Lifesaving equipment terminal guidance system and guidance method based on thermal imaging dynamic threshold analysis

Abstract

The invention discloses a lifesaving equipment terminal guidance system and method based on thermal imaging dynamic threshold analysis. The system comprises a life buoy, and a propeller is arranged at the rear end of the life buoy; the life buoy is provided with a thermal imaging module and an embedded processing unit, and the side surface of the thermal imaging module is respectively provided with three thermal imaging sensors; the embedded processing unit comprises a temperature quantification FPGA module, a vertical projection positioning module and a motion control module, the temperature quantification FPGA module carries out discretization grading on a pixel-level temperature difference based on a frame average temperature and a dynamic step length, and the vertical projection positioning module carries out column integral operation on a quantized temperature matrix; the motion control module generates a control signal according to the azimuth angle to drive the propeller to adjust the course of the life buoy. According to the invention, the defects of the prior art are overcome, and rapid identification and positioning of the thermal abnormal point are realized by combining the thermal imaging technology, the dynamic temperature quantification processing of the temperature quantification FPGA module and the vertical projection positioning of the vertical projection positioning module.

Description

Technical Field

[0001] The present invention relates to the technical field of surface vehicles, and in particular to a life-saving equipment terminal guidance system and guidance method based on thermal imaging dynamic threshold analysis. Background Art

[0002] Current marine search and rescue efforts rely primarily on visual observation, radar, or drone inspections. However, their efficiency plummets in low-visibility environments, such as at night, in fog, and under the influence of waves. Existing thermal imaging lifesaving equipment often utilizes a single, forward-facing thermal imager. However, these cameras only cover a conical area directly in front of them (typical field of view: 60° × 45°), making them unable to detect targets to the sides or rear. Furthermore, since forward-facing thermal imagers are typically fixed to the hull or lifebuoy, scanning relies on the carrier's motion, resulting in inefficient search paths. Furthermore, because water surface temperature is easily affected by ambient lighting, wind speed, and wave reflections, thermal imaging equipment suffers from low recognition accuracy in complex water environments, making it difficult to quickly lock onto targets. Summary of the Invention

[0003] In response to the shortcomings of the existing technology, the present invention provides a life-saving equipment terminal guidance system and guidance method based on thermal imaging dynamic threshold analysis, which overcomes the shortcomings of the existing technology. By combining thermal imaging technology, dynamic temperature quantization processing of the temperature quantification FPGA module and vertical projection positioning of the vertical projection positioning module, it realizes the rapid identification and positioning of thermal anomalies.

[0004] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0005] A lifesaving equipment terminal guidance system based on thermal imaging dynamic threshold analysis includes a lifebuoy with thrusters fixedly installed on both sides of the rear end of the lifebuoy;

[0006] A thermal imaging module is fixedly mounted on the front end of the upper surface of the lifebuoy. The horizontal cross-section of the thermal imaging module is a trapezoidal structure. The front side, left side, and right side of the thermal imaging module are respectively fixedly mounted with a first thermal imaging sensor, a second thermal imaging sensor, and a third thermal imaging sensor.

[0007] An embedded processing unit is fixedly installed on the lifebuoy, and the embedded processing unit includes a temperature quantification FPGA module, a vertical projection positioning module and a motion control module. The signal output ends of the first thermal imaging sensor, the second thermal imaging sensor and the third thermal imaging sensor are connected to the signal input end of the temperature quantification FPGA module, the signal output end of the temperature quantification FPGA module is connected to the signal input end of the vertical projection positioning module, the signal output end of the vertical projection positioning module is connected to the signal input end of the motion control module, and the signal output ends of the motion control module are respectively connected to the driving circuit of the thruster;

[0008] The temperature quantization FPGA module discretizes and grades the pixel-level temperature difference based on the frame average temperature and dynamic step size. The vertical projection positioning module performs column-wise integration operations on the quantized temperature matrix, extracts the azimuth corresponding to the maximum integral value as the target orientation, and the motion control module generates a control signal based on the azimuth to drive the thruster to adjust the lifebuoy's heading to achieve precise target tracking.

[0009] Preferably, the projections on the horizontal plane between the first thermal imaging sensor and the second thermal imaging sensor, and between the second thermal imaging sensor and the third thermal imaging sensor are respectively distributed at an angle of 60 degrees, the fields of view of the first thermal imaging sensor, the second thermal imaging sensor and the third thermal imaging sensor are respectively distributed at a horizontal angle of 55 degrees and a vertical angle of 35 degrees, and there is no overlapping area between adjacent first thermal imaging sensors and second thermal imaging sensors, and between adjacent second thermal imaging sensors and third thermal imaging sensors.

[0010] Preferably, the installation pitch angles of the first thermal imaging sensor, the second thermal imaging sensor and the third thermal imaging sensor are All meet the following requirements:

[0011]

[0012] Where h is the installation height of the thermal imaging module, d is the nominal detection distance, is the dynamic compensation angle.

[0013] Preferably, the temperature quantization FPGA module discretizes and grades the pixel-level temperature difference based on the frame average temperature and the dynamic step size, specifically comprising:

[0014] Calculate the average temperature value of all pixels in the current frame:

[0015]

[0016] The pixel temperature difference is quantified according to the following formula:

[0017]

[0018] Among them, Q is the quantized value of thermal anomaly intensity at pixel level, is the current pixel temperature value; is the environmental adaptive step size factor; is the ambient noise tolerance.

[0019] Preferably, the environment adaptive step factor The method of determining is:

[0020]

[0021] in, is the temperature standard deviation, is the ambient temperature, 、 is the adjustment coefficient, .

[0022] The present invention also discloses a guidance method based on the above-mentioned life-saving equipment terminal guidance system, comprising the following steps:

[0023] Step S1: Initialize the spiral search radius R, and use the spiral expansion search strategy to control the propeller to drive the lifebuoy to sail in a circle;

[0024] Step S2: Synchronously collect thermal imaging data of the target area through the first thermal imaging sensor, the second thermal imaging sensor, and the third thermal imaging sensor.

[0025] Step S3: using the temperature quantization FPGA module to independently perform dynamic temperature quantization processing on the thermal imaging data collected by the first thermal imaging sensor, the second thermal imaging sensor, and the third thermal imaging sensor;

[0026] Step S4: performing a column-wise integration operation on the quantized temperature matrix through a vertical projection positioning module to generate a one-dimensional array;

[0027] Step S5: extracting the maximum value of the feature vectors corresponding to the first thermal imaging sensor, the second thermal imaging sensor, and the third thermal imaging sensor respectively. If the maximum value exceeds a threshold, outputting the corresponding position ID;

[0028] Step S6: When the position ID is valid, the lifebuoy leaves the circling mode and adjusts the operating power of the two propellers through the motion control module according to the position ID to adjust the direction of the lifebuoy, accurately lock the target, start the straight-line navigation mode, and quickly approach the person in distress.

[0029] Preferably, the step S3 specifically includes the following steps:

[0030] Step S31: For the thermal imaging data collected by the first thermal imaging sensor, the second thermal imaging sensor, and the third thermal imaging sensor, the average temperature value of all pixels in the current frame is independently calculated:

[0031]

[0032] And quantify the pixel temperature difference according to the following formula:

[0033]

[0034] Among them, Q is the quantized value of thermal anomaly intensity at pixel level, is the current pixel temperature value; is the environmental adaptive step size factor;

[0035] Step S32: Quantize the value Morphological filtering is performed on the pixels to eliminate discrete thermal noise points.

[0036] Preferably, the step S4 specifically includes:

[0037] Step S41: Accumulate and sum the quantized values ​​Q of each column of pixels to obtain the column-wise integral value ;

[0038]

[0039] Where H is the image height, is the quantized value of the pixel at coordinate (x, y);

[0040] Step S42: Use Gaussian smoothing filter to process the column integral value , to reduce the impact of random noise;

[0041] Step S43: Generate a smoothed column-wise integral matrix.

[0042] Preferably, the step S5 specifically includes:

[0043] Step S51: Positioning satisfies The continuous column interval of is taken as the target area, where β is the threshold coefficient (β=0.7),

[0044] Step S52: Perform secondary screening on the pixels within the target area to ensure that the high-temperature abnormal points are concentrated.

[0045] Preferably, the step S6 specifically includes:

[0046] Step S61: Calculate the coordinates of the centroid of the target area as the guidance reference point;

[0047] Step S62: adjusting the operating power of the two propellers through the motion control module to adjust the direction of the lifebuoy so that the moving trajectory of the lifebuoy points to the center of mass coordinate;

[0048] Step S63: monitor the distance between the lifebuoy and the center of mass coordinates in real time, and dynamically adjust the propeller power to ensure accurate arrival at the target area.

[0049] The present invention provides a terminal guidance system and guidance method for life-saving equipment based on dynamic threshold analysis of thermal imaging, which has the following beneficial effects: by combining thermal imaging technology, dynamic temperature quantification processing of the temperature quantification FPGA module and vertical projection positioning of the vertical projection positioning module, the rapid identification and positioning of thermal anomalies are achieved. Thermal imaging technology first captures the thermal radiation images of the scene, and the temperature quantification FPGA module processes these images in real time, converting the thermal radiation into specific temperature values, and screening out points with abnormally high temperatures as potential targets. Subsequently, the vertical projection positioning module uses the spatial distribution information of these high-temperature points to determine the target area most likely to contain people in distress through calculation and analysis. A series of refined designs and steps are also used to ensure the efficiency and accuracy of the rescue operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for describing the present invention or the prior art.

[0051] Figure 1 A schematic structural diagram of the present invention;

[0052] Figure 2 A schematic structural diagram of the thermal imaging module of the present invention;

[0053] Figure 3 A block diagram showing the connection principle between the thermal imaging module and the embedded processing unit in the present invention;

[0054] Figure 4 A flowchart of the steps of the guidance method of the present invention;

[0055] Description of the numbers in the figure:

[0056] 1. Lifebuoy; 2. Thruster; 3. Thermal imaging module; 4. First thermal imaging sensor; 5. Second thermal imaging sensor; 6. Third thermal imaging sensor; 7. Embedded processing unit; 8. Temperature quantification FPGA module; 9. Vertical projection positioning module; 10. Motion control module. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solutions and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the present invention.

[0058] Examples, such as Figure 1-4 As shown, the present invention discloses a lifesaving equipment terminal guidance system based on thermal imaging dynamic threshold analysis, comprising a lifebuoy 1, with propellers 2 fixedly installed on both sides of the rear end of the lifebuoy 1;

[0059] A thermal imaging module 3 is fixedly mounted on the front end of the upper surface of the lifebuoy 1. The horizontal cross-section of the thermal imaging module 3 is a trapezoidal structure. A first thermal imaging sensor 4, a second thermal imaging sensor 5, and a third thermal imaging sensor 6 are fixedly mounted on the front side, left side, and right side of the thermal imaging module 3, respectively.

[0060] An embedded processing unit 7 is fixedly installed on the lifebuoy 1. The embedded processing unit 7 includes a temperature quantification FPGA module 8, a vertical projection positioning module 9 and a motion control module 10. The signal output ends of the first thermal imaging sensor 4, the second thermal imaging sensor 5 and the third thermal imaging sensor 6 are connected to the signal input end of the temperature quantification FPGA module 8, the signal output end of the temperature quantification FPGA module 8 is connected to the signal input end of the vertical projection positioning module 9, the signal output end of the vertical projection positioning module 9 is connected to the signal input end of the motion control module 10, and the signal output ends of the motion control module 10 are respectively connected to the drive circuit of the thruster 2;

[0061] The temperature quantization FPGA module 8 discretizes and grades the pixel-level temperature difference based on the frame average temperature and dynamic step size. The vertical projection positioning module 9 performs column-wise integration operation on the quantized temperature matrix and extracts the azimuth corresponding to the maximum integral value as the target orientation. The motion control module 10 generates a control signal according to the azimuth to drive the thruster 2 to adjust the heading of the lifebuoy 1 to achieve accurate tracking of the target.

[0062] Working principle:

[0063] In the present invention, a thermal imaging module 3 is installed above a lifebuoy 1, and a first thermal imaging sensor 4, a second thermal imaging sensor 5, and a third thermal imaging sensor 6 are installed on the front side, left inclined surface, and right inclined surface of the thermal imaging module 3, respectively. In this embodiment, the projections on the horizontal plane between the first thermal imaging sensor 4 and the second thermal imaging sensor 5, and between the second thermal imaging sensor 5 and the third thermal imaging sensor 6 are arranged at an angle of 60 degrees. The fields of view of the first thermal imaging sensor 4, the second thermal imaging sensor 5, and the third thermal imaging sensor 6 are arranged at 55 degrees horizontally and 35 degrees vertically, respectively. There is no overlap between adjacent first thermal imaging sensors 4 and second thermal imaging sensors 5, and between adjacent second thermal imaging sensors 5 and third thermal imaging sensors 6. Therefore, the layout design of the three thermal imaging sensors can achieve coverage and monitoring of the target area. When the lifesaving equipment is put into use, the thermal imaging module 3 begins operation, and the three thermal imaging sensors collect thermal imaging data for each target area. This data is then transmitted to the temperature quantization FPGA module 8 for dynamic temperature quantization processing to highlight potential thermal anomalies, which often correspond to the location of people in distress.

[0064] The quantized temperature matrix is ​​then sent to the vertical projection positioning module 9, which further extracts target features through column-wise integration. By comparing the values ​​of different column-wise integrals, the azimuth corresponding to the maximum integral value, i.e., the target position, can be determined. This information is crucial for subsequent tracking and rescue operations. Once the target position is determined, the motion control module 10 generates a corresponding control signal based on the azimuth to drive the propeller to adjust the lifebuoy's course. By monitoring the distance between the lifebuoy and the target position in real time and dynamically adjusting the propeller power, the lifebuoy can be ensured to accurately lock onto the target and quickly approach the person in distress via the shortest path.

[0065] Therefore, the terminal guidance system of the life-saving equipment of the present invention combines thermal imaging technology with advanced signal processing algorithms to achieve rapid and accurate positioning and tracking of people in distress. This not only greatly improves rescue efficiency, but also provides more timely and effective safety protection for people in distress.

[0066] In addition, in this embodiment, the installation pitch angles of the first thermal imaging sensor 4, the second thermal imaging sensor 5 and the third thermal imaging sensor 6 are All meet the following requirements:

[0067]

[0068] Where h is the installation height of the thermal imaging module, d is the nominal detection distance, is the dynamic compensation angle.

[0069] This ensures that the thermal imaging sensor maintains optimal detection performance at varying distances, improving the system's adaptability and accuracy. Furthermore, by introducing a dynamic compensation angle, Δα, the system takes into account various complex situations that may arise in real-world environments, such as the impact of external factors like wind direction and current on the trajectory of the lifesaving device. These effects are compensated by fine-tuning the installation pitch angle, ensuring that the device always tracks the target along the correct path.

[0070] Furthermore, the temperature quantization FPGA module 8 discretizes and grades the pixel-level temperature difference based on the frame average temperature and the dynamic step size, specifically including:

[0071] Calculate the average temperature value of all pixels in the current frame:

[0072]

[0073] The pixel temperature difference is quantified according to the following formula:

[0074]

[0075] Among them, Q is the quantized value of thermal anomaly intensity at pixel level, is the current pixel temperature value; is the environmental adaptive step size factor; is the environmental noise tolerance (0.1℃-0.5℃).

[0076] Environment Adaptive Step Factor The method of determining is:

[0077]

[0078] in, is the temperature standard deviation, is the ambient temperature, 、 is the adjustment coefficient, .

[0079] By dynamically adjusting the quantization step size, thermal anomalies can be more accurately identified, ensuring the accuracy of detection results even in complex and changing environmental conditions. The introduction of an adaptive step size factor (step) allows the quantization process to automatically adapt to changes in the external environment, further enhancing the system's robustness and practicality. By discretizing and grading pixel-level temperature differences, the system can efficiently screen potential thermal anomalies, providing a reliable data foundation for subsequent positioning and tracking.

[0080] In addition, if Figure 4 As shown, the present invention also discloses a guidance method based on the above-mentioned life-saving equipment terminal guidance system, comprising the following steps:

[0081] Step S1: Initialize the spiral search radius , a spiral expansion search strategy is adopted to control the thruster 2 to drive the lifebuoy 1 to sail in a circle;

[0082] Specifically, the spiral expansion search strategy includes:

[0083] Step S11: setting the initial radius R and the initial center coordinates; wherein R is the maximum field of view distance of the thermal imaging sensor;

[0084] Step S12: Spiral track generation: With the initial center coordinate as the center, gradually increase the search radius to generate a spiral track; wherein the recursive formula of the search radius is:

[0085]

[0086] Where n is the current circle number.

[0087] During the spiral search process, the lifebuoy 1 will circle along the generated spiral track. At the same time, the thermal imaging module 3 will continue to work and continuously collect thermal imaging data of the target area.

[0088] Step S13: According to the spiral track and the current position of the life buoy 1, the motion control module 10 adjusts the operating power and direction of the propeller 2 in real time to keep the life buoy 1 sailing stably along the spiral track.

[0089] During the spiral search, the first, second, and third thermal imaging sensors on the lifebuoy continuously collect thermal imaging data from the target area and transmit this data in real time to the temperature quantification FPGA module for dynamic temperature quantification processing. Once a thermal anomaly is successfully identified, the lifebuoy immediately exits the spiraling mode and adjusts the thruster power and the lifebuoy's direction based on the location of the thermal anomaly to accurately lock onto the target and quickly approach the person in distress.

[0090] By employing a spiral expansion search strategy, the present invention enables efficient search and coverage of the target area even when the target's specific location is unknown. Furthermore, combined with thermal imaging technology and advanced signal processing algorithms, the present invention can quickly and accurately identify potential thermally abnormal targets and provide strong support for subsequent rescue operations. This not only significantly improves rescue efficiency but also provides more timely and effective safety protection for those in distress.

[0091] Step S2: Synchronously collecting thermal imaging data of the target area through the first thermal imaging sensor 4, the second thermal imaging sensor 5 and the third thermal imaging sensor 6;

[0092] Step S3: using the temperature quantization FPGA module 8 to independently perform dynamic temperature quantization processing on the thermal imaging data collected by the first thermal imaging sensor 4, the second thermal imaging sensor 5, and the third thermal imaging sensor 6;

[0093] Specifically, the dynamic temperature quantization processing method of the temperature quantization FPGA module 8 includes the following steps:

[0094] Step S31: For the thermal imaging data collected by the first thermal imaging sensor 4, the second thermal imaging sensor 5, and the third thermal imaging sensor 6, the average temperature value of all pixels in the current frame is calculated independently:

[0095]

[0096] And the pixel temperature difference is quantified according to the following formula:

[0097]

[0098] Among them, Q is the quantized value of thermal anomaly intensity at pixel level, is the current pixel temperature value; is the environmental adaptive step size factor;

[0099] Step S32: Quantize the value Morphological filtering is performed on the pixels to eliminate discrete thermal noise points.

[0100] Step S4: performing a column-wise integration operation on the quantized temperature matrix through the vertical projection positioning module 9 to generate a one-dimensional array. In more detail, step S4 includes:

[0101] Step S41: For each column of pixels in the image, extract the corresponding quantization value Q one by one, and then perform a column-by-column accumulation operation on these quantization values ​​to calculate the column-wise integral of each column. value;

[0102]

[0103] Where H represents the total number of pixels in the vertical direction of the image, Refers to the value obtained after the pixel is quantized at the coordinate (x, y);

[0104] Step S42: In order to effectively reduce the influence of random noise on the column integral value, the obtained column integral is processed by using Gaussian smoothing filtering technology. The Gaussian smoothing filter can smooth the data through its unique weight distribution mechanism, thereby filtering out some random noise and making the integral value more stable and reliable;

[0105] Step S43: After the Gaussian smoothing filter is completed, a new column-wise integral matrix is ​​generated based on the filter results. Each element in the matrix is ​​the column-wise integral value after the smoothing process, thereby providing a more accurate and reliable data basis for subsequent image processing and analysis.

[0106] Step S5: extracting the maximum value of the feature vectors corresponding to the first thermal imaging sensor 4, the second thermal imaging sensor 5, and the third thermal imaging sensor 6 respectively. If the maximum value exceeds a threshold, outputting the corresponding position ID. More specifically, step S5 includes:

[0107] Step S51: Positioning satisfies The continuous column interval of is taken as the target area, where β is the threshold coefficient β=0.7,

[0108] Step S52: Perform secondary screening on the pixels within the target area to ensure that the high-temperature abnormal points are concentrated.

[0109] Step S6: When the position ID is valid, the lifebuoy leaves the circling mode and adjusts the operating power of the two propellers according to the position ID through the motion control module 10 to adjust the direction of the lifebuoy 1, so as to accurately lock the target, start the straight-line navigation mode, and quickly approach the person in distress.

[0110] In more detail, step S6 specifically includes:

[0111] Step S61: Calculate the coordinates of the centroid of the target area as the guidance reference point;

[0112] Step S62: The motion control module 10 is used to adjust the operating power of the two propellers to adjust the direction of the lifebuoy 1 so that the moving trajectory of the lifebuoy 1 points to the center of mass coordinate;

[0113] Step S63: monitor the distance between the lifebuoy and the center of mass coordinates in real time, and dynamically adjust the propeller power to ensure accurate arrival at the target area.

[0114] Therefore, the present invention realizes the rapid identification and positioning of thermal anomalies by combining thermal imaging technology, dynamic temperature quantification processing of the temperature quantification FPGA module 8 and vertical projection positioning of the vertical projection positioning module 9. Thermal imaging technology first captures the thermal radiation images of the scene, and the temperature quantification FPGA module 8 processes these images in real time, converts the thermal radiation into specific temperature values, and screens out points with abnormally high temperatures as potential targets. Subsequently, the vertical projection positioning module 9 uses the spatial distribution information of these high-temperature points to determine the target area most likely to contain people in distress through calculation and analysis. A series of refined designs and steps are also used to ensure the efficiency and accuracy of the rescue operation. During the spiral search stage, the life-saving equipment can autonomously use the spiral expansion search strategy to effectively cover and search the target area when the target location is unknown. Once the thermal anomaly is identified, the life-saving equipment can respond quickly, adjust its course, and directly lock on to and approach the people in distress.

[0115] Furthermore, the terminal guidance system for life-saving equipment of the present invention boasts high environmental adaptability and robustness. By dynamically adjusting the quantization step size, the system can automatically adapt to changes in the external environment, ensuring the accuracy of detection results even in complex and changing environmental conditions. This not only improves rescue efficiency but also provides more reliable safety for people in distress.

[0116] In summary, the terminal guidance system and guidance method of the present invention, by combining advanced thermal imaging technology with signal processing algorithms, enables rapid and precise positioning and tracking of people in distress, providing strong support for rescue operations. Its high environmental adaptability and robustness ensure efficient and accurate rescue operations, and it possesses extremely high practical value and social significance.

[0117] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A lifesaving equipment terminal guidance system based on thermal imaging dynamic threshold analysis, comprising a lifebuoy (1), wherein thrusters (2) are fixedly mounted on both sides of the rear end of the lifebuoy (1); Its characteristics are: A thermal imaging module (3) is fixedly mounted on the front end of the upper surface of the lifebuoy (1); the horizontal cross-section of the thermal imaging module (3) is a trapezoidal structure; a first thermal imaging sensor (4), a second thermal imaging sensor (5), and a third thermal imaging sensor (6) are fixedly mounted on the front side, left side, and right side of the thermal imaging module (3); An embedded processing unit (7) is fixedly mounted on the lifebuoy (1), and the embedded processing unit (7) includes a temperature quantification FPGA module (8), a vertical projection positioning module (9), and a motion control module (10); the signal output ends of the first thermal imaging sensor (4), the second thermal imaging sensor (5), and the third thermal imaging sensor (6) are connected to the signal input end of the temperature quantification FPGA module (8); the signal output end of the temperature quantification FPGA module (8) is connected to the signal input end of the vertical projection positioning module (9); the signal output end of the vertical projection positioning module (9) is connected to the signal input end of the motion control module (10); and the signal output end of the motion control module (10) is respectively connected to the drive circuit of the thruster (2); The temperature quantization FPGA module (8) discretizes and grades the pixel-level temperature difference based on the frame average temperature and the dynamic step size. The vertical projection positioning module (9) performs a column-wise integral operation on the quantized temperature matrix and extracts the azimuth angle corresponding to the maximum integral value as the target azimuth. The motion control module (10) generates a control signal according to the azimuth angle to drive the propeller (2) to adjust the heading of the lifebuoy (1) to achieve accurate tracking of the target.

2. The life-saving equipment terminal guidance system based on thermal imaging dynamic threshold analysis according to claim 1, characterized in that: The projections on the horizontal plane between the first thermal imaging sensor (4) and the second thermal imaging sensor (5), and between the second thermal imaging sensor (5) and the third thermal imaging sensor (6) are respectively distributed at an angle of 60 degrees, and the field of view ranges of the first thermal imaging sensor (4), the second thermal imaging sensor (5) and the third thermal imaging sensor (6) are respectively distributed at a horizontal angle of 55 degrees and a vertical angle of 35 degrees, and there is no overlapping area between the adjacent first thermal imaging sensor (4) and the second thermal imaging sensor (5) and the third thermal imaging sensor (6).

3. The life-saving equipment terminal guidance system based on thermal imaging dynamic threshold analysis according to claim 1, characterized in that: The installation pitch angles of the first thermal imaging sensor (4), the second thermal imaging sensor (5) and the third thermal imaging sensor (6) All meet the following requirements: Where h is the installation height of the thermal imaging module, d is the nominal detection distance, is the dynamic compensation angle.

4. The life-saving equipment terminal guidance system based on thermal imaging dynamic threshold analysis according to claim 1, characterized in that: The temperature quantization FPGA module (8) discretizes and grades the pixel-level temperature difference based on the frame average temperature and the dynamic step size, specifically including: calculating the average temperature value of all pixels in the current frame: The pixel temperature difference is quantified according to the following formula: Among them, Q is the quantized value of thermal anomaly intensity at pixel level, is the current pixel temperature value; is the environmental adaptive step size factor; is the ambient noise tolerance.

5. The life-saving equipment terminal guidance system based on thermal imaging dynamic threshold analysis according to claim 4, characterized in that: The environment adaptive step factor The method of determining is: in, is the temperature standard deviation, is the ambient temperature, 、 is the adjustment coefficient, .

6. A guidance method for a lifesaving equipment terminal guidance system based on thermal imaging dynamic threshold analysis according to any one of claims 1 to 5, characterized in that: The following steps are involved: Step S1: Initialize the spiral search radius R, and use the spiral expansion search strategy to control the propeller (2) to drive the lifebuoy (1) to sail in a circle; Step S2: Synchronously collecting thermal imaging data of the target area through the first thermal imaging sensor (4), the second thermal imaging sensor (5) and the third thermal imaging sensor (6). Step S3: using the temperature quantization FPGA module (8) to independently perform dynamic temperature quantization processing on the thermal imaging data collected by the first thermal imaging sensor (4), the second thermal imaging sensor (5), and the third thermal imaging sensor (6); Step S4: performing column-wise integration operation on the quantized temperature matrix through the vertical projection positioning module (9) to generate a one-dimensional array; Step S5: extracting the maximum values ​​of the feature vectors corresponding to the first thermal imaging sensor (4), the second thermal imaging sensor (5), and the third thermal imaging sensor (6), respectively, and outputting the corresponding orientation ID if the maximum value exceeds a threshold; Step S6: When the position ID is valid, the lifebuoy leaves the circling mode and adjusts the operating power of the two propellers according to the position ID through the motion control module (10) to adjust the direction of the lifebuoy (1) so as to accurately lock the target and start the straight-line navigation mode to quickly approach the person in distress.

7. The guidance method according to claim 6, characterized in that: The step S3 specifically includes the following steps: Step S31: For the thermal imaging data collected by the first thermal imaging sensor (4), the second thermal imaging sensor (5), and the third thermal imaging sensor (6), the average temperature value of all pixels in the current frame is calculated independently: And the pixel temperature difference is quantified according to the following formula: Among them, Q is the quantized value of thermal anomaly intensity at pixel level, is the current pixel temperature value; is the environmental adaptive step size factor; Step S32: Quantize the value Morphological filtering is performed on the pixels to eliminate discrete thermal noise points.

8. The guidance method according to claim 6, characterized in that: The step S4 specifically includes: Step S41: Accumulate and sum the quantized values ​​Q of each column of pixels to obtain the column-wise integral value ; Where H represents the number of pixels in the vertical direction of the image. is the quantized value of the pixel at the coordinate (x, y); Step S42: Use Gaussian smoothing filter to process the column-wise integral value , to reduce the impact of random noise; Step S43: Generate a smoothed column-wise integral matrix.

9. The guidance method according to claim 8, characterized in that: The step S5 specifically includes: step S51: positioning the The continuous column interval of is taken as the target area, where β is the threshold coefficient (β=0.7). Step S52: perform secondary screening on the pixels in the target area to ensure that the high-heat abnormal points are concentrated.

10. The guidance method according to claim 6, characterized in that: The step S6 specifically includes: step S61: calculating the coordinates of the center of mass of the target area as a reference point for guidance; step S62: adjusting the operating power of the two propellers through the motion control module (10) to adjust the direction of the lifebuoy (1) so that the moving trajectory of the lifebuoy (1) points to the coordinates of the center of mass; step S63: monitoring the distance between the lifebuoy and the coordinates of the center of mass in real time, and dynamically adjusting the propeller power to ensure that the target area is accurately reached.

Images