invention aims to overcome at least one of the defects (deficiencies) of the prior art and proposes a linear array remote sensing system and its data processing method that integrates seed sensing, storage and computation. It has the beneficial effect of accurate and rapid response to multispectral information of ground objects, and can effectively reduce the resource consumption and time cost of remote sensing data transmission, storage and processing, shorten data post-processing time, and improve remote sensing efficiency, image quality, projection registration and data accuracy.

[0009] To achieve the objectives of this invention, the technical solution adopted is as follows:

[0010] This technical solution proposes an integrated linear array remote sensing system that combines sensing, storage, and computation. It employs linear array imaging technology with the movement of an external carrier for wide-swath scanning acquisition, thereby improving measurement efficiency. Furthermore, it offers superior data processing speed compared to area imaging remote sensing technology and better data dimensionality than in-situ single-point measurements of multispectral data, better meeting user needs. In addition, this solution uses real-time data computation, eliminating the need for post-processing image correction and stitching, thus significantly reducing post-processing time.

[0011] Specifically, a linear array remote sensing system integrating sensing, storage, and computing, mounted on an external carrier, is characterized by comprising a host system, several slave systems, a housing, and a standard reflector.

[0012] Furthermore, the host system includes a central processing unit, a data storage device, a high-precision pose module, and an incident light calibration module;

[0013] Furthermore, the compartment is mounted on an external carrier; the back of the compartment is used to install the host system; the front of the compartment includes several compartments, each of which can accommodate a slave system; the geometric center normal of the slave system is coaxial with the geometric center normal of the compartment.

[0014] Furthermore, the high-precision pose module is used to acquire the geographical location information and attitude information of the linear array remote sensing system when collecting remote sensing data during the movement of the external carrier, and transmit it to the data storage device.

[0015] Furthermore, the slave system includes a linear array imaging sensor and a coprocessor, which are connected to the host system via an electrical bus within the housing; the coprocessor communicates with the central processing unit of the host system; the central processing unit is used to send instructions to the coprocessor to control the linear array imaging sensor to acquire linear array remote sensing data of ground features, and transmit the data to the data storage device;

[0016] Furthermore, a standard reflector is used to calibrate the linear array imaging sensor;

[0017] Furthermore, the host system synchronously triggers the incident light calibration module to collect ambient light information and the slave system to collect linear array remote sensing data. The host system automatically adjusts the shooting parameters of the slave system in real time based on the ambient light information.

[0018] Furthermore, the central processing unit is also used to parse the linear array remote sensing data in the data storage device and combine it with geographic location information and attitude information to construct orthophoto images.

[0019] Furthermore, the linear array imaging sensor includes a focal length of An imaging lens and a A group of photodiodes arranged in a ×1 linear array with identical spectral response curves. The number of photodiodes represents the total number of pixels in the linear array imaging sensor; the geometric center of the photodiode group is set as the center pixel of the linear array imaging sensor; the center normal of the photodiode group coincides with the optical axis of the imaging lens; the field of view of the linear array imaging sensor is... The linear array imaging sensor is equipped with a narrowband filter; the data acquired by the linear array imaging sensor at one time is one frame of linear array remote sensing data, including... Each pixel; the photodiode groups in the slave systems of all compartments in the compartment are arranged in the same direction. The geometric center normal of the slave system is coaxial with the geometric center normal of the compartment.

[0020] Furthermore, the incident light calibration module includes several photoelectric sensors, an auxiliary processor, and a multi-channel optical module; the multi-channel optical module is equipped with several narrowband filters; the spectral response curves of the photoelectric sensors are consistent with those of the photodiode array; the optical characteristics of the linear array imaging sensors of the several slave systems are consistent with those of the narrowband filters of the multi-channel optical module. The geographic location information includes latitude coordinates. Longitude coordinates ,high and heading angle The attitude information includes yaw angle. Pitch angle and roll angle ;

[0021] Furthermore, the heading angle With true north as 0°, the yaw angle increases clockwise from 0° to 360°. With the direction of the external carrier's nose as 0°, the pitch angle increases clockwise from 0° to 360°. and roll angle All references are to the horizontal plane of the earth; the pitch angles are... The roll angle is positive at elevation angles and negative at depression angles; The value is positive when rolling to the left and negative when rolling to the right.

[0022] Furthermore, the data processing method includes a correction phase and a remote sensing operation phase;

[0023] The correction phase includes the following steps:

[0024] J1 Initial Exposure Time Calculation: The initial exposure time is calculated during the sensor calibration stage to ensure that the linear array remote sensing data does not overflow or saturate.

[0025] J2 Linear Array Remote Sensing System Calibration: Based on the initial exposure time, the optical attenuation characteristics of the linear array imaging sensor are analyzed to clarify the attenuation coefficient of each pixel in the linear array remote sensing data;

[0026] The remote sensing operation phase includes the following steps:

[0027] S1 Initial pose information acquisition: Acquire the initial geographical location and attitude information of the linear array remote sensing system;

[0028] S2 Real-time Exposure Time Calculation: Calculates the exposure time for acquiring the current frame of linear array remote sensing data;

[0029] S3 collects the current frame array remote sensing data and simultaneously acquires real-time geographic location information and attitude information;

[0030] S4 Calculate relative height: that is, the difference between the height of the geographic location information corresponding to the current frame linear array remote sensing data in S3 and the height in the initial geographic location information in S1;

[0031] S5 Calculates ground resolution: That is, calculates the ground resolution of the current frame linear array remote sensing data based on the relative height in S4, the real-time geographical location information and attitude information collected in S3;

[0032] S6 performs initial processing of the current frame linear array remote sensing data: that is, it corrects and calculates reflectance of the current frame linear array remote sensing data acquired in S3, and constructs the data after initial processing;

[0033] S7 Calculates the projection coordinates of the current frame linear array remote sensing data: that is, calculates the projection coordinates of all pixels in the current frame linear array remote sensing data, which are used to further construct the linear array remote sensing image.

[0034] S8 saves the processed linear array remote sensing data;

[0035] Steps S1-S8 constitute the process for processing one frame of linear array remote sensing data. In practical applications, the entire process needs to be repeated until the task is completed. Furthermore, steps J1 and S2 employ the same exposure time calculation method. Specifically, the host system automatically adjusts the shooting parameters of each slave system during the linear array remote sensing data acquisition process. These shooting parameters are the exposure time. The specific steps are as follows:

[0036] During the acquisition of linear array remote sensing data by a slave system, the exposure time of each slave system is determined based on the exposure time of the corresponding channel (narrowband filter with the same optical characteristics) of the incident light calibration module; wherein, the exposure time of the slave system is consistent with the exposure time of the incident light calibration module.

[0037] The steps for calculating the exposure time of any channel in the incident light calibration module are as follows:

[0038] S01: Assume the power supply voltage of the linear array remote sensing system is... U ;

[0039] S02: Output voltage of any spectral channel of the incident light calibration module (V) The calculation formula is:

[0040] (1)

[0041] in, The dark voltage is (V). The sensitivity of a photosensitive element to a specific wavelength ( ), The incident light irradiance ( ), Exposure time (s);

[0042] S03: Dark processing and pre-sampling are performed before the acquisition of remote sensing data from the ground object linear array. Subtract dark voltage Then, equation (1) can be simplified to:

[0043] (2)

[0044] S04: exist Reaching the power supply voltage U The time is determined, and thus the exposure time of the corresponding slave system is determined. Consistent;

[0045] S05: Since the spectral response curves of the photodiode group and the incident light calibration module are consistent, the output voltage of any photodiode in the corresponding channel of the slave system is... (V) is calculated as follows:

[0046] (3)

[0047] in, The reflected light irradiance ( ).

[0048] Furthermore, step J2 calibrates the linear array remote sensing system. Taking a slave system as an example, the specific steps are as follows:

[0049] J21: First, obtain and lock the exposure time of a certain slave system determined in step S04;

[0050] J22: Place the standard reflector horizontally under a clear sky, align the slave system vertically with the standard reflector, and adjust the position and height of the slave system to ensure that its field of view is completely covered by the standard reflector.

[0051] J23: A slave system continuously acquires linear array remote sensing data at fixed time intervals. A single frame of linear array remote sensing data acquired in a single session contains... indivual data;

[0052] J24: Based on the peak output voltage of the photodiode corresponding to the center pixel. The output voltage of the photodiodes corresponding to other pixels Calculate the attenuation coefficient corresponding to each pixel position The mathematical relationship is as follows:

[0053] (4)

[0054] in The pixel index is [0, ..., ...]. The pixel number can be calculated sequentially using equation (4). Corresponding attenuation coefficient ;

[0055] J25: Output voltage after all pixels are corrected It can be calculated as:

[0056] (5)

[0057] J26: When correcting a single frame of actually acquired linear array remote sensing data. It can be obtained through table lookup or through model training. The following relationship exists:

[0058] (6)

[0059] exist After the model training is complete, it can be identified by pixel index. Calculate the attenuation coefficient of all pixels in a frame of linear array remote sensing data. ;

[0060] J27: Other slave systems use steps J21-J26 to perform similar calculations to determine their respective attenuation coefficients.

[0061] Furthermore, the method for calculating the ground resolution in step S5 is as follows:

[0062] When the height of the collected data is At that time, the unit pixel represents the actual distance of a ground object to a single photodiode. for:

[0063] (7)

[0064] Furthermore, the specific method for initial data processing in step S6 is as follows:

[0065] The data acquired by the linear array remote sensing system includes reflectance data. When a slave system is acquiring linear array remote sensing data, the incident light calibration module records the incident light irradiance of the corresponding channel of that slave system as follows: According to equation (2), the ground object reflected light irradiance collected by the slave system is an array. ,Include One element, For the first The reflected light irradiance of each pixel is calculated according to equations (3) and (5), and the linear array reflectance array is used for the calculation. ,Include Each element can be calculated as:

[0066] (8)

[0067] in The pixel index is [0, ..., ...]. Let N be the sequence number (i.e., frame number, starting from 0) of the current slave system linear array remote sensing data acquisition. Then the result array of the initial data processing is... for:

[0068] (9)

[0069] In the formula, The longitude coordinates of the center pixel of the linear array remote sensing data collected when the sequence number is N; The latitude coordinates of the center pixel of the linear array remote sensing data collected when the sequence number is N; This is the reflectance array calculated according to formula (8) when the acquisition sequence number is N. The above method is for the initial processing of data for a single slave system; other slave systems (other bands) use the same method.

[0070] Further, calculating the projection coordinates of the current frame linear array remote sensing data in step S7 includes calculating the projection coordinates of the center pixel, left center pixel, right center pixel, left pixel, and right pixel of the current frame linear array remote sensing data (e.g., Figure 8 The left and right center pixels are the first pixel to the left and the first pixel to the right of the center pixel in the linear array remote sensing data, respectively; the left and right pixels are all pixels to the left and right of the left and right center pixels, respectively; the linear array remote sensing data establishes a local coordinate system with the external carrier head as the zero axis, clockwise from 0 to 360°, with 90° as the right and 270° as the left; the calculation method for the projected coordinates of the center pixel of the linear array remote sensing data of all slave systems is as follows:

[0071] S71. Assign a one-to-one correspondence between all slave systems and bays, and designate bay 1 as bay number 1, meaning the normal of the slave system 1 coincides with the normal of the antenna geometric center of the high-precision pose module. Let the projected coordinates of the high-precision pose module at the current frame of linear array remote sensing data acquisition be... ;

[0072] S72. The origin is set as the center projection coordinates of cabin number 1. M is the angle between any other cabin and cabin 1 (origin), and M is the number of the other cabin (i.e., slave system).

[0073] S73. Let the projected coordinates of any slave system outside slave system 1 in this local coordinate system be... ;

[0074] S74. Projected coordinates of the center pixel of the linear array remote sensing data in the current frame of slave system No. 1 The calculation method is as follows:

[0075] (10)

[0076] (11)

[0077] The projection coordinates of the center pixel of the linear array remote sensing data in the current frame of any other slave system The calculation method is as follows:

[0078] (12)

[0079] (13)

[0080] Furthermore, under any motion state of the external carrier, the left pixel projection coordinates of the linear array remote sensing data of the current frame measured by any slave system are... The formula for calculating ) is:

[0081] (14)

[0082] (15)

[0083] The right pixel projection coordinates of the linear array remote sensing data of the current frame measured by any slave system. The formula for calculating ) is:

[0084] (16)

[0085] (17)

[0086] The left center pixel projection coordinates of the linear array remote sensing data of the current frame measured by any slave system. The formula for calculating ) is:

[0087] (18)

[0088] (19)

[0089] The right center pixel projection coordinates of the linear array remote sensing data of the current frame measured by any slave system. The formula for calculating ) is:

[0090] (20)

[0091] (twenty one)

[0092] According to equations (12) to (13), the projection coordinates of the center pixel of the current frame linear array remote sensing dataset obtained by the slave system in all cabins can be calculated. Then, according to equations (14) to (21), the projection coordinates of all other pixels of the linear array remote sensing dataset obtained by all slave systems can be calculated. Finally, the linear array remote sensing data collected by each slave system are integrated and converted into a two-dimensional orthophoto map.

[0093] Furthermore, when the heading angle of the high-precision pose module... When an anomaly is detected, the heading angle can be calculated based on adjacent latitude and longitude coordinates. Local calculation; let the center projection coordinates of the previous sampling point in cabin 1 be ( The planar projection coordinates of the next sampling point are ( Then the azimuth of the previous sampling point The calculation is as follows:

[0094] (twenty two)

[0095] (twenty three)

[0096] when =0 and When >0, =0°;

[0097] when =0 and When <0, =180°;

[0098] when =0 and When >0, =90°;

[0099] when =0 and When <0, =270°;

[0100] when >0 and When >0, = arctan ( / );

[0101] when <0 and When >0, = 360° + arctan ( / );

[0102] when <0 and <0 or when >0 and When <0, =180° + arctan ( / ).

[0103] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0104] Compared to area array imaging sensors, this invention significantly reduces the amount of data, enabling rapid spectral data acquisition and planar mapping. Compared to point sensors, it has a larger number of one-dimensional pixels, allowing the scanning area to fully cover the area to be measured, effectively avoiding omissions in the measurement area. During multispectral data acquisition, it can obtain the resolution required for actual production while also achieving higher acquisition efficiency and rapid acquisition.

[0105] The standard reflector calibration method used in this invention can correct the cosine-like effect that may be caused by aging or contamination of photosensitive semiconductor devices or optical defects in the lens itself, thus achieving accurate response of each pixel (photosensitive semiconductor photodiode device) to actual illumination.

[0106] This invention primarily utilizes a host system to synchronously trigger a slave system for accurate linear array remote sensing data acquisition. Simultaneously, the slave system's custom bands (narrowband filters) support the inversion and calculation of multiple vegetation indices and also support single-band reflectance data output. Furthermore, the proposed linear array remote sensing data correction and projection coordinate calculation method effectively reduces the resource consumption and time costs associated with remote sensing data transmission, storage, and processing. It enables on-board online calculations, achieving integrated sensing, storage, and computation, shortening data post-processing time, and improving remote sensing efficiency, image quality, projection registration, and data accuracy. Attached Figure Description

[0107] Figure 1 A schematic diagram of a linear array remote sensing system integrating sensing, storage, and computing mounted on a dual-rotor UAV.

[0108] Figure 2 This is a schematic diagram of the bottom of a five-band integrated linear array remote sensing system.

[0109] Figure 3 This is a schematic diagram of the top of a five-band integrated linear array remote sensing system.

[0110] Figure 4 This is a flowchart of linear array remote sensing data processing.

[0111] Figure 5 This is a schematic diagram of the pitch and roll angles of the slave system.

[0112] Figure 6 This is a schematic diagram of linear array remote sensing data correction.

[0113] Figure 7 Flowchart for determining the exposure time for the incident light calibration module and slave system.

[0114] Figure 8 This diagram illustrates the relationship between the left-side pixel, the left-center pixel, the right-side pixel, and the right-center pixel (using an even number of pixels as an example). Detailed Implementation

[0115] The accompanying drawings are for illustrative purposes only and should not be construed as limiting the invention. To better illustrate the following embodiments, some parts in the drawings may be omitted, enlarged, or reduced, and do not represent the actual product dimensions; it is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0116] Example

[0117] like Figure 1 , Figure 2 and Figure 3 As shown, this embodiment discloses a five-band integrated linear array remote sensing system that integrates sensing, storage, and computing, mounted on a dual-rotor UAV, including a main unit system 13, slave units (1, 6, 7, 8, 9), a housing 5, and a standard reflector.

[0118] The host system includes a central processing unit 12, a data storage device 10, a high-precision pose module 2, and an incident light calibration module 11, which are used to control the slave system to acquire linear array remote sensing data.

[0119] The housing is mounted on an external carrier; the back of the housing is used to install the host system; the front of the housing includes 6 compartments, and each compartment can accommodate and connect to a slave system. The compartments connect the host system 13 and the slave systems (1, 6, 7, 8, 9) via an SPI electrical bus. Specifically, each compartment is marked with a specific serial number. In this embodiment, the compartment corresponding to the red band slave system is designated as compartment 1, and the coordinate center of compartment 1 (i.e., the optical axis of the distortion-free imaging lens) coincides with the normal of the antenna geometric center of the high-precision pose module 2.

[0120] The high-precision pose module is used to acquire the slave system's geographical location information (latitude coordinates) when the linear array imaging sensor collects remote sensing data during the movement of the external carrier. Longitude coordinates ,high and heading angle ) and cabin attitude information (yaw angle) Pitch angle and roll angle ), and transmit it to the data storage device;

[0121] The slave system includes a linear array imaging sensor and a coprocessor, which are connected to the host system via an electrical bus inside the housing; the coprocessor communicates with the central processing unit of the host system; the central processing unit is used to send instructions to the coprocessor to control the linear array imaging sensor to perform linear array remote sensing data acquisition of ground features and transmit it to the data storage device.

[0122] like Figure 6 As shown, the slave system uses a standard reflector to calibrate the linear imaging sensor, and combines the incident light calibration module to calculate the linear array reflectivity data. Then, the calculated linear array reflectivity data is transmitted to the host system, and finally to the data storage device.

[0123] The host system synchronously triggers the incident light calibration module to collect ambient light information and the linear array imaging sensor to collect linear array remote sensing data. The host system automatically and in real time adjusts the shooting parameters of the slave system according to the ambient light information.

[0124] The central processing unit is also used to parse linear array remote sensing data in the data storage device and combine it with geographic location information and attitude information to construct orthophoto images.

[0125] like Figure 2 As shown, specifically, the linear array imaging sensor is equipped with five bands: red, green, blue, near-infrared, and red-edge. These bands are mainly related to the vegetation index; the power supply voltage throughout the process... U Set to 3.3V.

[0126] Furthermore, in this embodiment, the linear array imaging sensor includes a focal length... For a 6.00mm distortion-free imaging lens and a A photodiode array with high sensitivity, ultra-low dark current, and identical spectral response curves arranged in a ×1 linear array. Among them, The number of photodiodes represents the total number of pixels in the linear array imaging sensor; the geometric center of the photodiode group is set as the center pixel of the linear array imaging sensor; the center normal of the photodiode group coincides with the optical axis of the distortion-free imaging lens; the field of view of the linear array imaging sensor is... The linear array imaging sensor is equipped with a narrowband filter; the data acquired by the linear array imaging sensor at one time is one frame of linear array remote sensing data, including... Each pixel; the photodiode groups in the slave systems of all compartments in the compartment are arranged in the same direction.

[0127] The incident light calibration module includes several photoelectric sensors, an auxiliary processor, and a multi-channel optical module; the multi-channel optical module is equipped with several narrowband filters; the spectral response curves of the photoelectric sensors are consistent with those of the photodiode array; the optical characteristics of the linear array imaging sensors of the several slave systems are consistent with those of the narrowband filters of the multi-channel optical module; each channel of the linear array imaging sensor and the incident light calibration module is equipped with narrowband filters of the same wavelength; the photodiode arrays of the slave systems in all compartments of the compartment are arranged in the same direction;

[0128] Specifically, the working mode of this acquisition system is line scanning. The data sampling method of this acquisition system can be set to timed acquisition and equidistant acquisition through the time information and latitude and longitude information of the high-precision pose module 2 in the host system. In this embodiment, it is set to timed acquisition at 20 Hz.

[0129] Furthermore, in this embodiment, the heading angle With true north as 0°, the yaw angle increases clockwise from 0° to 360°. With the direction of the external carrier's nose as 0°, the pitch angle increases clockwise from 0° to 360°. and roll angle All references are to the horizontal plane of the earth; the pitch angles are... The roll angle is positive at elevation angles and negative at depression angles; The value is positive when rolling to the left and negative when rolling to the right (e.g., ...). Figure 5 (As shown).

[0130] Furthermore, this embodiment also discloses an integrated linear array remote sensing data processing method (such as...). Figure 4 (As shown). The data processing method includes a correction phase and a remote sensing operation phase;

[0131] The correction phase includes the following steps:

[0132] J1 Initial Exposure Time Calculation: The initial exposure time is calculated during the sensor calibration stage to ensure that the linear array remote sensing data does not overflow or saturate.

[0133] J2 Linear Array Remote Sensing System Calibration: Based on the initial exposure time, the optical attenuation characteristics of the linear array imaging sensor are analyzed to clarify the attenuation coefficient of each pixel in the linear array remote sensing data;

[0134] The remote sensing operation phase includes the following steps:

[0135] S1 Initial pose information acquisition: Acquire the initial geographical location and attitude information of the linear array remote sensing system;

[0136] S2 Real-time Exposure Time Calculation: Calculates the exposure time for acquiring the current frame of linear array remote sensing data;

[0137] S3 collects the current frame array remote sensing data and simultaneously acquires real-time geographic location information and attitude information;

[0138] S4 Calculate relative height: that is, the difference between the height of the geographic location information corresponding to the current frame linear array remote sensing data in S3 and the height in the initial geographic location information in S1;

[0139] S5 Calculates ground resolution: That is, calculates the ground resolution of the current frame linear array remote sensing data based on the relative height in S4, the real-time geographical location information and attitude information collected in S3;

[0140] S6 performs initial processing of the current frame linear array remote sensing data: that is, it corrects and calculates reflectance of the current frame linear array remote sensing data acquired in S3, and constructs the data after initial processing;

[0141] S7 Calculates the projection coordinates of the current frame linear array remote sensing data: that is, calculates the projection coordinates of all pixels in the current frame linear array remote sensing data, which are used to further construct the linear array remote sensing image.

[0142] S8 saves the processed linear array remote sensing data;

[0143] Steps S1-S8 constitute the process of processing one frame of linear array remote sensing data. In practical applications, the entire process needs to be repeated until the task is completed.

[0144] Steps J1 and S2 use the same exposure time calculation method. Specifically, the host system automatically adjusts the shooting parameters of each slave system during the online array remote sensing data acquisition process. The shooting parameters are the exposure time. The specific steps are as follows:

[0145] During the acquisition of linear array remote sensing data by a slave system, the exposure time of the slave system is determined according to the corresponding channel of the incident light calibration module;

[0146] The exposure time calculation steps for any channel of the incident light calibration module (e.g.) Figure 7 )for:

[0147] S01: Assume the power supply voltage of the linear array remote sensing system is... U ;

[0148] S02: Output voltage of any spectral channel of the incident light calibration module (V) The calculation formula is:

[0149] (1)

[0150] in, The dark voltage is (V). The sensitivity of a photosensitive element to a specific wavelength ( ), The incident light irradiance ( ), Exposure time (s);

[0151] S03: Dark processing and pre-sampling are performed before the acquisition of remote sensing data from the ground object linear array. Subtract dark voltage Then, equation (1) can be simplified to:

[0152] (2)

[0153] S04: exist Reaching the power supply voltage U The time is determined, and thus the exposure time of the corresponding slave system is determined. Consistent;

[0154] S05: Since the spectral response curves of the photodiode group in the slave system are consistent with those of the photoelectric sensor in the incident light calibration module, the output voltage of any photodiode in the corresponding channel of the slave system is... (V) is calculated as follows:

[0155] (3)

[0156] in, The reflected light irradiance ( ).

[0157] Furthermore, step J2 calibrates the linear array remote sensing system. Taking a slave system as an example, the specific steps are as follows:

[0158] J21: First, obtain and lock the exposure time of a certain slave system determined in step S04;

[0159] J22: Place the standard reflector horizontally under a clear sky, align the slave system vertically with the standard reflector, and adjust the position and height of the slave system to ensure that its field of view is completely covered by the standard reflector.

[0160] J23: Using this slave system, linear array remote sensing data is continuously acquired at fixed time intervals. A single frame of linear array remote sensing data acquired in a single session contains... indivual data;

[0161] J24: Based on the peak output voltage of the photodiode corresponding to the center pixel. The output voltage of the photodiodes corresponding to other pixels Calculate the attenuation coefficient corresponding to each pixel position The mathematical relationship is as follows:

[0162] (4)

[0163] in The pixel index is [0, ..., ...]. The pixel number can be calculated sequentially using equation (4). Corresponding attenuation coefficient ;

[0164] J25: Output voltage after all pixels are corrected It can be calculated as:

[0165] (5)

[0166] J26: When correcting a single frame of actually acquired linear array remote sensing data. It can be obtained through table lookup or through model training. The following relationship exists:

[0167] (6)

[0168] exist After the model training is complete, it can be identified by pixel index. Calculate the attenuation coefficient of all pixels in a frame of linear array remote sensing data. ;

[0169] J27: Other slave systems use steps J21-J26 to perform similar calculations to determine their respective attenuation coefficients.

[0170] Furthermore, the ground resolution calculation in step S5 is performed as follows:

[0171] When the height of the collected data is At that time, the actual distance of a ground object corresponding to a unit pixel (i.e., a single photodiode) for:

[0172] (7)

[0173] Furthermore, the initial data processing in step S6 is specifically carried out as follows:

[0174] The data acquired by the linear array remote sensing system includes reflectance data. When a slave system is acquiring linear array remote sensing data, the incident light calibration module records the incident light irradiance of the corresponding channel of that slave system as follows: According to equation (2), the ground object reflected light irradiance collected by the slave system is an array. ,Include One element, For the first The reflected light irradiance of each pixel is calculated according to equations (3) and (5), and the linear array reflectance array is used for the calculation. ,Include Each element can be calculated as:

[0175] (8)

[0176] in The pixel index is [0, ..., ...]. Let N be the sequence number (i.e., frame number, starting from 0) of the current slave system linear array remote sensing data acquisition. Then the result array of the initial data processing is... for:

[0177] (9)

[0178] In the formula, The longitude coordinates of the center pixel of the linear array remote sensing data collected when the sequence number is N; The latitude coordinates of the center pixel of the linear array remote sensing data collected when the sequence number is N; This is the reflectance array calculated according to formula (8) when the acquisition sequence number is N. The above method is for the initial processing of data for a single slave system; other slave systems (other bands) use the same method.

[0179] Further, step S7 calculates the projection coordinates of the current frame linear array remote sensing data, including calculating the projection coordinates of the center pixel, left center pixel, right center pixel, left side pixel, and right side pixel of the current frame linear array remote sensing data; the left center pixel and right center pixel are respectively the first pixel to the left and the first pixel to the right of the center pixel of the linear array remote sensing data; the left side pixel and right side pixel are respectively all pixels to the left of the left center pixel and to the right of the right center pixel of the linear array remote sensing data; the linear array remote sensing data establishes a local coordinate system with the external carrier head as the zero axis, clockwise from 0 to 360°, with 90° as the right and 270° as the left; the calculation method for the projection coordinates of the center pixel of the linear array remote sensing data of all slave systems is as follows:

[0180] S71. Assign a one-to-one correspondence between all slave systems and bays, and designate bay 1 as bay number 1, meaning the normal of the slave system 1 coincides with the normal of the antenna geometric center of the high-precision pose module. Let the projected coordinates of the high-precision pose module at the current frame of linear array remote sensing data acquisition be... ;

[0181] S72. The origin is set as the center projection coordinates of cabin number 1. M is the angle between any other cabin and cabin 1 (origin), and M is the number of the other cabin (i.e., slave system).

[0182] S73. Let the projected coordinates of any slave system outside slave system 1 in this local coordinate system be... ;

[0183] S74. Projected coordinates of the center pixel of the linear array remote sensing data in the current frame of slave system No. 1 The calculation method is as follows:

[0184] (10)

[0185] (11)

[0186] The projection coordinates of the center pixel of the linear array remote sensing data in the current frame of any other slave system The calculation method is as follows:

[0187] (12)

[0188] (13)

[0189] Furthermore, under any motion state of the external carrier, the left pixel projection coordinates of the linear array remote sensing data of the current frame measured by any slave system are... The formula for calculating ) is:

[0190] (14)

[0191] (15)

[0192] The right pixel projection coordinates of the linear array remote sensing data of the current frame measured by any slave system. The formula for calculating ) is:

[0193] (16)

[0194] (17)

[0195] The left center pixel projection coordinates of the linear array remote sensing data of the current frame measured by any slave system. The formula for calculating ) is:

[0196] (18)

[0197] (19)

[0198] The right center pixel projection coordinates of the linear array remote sensing data of the current frame measured by any slave system. The formula for calculating ) is:

[0199] (20)

[0200] (twenty one)

[0201] According to equations (12) to (13), the projection coordinates of the center pixel of the current frame linear array remote sensing dataset obtained by the slave system in all cabins can be calculated. Then, according to equations (14) to (21), the projection coordinates of all other pixels of the linear array remote sensing dataset obtained by all slave systems can be calculated. Finally, the linear array remote sensing data collected by each slave system are integrated and converted into a two-dimensional orthophoto map.

[0202] Furthermore, when the heading angle of the high-precision pose module... When an anomaly is detected, the heading angle is calculated based on adjacent latitude and longitude coordinates. For local calculations, let the center projection coordinates of the sampling point preceding cabin 1 be ( The planar projection coordinates of the next sampling point are ( Then the azimuth of the previous sampling point The calculation is as follows:

[0203] (twenty two)

[0204] (twenty three)

[0205] when =0 and When >0, =0°;

[0206] when =0 and When <0, =180°;

[0207] when =0 and When >0, =90°;

[0208] when =0 and When <0, =270°;

[0209] when >0 and When >0, = arctan ( / );

[0210] when <0 and When >0, = 360° + arctan ( / );

[0211] when <0 and <0 or when >0 and When <0, =180° + arctan ( / ).

[0212] Obviously, the above embodiments of the present invention are merely examples to clearly illustrate the technical solutions of the present invention, and are not intended to limit the specific implementation of the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the claims of the present invention should be included within the protection scope of the claims of the present invention.

Claims

1. A method for processing linear array remote sensing data based on an integrated sensing, storage, and computing linear array remote sensing system, wherein the integrated sensing, storage, and computing linear array remote sensing system is mounted on an external carrier, characterized in that... Includes the main system, several slave systems, a housing, and a standard reflector; The host system includes a central processing unit, a data storage device, a high-precision pose module, and an incident light calibration module. The compartment is mounted on an external carrier; the back of the compartment is used to install the host system; the front of the compartment includes several compartments, each of which can accommodate a slave system. The high-precision pose module is used to acquire the geographical location information and pose information of the linear array remote sensing system when collecting data during the movement of the external carrier, and transmit it to the data storage device. The slave system includes a linear array imaging sensor and a coprocessor, which are connected to the host system via an electrical bus inside the housing; the coprocessor communicates with the central processing unit of the host system; the central processing unit is used to send instructions to the coprocessor to control the linear array imaging sensor to perform linear array remote sensing data acquisition of ground features and transmit it to the data storage device. The standard reflector is used to calibrate the linear array imaging sensor; The host system synchronously triggers the incident light calibration module to collect ambient light information and the slave system to collect linear array remote sensing data. The host system automatically adjusts the shooting parameters of the slave system in real time according to the ambient light information. The central processing unit is also used to parse linear array remote sensing data in the data storage device and combine geographic location information and attitude information to construct orthophoto remote sensing images. The data processing method includes a correction phase and a remote sensing operation phase; The correction phase includes the following steps: J1 Initial Exposure Time Calculation: The initial exposure time is calculated during the sensor calibration stage to ensure that the linear array remote sensing data does not overflow or saturate. J2 Linear Array Remote Sensing System Calibration: Based on the initial exposure time, the optical attenuation characteristics of the linear array imaging sensor are analyzed to clarify the attenuation coefficient of each pixel in the linear array remote sensing data; The remote sensing operation phase includes the following steps: S1 Initial pose information acquisition: Acquire the initial geographical location and attitude information of the linear array remote sensing system; S2 Real-time Exposure Time Calculation: Calculates the exposure time for acquiring the current frame of linear array remote sensing data; S3 collects the current frame array remote sensing data and simultaneously acquires real-time geographic location information and attitude information; S4 Calculate relative height: that is, the difference between the height of the geographic location information corresponding to the current frame linear array remote sensing data in S3 and the height in the initial geographic location information in S1; S5 Calculates ground resolution: That is, calculates the ground resolution of the current frame linear array remote sensing data based on the relative height in S4, the real-time geographical location information and attitude information collected in S3; S6 performs initial processing of the current frame linear array remote sensing data: that is, it corrects and calculates reflectance of the current frame linear array remote sensing data acquired in S3, and constructs the data after initial processing; S7 Calculates the projection coordinates of the current frame linear array remote sensing data: that is, calculates the projection coordinates of all pixels in the current frame linear array remote sensing data, which are used to further construct the linear array remote sensing image. S8 saves the processed linear array remote sensing data; Steps S1-S8 constitute the process of processing one frame of linear array remote sensing data. In practical applications, the entire process needs to be repeated until the task is completed. Steps J1 and S2 use the same exposure time calculation method, specifically, the host system automatically adjusts the exposure time of each slave system during the online array remote sensing data acquisition process. The specific steps are as follows: During the acquisition of linear array remote sensing data by a slave system, the exposure time of the slave system is determined according to the corresponding channel of the incident light calibration module; The steps for calculating the exposure time of any channel in the incident light calibration module are as follows: S01: Assume the power supply voltage of the linear array remote sensing system is... U ; S02: Output voltage of any spectral channel of the incident light calibration module The formula for calculating V is: （1） in, The dark voltage is (V). Sensitivity of a photosensitive element for a specific wavelength, expressed in units of 1000 nm. , The incident light irradiance is expressed in units of . , Exposure time, in seconds; S03: Dark processing and pre-sampling are performed before the acquisition of remote sensing data from the ground object linear array. Subtract dark voltage Then, equation (1) can be simplified to: （2） S04: exist Reaching the power supply voltage U The time is determined, and thus the exposure time of the corresponding slave system is determined. Consistent; S05: Since the spectral response curves of the photodiode group in the slave system are consistent with those of the photoelectric sensor in the incident light calibration module, the output voltage of any photodiode in the corresponding channel of the slave system is... The calculation, in units of V, is as follows: （3） in, Reflected radiant illuminance, in units of . .

2. The method for processing linear array remote sensing data based on an integrated sensing, storage, and computing linear array remote sensing system according to claim 1, characterized in that, Step J2 calibrates the linear array remote sensing system, and the specific steps are as follows: J21: First, obtain and lock the exposure time of a certain slave system determined in step S04; J22: Place the standard reflector horizontally under a clear sky, align the slave system vertically with the standard reflector, and adjust the position and height of the slave system to ensure that its field of view is completely covered by the standard reflector. J23: Using this slave system, linear array remote sensing data is continuously acquired at fixed time intervals. A single frame of linear array remote sensing data acquired in a single session contains... indivual data; J24: Based on the peak output voltage of the photodiode corresponding to the center pixel. The output voltage of the photodiodes corresponding to other pixels Calculate the attenuation coefficient corresponding to each pixel position The mathematical relationship is as follows: （4） in The pixel index is [0, ..., ...]. The pixel number can be calculated sequentially using equation (4). Corresponding attenuation coefficient ; J25: Output voltage after all pixels are corrected It can be calculated as: （5） J26: When correcting a single frame of actually acquired linear array remote sensing data. It can be obtained through table lookup or through model training. The following relationship exists: （6） exist After the model training is complete, it can be identified by pixel index. Calculate the attenuation coefficient of all pixels in a frame of linear array remote sensing data. ; J27: Other slave systems use similar calculations, following steps J21-J26, to determine their respective attenuation coefficients.

3. The linear array remote sensing data processing method based on an integrated sensing, storage, and computing linear array remote sensing system according to claim 1, characterized in that, The ground resolution calculation in step S5 is performed as follows: When the height of the collected data is At that time, the unit pixel represents the actual distance of a ground object to a single photodiode. for: （7）。 4. The method for processing linear array remote sensing data based on an integrated sensing, storage, and computing linear array remote sensing system according to claim 1, characterized in that, The initial data processing in step S6 is as follows: The data acquired by the linear array remote sensing system includes reflectance data. When a slave system is acquiring linear array remote sensing data, the incident light calibration module records the incident light irradiance of the corresponding channel of that slave system as follows: According to equation (2), the ground object reflected light irradiance collected by the slave system is an array. ,Include One element, For the first The reflected light irradiance of each pixel is calculated according to equations (3) and (5), and the linear array reflectance array is used. ,Include Each element can be calculated as: （8） in The pixel index is [0, ..., ...]. Let N be the sequence number of the current slave system's linear array remote sensing data acquisition, i.e., the frame number, starting from 0. Then the result array of the initial data processing is... for: （9） In the formula, The longitude coordinates of the center pixel of the linear array remote sensing data collected when the sequence number is N; The latitude coordinates of the center pixel of the linear array remote sensing data collected when the sequence number is N; This is the reflectance array calculated according to formula (8) when the acquisition sequence number is N. The above method is for the initial data processing of a single slave system; other slave systems (other bands) use the same method.

5. The linear array remote sensing data processing method based on an integrated sensing, storage, and computing linear array remote sensing system according to claim 1, characterized in that, Step S7 calculates the projection coordinates of the current frame linear array remote sensing data, including calculating the projection coordinates of the center pixel, left center pixel, right center pixel, left side pixel, and right side pixel of the current frame linear array remote sensing data; the left center pixel and right center pixel are the first pixel to the left and the first pixel to the right of the center pixel of the linear array remote sensing data, respectively; the left side pixel and right side pixel are all pixels to the left of the left center pixel and the right side of the right center pixel of the linear array remote sensing data, respectively; the linear array remote sensing data establishes a local coordinate system with the external carrier head as the zero axis, clockwise from 0 to 360°, with 90° as the right and 270° as the left; the calculation method for the projection coordinates of the center pixel of the linear array remote sensing data of all slave systems is as follows: S71. Assign a one-to-one correspondence between all slave systems and bays, and designate bay 1 as bay number 1, meaning the normal of the slave system 1 coincides with the normal of the antenna geometric center of the high-precision pose module. Let the projected coordinates of the high-precision pose module at the current frame of linear array remote sensing data acquisition be... ; S72. The origin is set as the center projection coordinates of cabin number 1. M is the angle between any other cabin and cabin 1 (origin), and M is the number of the other cabin (i.e., slave system). S73. Let the projected coordinates of any slave system outside slave system 1 in this local coordinate system be... ; S74. Projected coordinates of the center pixel of the linear array remote sensing data in the current frame of slave system No. 1 The calculation method is as follows: （10） （11） The projection coordinates of the center pixel of the linear array remote sensing data in the current frame of any other slave system The calculation method is as follows: （12） （13） Furthermore, under any motion state of the external carrier, the left pixel projection coordinates of the linear array remote sensing data of the current frame measured by any slave system are... The formula for calculating ) is: （14） （15） The right pixel projection coordinates of the linear array remote sensing data of the current frame measured by any slave system. The formula for calculating ) is: （16） （17） The left center pixel projection coordinates of the linear array remote sensing data of the current frame measured by any slave system. The formula for calculating ) is: （18） （19） The right center pixel projection coordinates of the linear array remote sensing data of the current frame measured by any slave system. The formula for calculating ) is: （20） （21） According to equations (12) to (13), the projection coordinates of the center pixel of the current frame linear array remote sensing dataset obtained by the slave system in all cabins can be calculated. Then, according to equations (14) to (21), the projection coordinates of all other pixels of the linear array remote sensing dataset obtained by all slave systems can be calculated. Finally, the linear array remote sensing data collected by each slave system are integrated and converted into a two-dimensional orthophoto map.

6. A method for processing linear array remote sensing data in a linear array remote sensing system based on integrated sensing, storage, and computing, as described in claim 4, wherein the heading angle of the high-precision pose module... When an anomaly is detected, the heading angle can be calculated based on adjacent latitude and longitude coordinates. Local calculation; let the center projection coordinates of the previous sampling point in cabin 1 be ( The planar projection coordinates of the next sampling point are ( Then the azimuth of the previous sampling point The calculation is as follows: （22） （23） when =0 and When >0, =0°; when =0 and When <0, =180°; when =0 and When >0, =90°; when =0 and When <0, =270°; when >0 and When >0, = arctan ( / ); when <0 and When >0, = 360° + arctan ( / ); when <0 and <0 or when >0 and When <0, =180° + arctan ( / ).

7. The system for processing linear array remote sensing data based on an integrated sensing, storage, and computing linear array remote sensing system according to claim 1, characterized in that, The linear array imaging sensor includes a focal length of An imaging lens and a A group of photodiodes arranged in a ×1 linear array with identical spectral response curves; wherein, The number of photodiodes represents the total number of pixels in the linear array imaging sensor; the geometric center of the photodiode group is set as the center pixel of the linear array imaging sensor; the center normal of the photodiode group coincides with the optical axis of the imaging lens; the field of view of the linear array imaging sensor is... The linear array imaging sensor is equipped with a narrowband filter; the data acquired by the linear array imaging sensor at one time is one frame of linear array remote sensing data, including... Each pixel; the photodiode groups in the slave systems of all compartments in the compartment are arranged in the same direction; the geometric center normal of the slave system is coaxial with the geometric center normal of the compartment.

8. The system for processing linear array remote sensing data based on an integrated sensing, storage, and computing linear array remote sensing system according to claim 7, characterized in that, The incident light calibration module includes several photoelectric sensors, an auxiliary processor, and a multi-channel optical module; the multi-channel optical module is equipped with several narrowband filters; the spectral response curves of the photoelectric sensors are consistent with those of the photodiode array; the optical characteristics of the linear array imaging sensors of the several slave systems are consistent with those of the narrowband filters of the multi-channel optical module; the geographic location information includes latitude coordinates. Longitude coordinates ,high and heading angle The attitude information includes yaw angle. Pitch angle and roll angle The heading angle With true north as 0°, the yaw angle increases clockwise from 0° to 360°. With the direction of the external carrier's nose as 0°, the pitch angle increases clockwise from 0° to 360°. and roll angle All references are to the horizontal plane of the earth; the pitch angles are... The roll angle is positive at elevation angles and negative at depression angles; The value is positive when rolling to the left and negative when rolling to the right.

Images