A method and system for optical axis stability detection for thermal imaging devices

Abstract

The present application relates to a kind of optical axis stability detection method and system for thermal imaging device, comprising: build optical axis detection platform and obtain the initial image of the thermal imaging device to be measured;The preset evaluation operation is carried out to the thermal imaging device to be measured, and secondary image is obtained;Using distinguishing rule to identify feature area, form first layer target feature set and second layer target feature set;Calculate multidimensional offset, according to multidimensional offset using judgment logic to judge distortion type, calculate comprehensive offset, according to comprehensive offset to identify optical axis stability, through translation offset, rotation offset, distortion offset and axial offset, improve the detection precision, and identify potential internal damage, through double-layer target structure and differentiation shape design, detection dimension is expanded to three-dimensional space.

Description

Technical Field

[0001] This invention relates to the field of thermal imaging technology, and in particular to a method and system for detecting the optical axis stability of thermal imaging equipment. Background Technology

[0002] In the field of thermal imaging technology, thermal imaging equipment, as a device capable of capturing and converting infrared radiation into visible images, is widely used in various fields such as military reconnaissance, security monitoring, medical diagnosis, industrial inspection, and scientific research. Thermal imaging equipment receives infrared radiation emitted by a target object, focuses it through an optical system, converts it using an infrared detector, and processes the signal to ultimately form a thermal image reflecting the temperature distribution on the object's surface. In this process, the stability of the optical axis is crucial for ensuring the imaging quality and measurement accuracy of the thermal imaging equipment.

[0003] Optical axis stability refers to the ability of a thermal imaging device's optical system to maintain its initial alignment without significant shifts or jitter during prolonged operation or exposure to various environmental changes. However, in practical applications, thermal imaging devices are often affected by a variety of factors, leading to a decrease in optical axis stability. These factors include, but are not limited to, mechanical vibration, temperature changes, shock vibration, repeated disassembly and assembly, and inhomogeneities in the internal optical path medium.

[0004] Mechanical vibration and shock are common external disturbances encountered by thermal imaging equipment during use. For example, in military reconnaissance or security monitoring scenarios, thermal imaging equipment may be mounted on mobile carriers, such as vehicles or drones. Vibrations generated by these carriers during travel or flight are directly transmitted to the thermal imaging equipment, affecting its optical axis stability. Furthermore, the equipment may also be subjected to shock and vibration during transportation, installation, or maintenance, leading to optical axis misalignment. Summary of the Invention

[0005] This invention addresses the technical problems existing in the prior art by providing a method and system for detecting the optical axis stability of thermal imaging equipment. It achieves multi-dimensional, refined, and digital detection and diagnosis of the optical axis stability of thermal imaging equipment through translational offset, rotational offset, torsional offset, and axial offset, thereby improving detection accuracy and identifying potential internal damage. Through a double-layer target structure and differentiated shape design, the detection dimension is extended to three-dimensional space.

[0006] The technical solution of the present invention to solve the above-mentioned technical problems is as follows: a method for detecting the optical axis stability of thermal imaging equipment is provided. S101, Build an optical axis detection platform, and obtain the initial image of the thermal imaging device under test based on the optical axis detection platform; perform a preset evaluation operation on the thermal imaging device under test, and obtain a secondary image after the preset evaluation operation. The optical axis detection platform includes a thermal imaging device under test, a collimator, a multi-layer irregular target, a first stage, a second stage, and a controller. The preset evaluation operations include repeated disassembly and assembly, impact and vibration, temperature cycling, and focusing actions; S102, process the initial image and the secondary image, use the distinction rule to identify the feature region, and extract the center coordinates of the feature region to form the first layer target feature set and the second layer target feature set; S103, calculate multidimensional offset based on the first layer target feature set and the second layer target feature set, wherein the multidimensional offset includes translation offset, rotation offset, twist offset and axial offset; S104 uses judgment logic to determine the distortion type based on the multidimensional offset, calculates the comprehensive offset based on translation offset, rotation offset, torsion offset and axial offset, and identifies the optical axis stability based on the comprehensive offset.

[0007] Preferably, the method for acquiring the initial image is as follows: the controller sends a working command to the thermal imaging device under test, the device observes the multi-layer irregular target through a collimator, the controller controls the thermal imaging device under test to perform focusing operation until the device achieves the clearest imaging of the multi-layer irregular target, and after focusing is completed, the controller acquires the image output by the thermal imaging device under test.

[0008] Preferably, the distinction rule is to calculate the roundness based on the extracted contour. When the roundness is greater than a preset recognition threshold, the contour is round; when the roundness is less than the preset recognition threshold, the contour is square. For the remaining contours, a skeleton extraction algorithm is used for recognition to obtain the number of skeleton endpoints. Contours with four skeleton endpoints are identified as cross-shaped feature regions, and contours with three skeleton endpoints are identified as L-shaped feature regions.

[0009] Preferably, the step of identifying optical axis stability based on the comprehensive offset is as follows: the comprehensive offset is compared with a preset comprehensive threshold; if the comprehensive offset is less than the preset comprehensive threshold, the optical axis stability is deemed acceptable; if the comprehensive offset is greater than or equal to the preset comprehensive threshold, the optical axis stability is deemed unacceptable.

[0010] Preferably, the stability testing method further includes: S201, Set up a monitoring module, use the monitoring module to acquire historical data, use judgment rules to determine the convection state based on the historical data; use sensors to identify the working posture under the historical convection state, and construct a posture-convection correspondence table according to the working posture and the corresponding convection state; the historical data includes flow velocity data and temperature data. S202, determine the current convection state using the attitude convection correspondence table based on the current device attitude, adjust the optical axis detection strategy according to the current convection state, and acquire the target image based on the adjusted optical axis detection; S203, based on the target image, execute steps S103 and S104 to determine the stability of the optical axis.

[0011] Preferably, the rules for determining the convection state are as follows: Within 0-5 minutes after startup, if the temperature gradient value is in the initial fluctuation stage and the flow rate is low and gradually increasing, the convection state is determined to be the initial stage of convection; within 5-20 minutes after startup, if the temperature gradient value gradually increases and stabilizes within the range, and the flow rate continues to rise and reaches a high threshold range, the convection state is determined to be the convection development stage; after 20 minutes of startup, if the temperature gradient value reaches its maximum value and tends to stabilize, and the flow rate also reaches its maximum value and tends to stabilize, the convection state is determined to be the convection stabilization stage; wherein, the initial fluctuation stage is defined as the absolute value of the temperature gradient being less than a preset threshold and the rate of change of the absolute value being greater than the rate of change threshold; the high threshold range refers to the range greater than the maximum value of the upward trend in the initial stage of convection.

[0012] Preferably, the stability testing method further includes: S301, install ultrasonic sensors and strain sensors to collect reflection data and strain data respectively, calculate the adhesive layer stress index and pressure ring stress index based on the reflection data and strain data, and obtain the stress competition index through the adhesive layer stress index and pressure ring stress index; S302, Construct a jump warning model based on the stress competition index, use the jump warning model to predict optical axis jump, and issue a warning based on the prediction results; S303, based on the stress competition index, uses the jump rule to identify the jump source; calculates the fixation health based on the adhesive layer stress index and the pressure ring stress index, and judges the optical axis stability based on the fixation health.

[0013] This application also provides an optical axis stability detection system for thermal imaging equipment, the system including an image acquisition module, a calculation module and a stability judgment module; The image acquisition module is used to control the thermal imaging device under test through the controller, observe the multi-layer irregular target and acquire the initial image and the secondary image after the preset evaluation operation; The calculation module is used to calculate multi-dimensional offsets such as translation offset, rotation offset, torsion offset, and axial offset; The stability judgment module is used to judge the distortion type using preset judgment logic, and at the same time, calculate the comprehensive offset and judge whether the optical axis stability is qualified according to the preset comprehensive threshold. The image acquisition module is connected to the calculation module, and the calculation module is connected to the stability judgment module.

[0014] The beneficial effects of this invention are: by using translation offset, rotation offset, torsion offset and axial offset to achieve multi-dimensional, refined and digital detection and diagnosis of the optical axis stability of thermal imaging equipment, the detection accuracy is improved and potential internal damage is identified. Through the double-layer target structure and differentiated shape design, the detection dimension is extended to three-dimensional space. By identifying the convection state and adjusting the detection strategy according to the attitude, the problems of poor detection repeatability caused by convection disturbance, sudden changes in optical axis caused by attitude change, uncertain detection time window after power-on, and inability to distinguish between true offset and convection pseudo offset are effectively solved, thus improving the quality and reliability of optical axis stability detection. By using a jump warning model to predict the probability of jumps in the next 24 hours, the detection is upgraded from post-discovery to pre-event warning. Detection or shutdown inspection can be arranged in advance to avoid sudden loss of aiming accuracy, improve the reliability and safety of equipment operation, achieve comprehensive perception of the composite fixed state of the folding mirror and accurate source tracing and early warning of optical axis jumps, and improve the accuracy and timeliness of optical axis stability detection. Attached Figure Description

[0015] Figure 1 This is a schematic flowchart of a method for detecting the optical axis stability of a thermal imaging device according to the present invention. Figure 2 This is a schematic diagram of the process for determining the stability of the optical axis based on the convection state according to the present invention; Figure 3 This is a schematic diagram of the process for calculating a fixed health score according to the present invention; Figure 4 This is a block diagram of an optical axis stability detection system for a thermal imaging device according to the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0017] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0018] In the description of this application, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed in this application.

[0019] Example 1: Figure 1 This is a flowchart illustrating a method for detecting the optical axis stability of a thermal imaging device according to Embodiment 1 of the present invention, comprising the following steps: S101, Build an optical axis detection platform, and obtain the initial image of the thermal imaging device under test based on the optical axis detection platform; perform a preset evaluation operation on the thermal imaging device under test, and obtain a secondary image after the preset evaluation operation. The optical axis detection platform includes a thermal imaging device under test, a collimator, a multi-layer irregular target, a first stage, a second stage, and a controller. The preset evaluation operations include repeated disassembly and assembly, impact and vibration, temperature cycling, and focusing actions; Specifically, an optical axis detection platform is constructed, comprising a thermal imaging device under test, a collimator, a multi-layer irregularly shaped target, a first stage, a second stage, and a controller. The thermal imaging device under test refers to an infrared thermal imager or thermal imager that requires optical axis stability testing. The thermal imaging device has a focusing function and can clearly image the target at different object distances. The collimator is used to provide a simulated optical device for targets at infinity. The optical axis of the collimator is aligned with the optical axis of the thermal imaging device under test to ensure that the device can receive light from the simulated infinity-distance target. The multi-layer irregularly shaped target serves as the target object to be imaged and is configured as a double-layer target. The first layer target is located near the collimator, with a circular feature pattern in its central area, a cross-shaped feature pattern in its edge area, and four L-shaped feature patterns with four different orientations in its four corner areas. The second layer target is located away from the collimator and is aligned with the first stage. A target layer is offset by a predetermined distance along the optical axis. Its central area has a square feature pattern, its edge area has a cross-shaped feature pattern rotated 45° from the first layer's cross shape, and its four corner areas have L-shaped feature patterns complementary to the first layer's L-shape orientation. The first platform is used to support the multi-layered irregularly shaped targets and can move the target position via control commands. The second platform is used to support the thermal imaging device under test and has height and angle adjustment functions. The controller is electrically connected to the thermal imaging device under test and the first platform, used to send control commands and receive image data. The specific method for building the optical axis detection platform is as follows: the thermal imaging device under test is fixedly installed on the second platform. By adjusting the height and angle, the optical axis of the thermal imaging device under test is aligned with the optical axis of the collimator. The multi-layered irregularly shaped targets are fixedly installed on the first platform, and the initial position of the first platform is set at the imaging distance of the thermal imaging device under test.

[0020] After the optical axis detection platform is set up, the initial image is acquired. Specifically, the controller sends a working command to the thermal imaging device under test. The device observes through a multi-layer irregular target with a collimator. The controller controls the thermal imaging device under test to perform focusing operations until the imaging of the multi-layer irregular target is at its clearest state. After focusing is completed, the controller acquires the image output by the thermal imaging device under test, which is the initial image. The initial image includes the circular feature pattern, cross-shaped feature pattern, and L-shaped feature pattern with four orientations of the first layer target; the square feature pattern, the cross-shaped feature pattern rotated 45° from the first layer cross-shaped feature pattern, and the four L-shaped feature patterns that are complementary to the orientation of the first layer L-shaped feature pattern.

[0021] After initial image acquisition, a preset evaluation operation is performed on the thermal imaging device under test to simulate various conditions affecting optical axis stability experienced during actual use. This preset evaluation operation includes repeated disassembly and assembly, impact and vibration, temperature cycling, and focusing. Repeated disassembly and assembly is used to detect the impact of the mechanical interface's repeatability accuracy on optical axis stability. Impact and vibration are used to detect the device's internal structure's resistance to mechanical disturbances. Temperature cycling simulates the device's thermal stability under different climatic conditions. Focusing simulates the device's ability to maintain optical axis stability under frequent focusing scenarios. After completing the preset evaluation operation... Keeping the position of the multi-layered irregular target unchanged, the controller sends a working command to the thermal imaging device under test again. The device observes the multi-layered irregular target again through the collimator. The controller refocuses the thermal imaging device under test. After the focus is clear, the controller acquires the image output by the thermal imaging device under test, i.e., the secondary image. The secondary image includes the circular, cross-shaped, and L-shaped feature patterns of the first layer of targets, and the square, cross-shaped, and L-shaped feature patterns of the second layer of targets. The secondary image records the position and clarity information of the thermal imaging device imaging the same target after undergoing the preset evaluation operation, which serves as a basis for comparison with the initial image.

[0022] S102, process the initial image and the secondary image, use the distinction rule to identify the feature region, and extract the center coordinates of the feature region to form the first layer target feature set and the second layer target feature set; The feature regions include circular, square, cross-shaped, and L-shaped regions; the first layer of target feature set refers to the coordinates extracted from all feature points on the target layer closest to the collimator, and the second layer of target feature set refers to the coordinates extracted from all feature points on the target layer furthest from the collimator. Furthermore, grayscale and binarization processing is performed on the initial and secondary images. Grayscale processing preserves the brightness difference information between the feature regions and the background. After binarization, each feature pattern of the multi-layered irregular target appears as a white area, completely separated from the black background, forming a clear binarized outline. After binarization, the controller extracts the outline of the white area in the binarized image and uses a distinction rule to identify the feature region. The distinction rule calculates the roundness based on the extracted outline. When the roundness is greater than a preset recognition threshold, the outline is considered circular. Circular feature regions only exist in the first... The center position of the first layer target is determined by the shape of the outline. Therefore, a circular outline corresponds to the central region of the first layer target. When the circularity is less than a preset recognition threshold, the outline is square. Square feature regions only exist at the center of the second layer target; therefore, outlines identified as square correspond to the central region of the second layer target. For the remaining outlines, a skeleton extraction algorithm is used to identify them, obtaining the number of skeleton endpoints. Outlines with four skeleton endpoints are identified as cross-shaped feature regions, and outlines with three skeleton endpoints are identified as L-shaped feature regions. Specifically, the skeleton extraction algorithm gradually peels away the outline of a connected region until a cross-shaped feature region is obtained. A single-pixel-width centerline structure preserves the topological features and geometric orientation of the original shape. After obtaining the skeleton of each contour, the number of branch endpoints of the skeleton is counted. An endpoint is a point on the skeleton line that is connected to only one adjacent pixel, reflecting the number of protruding parts of the shape. The cross-shaped pattern has four protruding branches. When the four protruding branches point to the four positive directions (up, down, left, right), there is one horizontal arm and one vertical arm. This cross-shaped pattern is identified as belonging to the first layer of targets. When the four arms of the cross-shaped pattern point to the diagonal direction, i.e., after rotating 45° to form an X shape, this cross-shaped pattern is identified as belonging to the first layer of targets. The shape is identified as belonging to the second layer of targets. Therefore, the cross-shaped skeleton structure must contain four endpoints. The L-shaped pattern has two mutually perpendicular arms, forming a right angle shape. Its skeleton structure contains three endpoints, located at the ends of the two arms and the right angle vertex. The vertex position and the extension direction of the two arms are determined by the corner detection algorithm. For example, if the two arms of one L-shape point downward and to the right respectively, and the right angle opening points to the lower right direction, then this L-shape belongs to the upper left corner of the first layer of targets; if the two arms of another L-shape point to the right and downward respectively, and the right angle opening points to the lower left direction, then this L-shape belongs to the upper left corner of the second layer of targets.

[0023] Center coordinates are extracted for each type of feature region. For circular and square feature regions, the centroid method is used, which calculates the average of the coordinates of all pixels within the contour to obtain the center coordinates of the region. For cross-shaped feature regions, the skeleton intersection method is used, which finds the point where the four branches intersect by detecting the intersection of the skeleton lines, which is the center of the cross. For L-shaped feature regions, the center coordinates are extracted using the corner point and arm length weighted method. The vertex of the L-shape (i.e., the intersection of the two arms) is used as the reference, and the weighted average of the lengths of the two arms is used to determine the equivalent center point. This equivalent center point is located inside the L-shape near the vertex. The extracted center coordinates form the first layer of target feature set and the second layer of target feature set. The first layer of target feature set includes circular, cross-shaped, and L-shaped feature regions, and the second layer of target feature set includes square, cross-shaped, and L-shaped feature regions.

[0024] S103, calculate multidimensional offset based on the first layer target feature set and the second layer target feature set, wherein the multidimensional offset includes translation offset, rotation offset, twist offset and axial offset; Specifically, multidimensional offsets are calculated based on the first-layer target feature set and the second-layer target feature set. These multidimensional offsets include translational offsets, rotational offsets, torsional offsets, and axial offsets. The formula for calculating the translational offset is as follows: ,in, The translation offset represents the degree of lateral drift of the optical axis of the thermal imaging device within a plane perpendicular to the optical axis. The x-coordinate of the center point of the circular feature region of the first layer target in the initial image represents the central reference position of the first layer target in the initial state. The ordinate of the center point of the first layer of the target circular feature region in the initial image. The x-coordinate of the center point of the circular feature region of the first layer target in the secondary image represents the reference position of the center of the first layer target during secondary imaging. Let be the ordinate of the center point of the first layer of the target's circular feature region in the secondary image. The above translational offset can be obtained by comparing the positional differences of the same circular feature point in the two images to obtain the overall translational amount of the optical axis; the formula for calculating the rotational offset is: ,in, The rotation offset represents the overall rotation angle of the image between the two imaging operations. , This represents the x and y coordinates of the center point of the first cross-shaped feature region in the initial image. , This represents the x and y coordinates of the center point of the second cross-shaped feature region in the initial image. , This represents the center coordinates of the feature point corresponding to the first cross in the quadratic image. , The coordinates of the center of the feature point corresponding to the second cross in the secondary image are represented. The above formula for rotation offset is based on the center coordinates of the cross in the first layer of target feature set. By calculating the tilt angle of the line connecting the two cross feature points in the image and comparing the change of this angle in the two imaging processes, the overall rotation of the image can be obtained. The formula for calculating the distortion offset is: in, The distortion offset represents the maximum percentage change in relative distance between all L-shaped feature points. Let be the Euclidean distance between the i-th L-shaped feature point and the j-th L-shaped feature point in the initial image. The Euclidean distance between the i-th and j-th L-shaped feature points in the quadratic image is calculated using the above formula for twist offset. This formula captures the most severe local distortion by calculating the maximum rate of change of distance between all point pairs. The formula for calculating the axial offset is: ,in, This represents the modulation transfer function value of the circular region of the first-layer target, reflecting the image sharpness of that region. This represents the modulation transfer function value of the square region of the second-layer target, reflecting the image sharpness of that region. The sharpness ratio represents the relative sharpness of the first-layer target and the second-layer target in the same image. ,in, The axial offset represents the degree of displacement of the focal plane along the optical axis. The sharpness ratio of the secondary image. The sharpness ratio of the initial image is given by k, which is a calibration coefficient obtained through prior experimental measurement. The formula for the axial offset is based on the distance between the two target layers along the optical axis. When the focal plane position of the thermal imaging device changes, the image sharpness of the two target layers will change in opposite directions.

[0025] S104 uses judgment logic to determine the distortion type based on the multidimensional offset, calculates the comprehensive offset based on translation offset, rotation offset, torsion offset and axial offset, and identifies the optical axis stability based on the comprehensive offset.

[0026] Furthermore, based on the technical specifications, application scenario requirements, and quality control standards of the thermal imaging equipment, translation threshold, rotation threshold, torsion threshold, and axial offset threshold are set. When the translation offset is greater than the preset translation threshold and the rotation offset is less than the preset rotation threshold, and the torsion offset is less than the preset torsion threshold, it is determined to be a pure translation type, indicating that the optical axis has undergone overall translation without relative rotation or local deformation; when the rotation offset is greater than or equal to the preset rotation threshold and less than the preset rotation threshold, it is determined to be a pure rotation type, indicating that the optical axis has undergone overall rotation; when the torsion offset is greater than or equal to the preset torsion threshold... When the axial offset is greater than or equal to a preset axial offset threshold, it is determined to be an axial offset type, indicating that the focal plane has undergone significant displacement along the optical axis. The sum of the squares of the translational offset, rotational offset, torsional offset, and axial offset is calculated and then squared to obtain the comprehensive offset. The comprehensive offset is compared with a preset comprehensive threshold. When the comprehensive offset is less than the preset comprehensive threshold, the optical axis stability is deemed acceptable. When the comprehensive offset is greater than or equal to the preset comprehensive threshold, the optical axis stability is deemed unacceptable. The comprehensive threshold is set according to industry standards, product specifications, or user requirements.

[0027] The technical solutions in the above embodiments of this application have at least the following technical effects or advantages: multi-dimensional, refined, and digital detection and diagnosis of the optical axis stability of thermal imaging equipment is achieved by translation offset, rotation offset, torsion offset and axial offset, which improves the detection accuracy and identifies potential internal damage. Through the double-layer target structure and differentiated shape design, the detection dimension is extended to three-dimensional space.

[0028] Example 2: In Example 1, when performing optical axis stability testing, it was assumed that the optical path medium inside the thermal imaging device was uniform and stable. However, in actual operation, residual gas in the sealed cavity of the thermal imaging device can cause convection errors. This example identifies the current convection state by real-time measurement of the gas flow rate and temperature distribution within the sealed cavity, thus eliminating the influence of convection disturbances on the detection results. Figure 2 As shown.

[0029] S201, Set up a monitoring module, use the monitoring module to acquire historical data, identify historical convection states based on historical data; use sensors to identify the working posture under historical convection states, and construct a posture-convection correspondence table based on the working posture and the corresponding convection state; where historical data includes flow velocity data and temperature data; Specifically, the monitoring module is a gas state monitoring module, which consists of a flow velocity sensor and a temperature sensor to monitor flow velocity data and temperature data, respectively. The temperature sensor is installed at the top and bottom of the thermal imaging device, while the flow velocity sensor is installed in the deflection optical path area and inside the lens barrel. The direction of the flow velocity sensor is consistent with the direction of gas flow. Historical flow velocity data and historical temperature data are collected at set time intervals using the flow velocity and temperature sensors. The collected data is stored, and the historical temperature data is organized into a temperature data sequence according to time order. For the temperature data at each time point, the difference between the top and bottom temperatures (i.e., the temperature gradient value) is calculated. Based on the calculated difference, the convection temperature rule is used to determine the convection direction and convection intensity trend. Specifically, a temperature threshold range is set, including a minimum temperature threshold and a maximum temperature threshold. When the calculated difference is greater than the preset maximum temperature threshold, hot air rises, and convection is mainly in the vertical direction; when the calculated difference is less than the maximum temperature threshold, the convection is less than the maximum temperature threshold. When the temperature reaches the preset minimum threshold, the hot air decreases (abnormal situation), and the thermal imaging equipment is checked for any abnormal operation. When the absolute value of the difference is less than the preset maximum temperature threshold, the convection intensity is weak. Historical flow velocity data is organized into a flow velocity data sequence according to time order. Based on the flow velocity data at each time point, the convection velocity rule is used to accurately determine the convection intensity level. Specifically, a flow velocity threshold range is set, which includes a minimum flow velocity threshold and a maximum flow velocity threshold. When the flow velocity of all sensors is less than the minimum flow velocity threshold, the convection intensity is weakest. When the flow velocity of any sensor is between the minimum and maximum flow velocity thresholds, the convection intensity is moderate. When the flow velocity of any sensor is greater than the maximum flow velocity threshold, the convection intensity is strongest. The power-on time information is extracted from historical data and combined with the data acquisition timestamp to determine the power-on time corresponding to each data point. Based on the power-on time, temperature gradient measurement results, and flow velocity measurement results, the convection state is comprehensively judged. The specific judgment rule is: after power-on, 0 -5 minutes after startup, the temperature gradient value is in the initial fluctuation stage (the absolute value of the temperature gradient is less than the preset gradient threshold and the rate of change of the absolute value is greater than the rate of change threshold), the flow rate is relatively low and shows a gradual upward trend, indicating that the convection state is in the initial stage of convection, during which convection is weak; 5-20 minutes after startup, the temperature gradient value gradually stabilizes in a range (the range refers to the range greater than the preset gradient threshold), and the flow rate also continues to rise and reaches the high threshold range (the high threshold range refers to the range greater than the maximum value of the upward trend in the initial stage of convection), but is still fluctuating, indicating that the convection state is in the development stage of convection, during which convection is moderate; 20 minutes after startup, the temperature gradient value tends to stabilize, the fluctuation range is small, and the flow rate also stabilizes near a relatively constant value, indicating that the convection state is in the stable stage of convection, during which convection is strong.

[0030] Tilt and acceleration sensors are installed on the thermal imaging device to collect data synchronously with temperature and flow rate sensors. Working posture data is recorded at the same time intervals. Based on the aforementioned determined convection state (i.e., 0-5 minutes after power-on is the initial stage of convection, 5-20 minutes is the development stage of convection, and 20 minutes after power-on is the stable stage of convection), the working posture data is divided into corresponding time intervals (each corresponding to the convection state of different time intervals). An attitude-convection correspondence table is constructed according to the working posture and the corresponding convection state. The attitude-convection correspondence table is used to record the working posture information of the device under different convection states.

[0031] S202, determine the current convection state using the attitude convection correspondence table based on the current device attitude, adjust the optical axis detection strategy according to the current convection state, and acquire the target image based on the adjusted optical axis detection; Furthermore, the current working posture data is acquired through tilt and acceleration sensors. This data is then compared and analyzed with a posture-convection correspondence table. By searching for and matching corresponding posture features, the current convection state of the device is determined. When the current convection state is determined to be in the early stage, the corresponding image jitter is less than 0.1 pixels, and the convection intensity is determined to be level 1. In this case, 10 frames of images are continuously acquired. Because the image jitter is relatively small under weak convection, acquiring a smaller number of frames and averaging them can effectively reduce the impact of image jitter caused by convection on the detection results, while also ensuring detection efficiency. When the current convection state is determined to be in the development stage, the corresponding image jitter is between 0.1 and 0.3 pixels, and the convection intensity is determined to be level 2. In this case, 30 frames of images are continuously acquired. Frame images can more comprehensively cover the changes in the image during convection. Statistical averaging further improves the suppression of image jitter, thereby enhancing the accuracy of optical axis detection. When the current convection state is determined to be stable, and the corresponding image jitter is greater than 0.3 pixels, the convection intensity is determined to be level 3. In this case, 50 frames of images are continuously acquired. Strong convection will cause more severe image jitter. Acquiring a larger number of frames can better capture the dynamic changes of the image. Through a complex statistical averaging algorithm, the impact of convection on image quality is minimized, ensuring the reliability of the optical axis detection results. After determining the number of frames to be acquired, the target image is continuously acquired using a thermal imaging device according to the predetermined number of frames.

[0032] S203, based on the target image, execute steps S103 and S104 to determine the stability of the optical axis.

[0033] Specifically, based on the acquired target image, steps S103 and S104 are executed to calculate the translational offset, rotational offset, torsional offset, axial offset, and comprehensive offset, and the stability of the optical axis is determined based on the comprehensive offset.

[0034] The technical solutions in the above embodiments of this application have at least the following technical effects or advantages: by identifying the convection state and adjusting the detection strategy according to the attitude, the problems of poor detection repeatability caused by convection disturbance, sudden changes in optical axis caused by attitude change, uncertain detection time window after power-on, and inability to distinguish between true offset and convection pseudo offset are effectively solved, thereby improving the quality and reliability of optical axis stability detection.

[0035] Example 3: Examples 1 and 2 above can only obtain the final offset after the optical axis jump occurs, and cannot determine whether the jump is caused by adhesive layer creep or pressure ring slippage. This example improves the accuracy and timeliness of optical axis stability detection by comprehensively sensing the composite fixation state of the folding mirror and accurately tracing and warning of optical axis jumps. Figure 3 As shown.

[0036] S301, install ultrasonic sensors and strain sensors to collect reflection data and strain data respectively, calculate the adhesive layer stress index and pressure ring stress index based on the reflection data and strain data, and obtain the stress competition index through the adhesive layer stress index and pressure ring stress index; Specifically, an ultrasonic sensor is installed at the adhesive layer to collect reflection data. This reflection data includes a first characteristic peak and a second characteristic peak. The first characteristic peak corresponds to the interface between the adhesive layer and the mirror mount, and is sensitive to the acoustic impedance of the adhesive layer. The second characteristic peak corresponds to the contact interface between the pressure ring and the mirror mount, and is sensitive to the contact stiffness. A strain sensor is installed on the pressure ring to collect strain data, which represents the strain change of the pressure ring at different times. Based on the collected reflection data and strain data, the adhesive layer stress index is calculated using the following formula: ,in, This represents the adhesive layer stress index at time t, used to quantify the stress state of the adhesive layer at a given moment. The frequency shift of the first characteristic peak at time t, i.e., the difference between the frequency of the first characteristic peak at the current time and the reference frequency. The difference reflects the change in ultrasonic frequency of the adhesive layer under the action of factors such as force. The center frequency of the first characteristic peak under the reference state is the frequency value of the first characteristic peak measured under the initial or steady state of the system. This is the moisture absorption amplification factor, used to account for the degree of influence of moisture absorption by the adhesive layer on the stress index of the adhesive layer. The moisture absorption of the adhesive layer at time t reflects the amount of moisture absorbed by the adhesive layer at that current moment. This represents the saturated moisture absorption capacity of the adhesive layer, i.e., the maximum amount of moisture the adhesive layer can absorb under specific environmental conditions; the formula for calculating the stress index of the pressure ring is: ,in, This represents the stress index of the pressure ring at time t, used to measure the stress state of the pressure ring at a certain moment. This represents the change in amplitude of the second characteristic peak at time t, i.e., the difference between the amplitude of the second characteristic peak at the current time and the reference amplitude. The difference reflects the influence of factors such as the stress on the pressure ring on the amplitude of the second characteristic peak. The amplitude of the second characteristic peak under the reference state is the amplitude of the second characteristic peak measured under the initial or steady state of the system. This is the strain amplification factor. The maximum absolute value of the strain change of the pressure ring collected at time t reflects the maximum strain change experienced by the pressure ring at that time. To determine the maximum strain of the pressure ring, a stress competition index is obtained based on the calculated adhesive layer stress index and pressure ring stress index. Specifically, the pressure ring stress index is divided by the adhesive layer stress index to reflect the relative competition between the pressure ring stress and the adhesive layer stress. This index allows us to understand the contribution of the pressure ring and adhesive layer to the overall structural stress at a given moment.

[0037] S302, Construct a jump warning model based on the stress competition index, use the jump warning model to predict optical axis jump, and issue a warning based on the prediction results; Furthermore, stress competition index data are collected at certain time intervals. The central difference method is then used to calculate the ratio of the difference in stress competition index between adjacent moments to the time interval, thereby obtaining the stress index change rate. The stress competition index, stress index change rate, and real-time data used to calculate the pressure ring stress index and adhesive layer stress index are combined to form an input feature vector. A jump warning model is constructed using the random forest algorithm. Historical data is collected and divided into training and testing sets. The training set is used to train the model, and the testing set is used to evaluate the model's performance. The input feature vector is then input into the trained jump warning model. In this model, the optical axis jump warning model outputs the probability of a future jump. Based on actual application needs and risk tolerance, a warning threshold is set. When the jump probability is greater than 0.7, an orange warning is issued, indicating that the probability of the optical axis jumping within the next 24 hours is relatively high. The optical axis should be detected within 24 hours to further assess its status, identify potential problems in a timely manner, and take corresponding measures. When the jump probability is greater than 0.9, a red warning is issued, indicating that the probability of the optical axis jumping within the next 24 hours is the highest. It is recommended to immediately stop the system for inspection to avoid serious damage to the system caused by the optical axis jump and ensure the safe operation of the system.

[0038] S303, based on the stress competition index, uses the jump rule to identify the jump source; calculates the fixation health based on the adhesive layer stress index and the pressure ring stress index, and judges the optical axis stability based on the fixation health.

[0039] Specifically, the source of stress jumps is identified using jump rules based on the stress competition index. The jump rules are as follows: a jump threshold range is set, including a minimum jump threshold and a maximum jump threshold. When the stress competition index is greater than the minimum jump threshold but less than the maximum jump threshold, it indicates that the adhesive layer stress and the pressure ring stress are at a similar level. In this case, the stress competition index value alone cannot definitively determine whether the jump is caused by the adhesive layer or the pressure ring. It is necessary to check the moisture absorption data of the adhesive layer. If the moisture absorption increases, it indicates that the adhesive layer performance has changed due to moisture absorption, thus causing the jump. Analyzing the displacement data of the pressure ring, if there is abnormal slight movement of the pressure ring, it indicates a problem with the pressure ring installation, causing the jump. When the stress competition index is greater than the maximum jump threshold, it indicates that the adhesive layer stress dominates the overall stress. In this case, the jump is determined to be caused by adhesive layer creep. Adhesive layer creep refers to the irreversible deformation of the adhesive layer under long-term stress. Over time, the creep of the adhesive layer weakens its fixing effect on the optical axis. When the creep reaches a certain level, it triggers the jump of the optical axis. When the stress competition index is less than the minimum jump threshold, it indicates that the pressure ring stress dominates the overall stress. At this time, the jump is determined to be caused by pressure ring slippage. Pressure ring slippage is caused by reasons such as loose installation of the pressure ring, impact from external forces, or long-term vibration leading to loosening of the connection between the pressure ring and the optical axis. Pressure ring slippage directly destroys the fixed state of the optical axis, causing the optical axis position to change, thus triggering the jump. The formula for calculating the fixation health based on the adhesive layer stress index and the pressure ring stress index is: ,in, To maintain a fixed health level, This is a weighting coefficient used to measure the proportion of adhesive layer stress in the calculation of composite fixation health indicators. This represents the stress value of the adhesive layer at the current moment. This represents the maximum stress in the adhesive layer. This is a weighting coefficient used to measure the proportion of stress in the calculation of the composite fixation health index. This represents the stress value of the pressure ring at the current moment. The maximum value of the pressure ring stress is given. Based on the above-derived fixation health level, and considering practical application scenarios and experience, three different health level thresholds are set to determine the optical axis stability: high stability threshold, medium stability threshold, and low stability threshold. When the fixation health level is greater than or equal to the high stability threshold, the optical axis is considered to be in a high stability state. At this time, the stress distribution of the adhesive layer and pressure ring is reasonable, and the fixation effect on the optical axis is good, maintaining the existing maintenance plan and monitoring frequency. When the fixation health level is greater than or equal to the medium stability threshold but less than the high stability threshold, the optical axis is in a medium stability state, indicating that the stress of the adhesive layer and pressure ring is low. The overall condition meets the requirements, but there are some potential factors that may affect the stability of the optical axis under long-term use or specific working conditions. Although the optical axis is currently functioning normally, we should increase the monitoring frequency and pay attention to the changing trend of the fixation health. When the fixation health is less than the low stability threshold, the optical axis is in a low stability state. At this time, there are obvious problems with the stress state of the adhesive layer or pressure ring, resulting in poor fixation of the optical axis. The optical axis is prone to displacement or vibration, and immediate measures need to be taken for maintenance and repair. A comprehensive inspection and maintenance of the folding mirror composite fixation system should be carried out, including checking the bonding quality of the adhesive layer, the installation tightness of the pressure ring, and the performance of the materials.

[0040] The technical solutions in the above embodiments of this application have at least the following technical effects or advantages: by using the jump warning model to predict the probability of jump in the next 24 hours, the detection is upgraded from post-discovery to pre-warning, and detection or shutdown inspection can be arranged in advance to avoid sudden loss of aiming accuracy, improve the reliability and safety of equipment operation, realize comprehensive perception of the composite fixed state of the folding mirror and accurate tracing and early warning of optical axis jump, and improve the accuracy and timeliness of optical axis stability detection.

[0041] Example 4: Figure 4 As shown, this embodiment provides an optical axis stability detection system for a thermal imaging device, including: an image acquisition module, a calculation module, and a stability judgment module. The image acquisition module is used to control the thermal imaging device under test through a controller, observe a multi-layer irregular target, and acquire an initial image and a secondary image after a preset evaluation operation. The calculation module is used to calculate multi-dimensional offsets such as translational offset, rotational offset, torsional offset, and axial offset. The stability judgment module is used to judge the distortion type using preset judgment logic, calculate the comprehensive offset, and judge whether the optical axis stability is qualified according to a preset comprehensive threshold. The image acquisition module is connected to the calculation module, and the calculation module is connected to the stability judgment module.

[0042] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0043] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0044] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0045] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0046] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0047] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0048] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for detecting the optical axis stability of a thermal imaging device, characterized in that, include: S101, Build an optical axis detection platform and obtain the initial image of the thermal imaging device under test based on the optical axis detection platform; Perform a preset evaluation operation on the thermal imaging device under test, and obtain a secondary image after the preset evaluation operation; The optical axis detection platform includes a thermal imaging device under test, a collimator, a multi-layer irregular target, a first stage, a second stage, and a controller. The preset evaluation operations include repeated disassembly and assembly, impact and vibration, temperature cycling, and focusing actions; S102, process the initial image and the secondary image, use the distinction rule to identify the feature region, and extract the center coordinates of the feature region to form the first layer target feature set and the second layer target feature set; S103, calculate multidimensional offset based on the first layer target feature set and the second layer target feature set, wherein the multidimensional offset includes translation offset, rotation offset, twist offset and axial offset; S104 uses judgment logic to determine the distortion type based on the multidimensional offset, calculates the comprehensive offset based on translation offset, rotation offset, torsion offset and axial offset, and identifies the optical axis stability based on the comprehensive offset.

2. The method for detecting the optical axis stability of a thermal imaging device according to claim 1, characterized in that, The method for acquiring the initial image is as follows: the controller sends a working command to the thermal imaging device under test, the device observes the multi-layer irregular target through the collimator, the controller controls the thermal imaging device under test to perform focusing operation until the device achieves the clearest imaging of the multi-layer irregular target, and after focusing is completed, the controller acquires the image output by the thermal imaging device under test.

3. The method for detecting the optical axis stability of a thermal imaging device according to claim 2, characterized in that, The distinction rule is to calculate the roundness of the extracted contour. When the roundness is greater than the preset recognition threshold, the contour is round. When the roundness is less than the preset recognition threshold, the contour is square. For the remaining contours, a skeleton extraction algorithm is used for recognition to obtain the number of skeleton endpoints. Contours with four skeleton endpoints are identified as cross-shaped feature regions, and contours with three skeleton endpoints are identified as L-shaped feature regions.

4. The method for detecting the optical axis stability of a thermal imaging device according to claim 3, characterized in that, The steps for identifying optical axis stability based on the comprehensive offset are as follows: compare the comprehensive offset with a preset comprehensive threshold. If the comprehensive offset is less than the preset comprehensive threshold, the optical axis stability is deemed acceptable; if the comprehensive offset is greater than or equal to the preset comprehensive threshold, the optical axis stability is deemed unacceptable.

5. The method for detecting the optical axis stability of a thermal imaging device according to claim 1, characterized in that, Stability testing methods also include: S201, Set up a monitoring module, use the monitoring module to acquire historical data, use judgment rules to determine the convection state based on the historical data; use sensors to identify the working posture under the historical convection state, and construct a posture-convection correspondence table according to the working posture and the corresponding convection state; the historical data includes flow velocity data and temperature data. S202, determine the current convection state using the attitude convection correspondence table based on the current device attitude, adjust the optical axis detection strategy according to the current convection state, and acquire the target image based on the adjusted optical axis detection; S203, based on the target image, execute steps S103 and S104 to determine the stability of the optical axis.

6. The method for detecting the optical axis stability of a thermal imaging device according to claim 5, characterized in that, The rules for determining the convection state are as follows: Within 0-5 minutes after startup, if the temperature gradient is in the initial fluctuation stage and the flow rate is low and gradually increasing, the convection state is determined to be in the initial stage of convection. Within 5-20 minutes after startup, if the temperature gradient gradually increases and stabilizes within the range, and the flow rate continues to rise and reaches a high threshold range, the convection state is determined to be in the development stage of convection. After 20 minutes of startup, if the temperature gradient reaches its maximum value and tends to stabilize, and the flow rate also reaches its maximum value and tends to stabilize, the convection state is determined to be in the stable stage of convection. The initial fluctuation stage is defined as the absolute value of the temperature gradient being less than a preset threshold and the rate of change of the absolute value being greater than the rate of change threshold. The high threshold range refers to the range greater than the maximum value of the upward trend in the initial stage of convection.

7. The method for detecting the optical axis stability of a thermal imaging device according to claim 5, characterized in that, Stability testing methods also include: S301, install ultrasonic sensors and strain sensors to collect reflection data and strain data respectively, calculate the adhesive layer stress index and pressure ring stress index based on the reflection data and strain data, and obtain the stress competition index through the adhesive layer stress index and pressure ring stress index; S302, Construct a jump warning model based on the stress competition index, use the jump warning model to predict optical axis jump, and issue a warning based on the prediction results; S303, based on the stress competition index, uses the jump rule to identify the jump source; calculates the fixation health based on the adhesive layer stress index and the pressure ring stress index, and judges the optical axis stability based on the fixation health.

8. The method for detecting the optical axis stability of a thermal imaging device according to claim 7, characterized in that, The formula for calculating the stress index of the adhesive layer is: ,in, This represents the stress index of the adhesive layer at time t. The frequency shift of the first characteristic peak at time t The center frequency of the first characteristic peak under the reference state. This is the hygroscopic amplification factor. Let t be the amount of moisture absorbed by the adhesive layer. This represents the saturated moisture absorption of the adhesive layer.

9. The method for detecting the optical axis stability of a thermal imaging device according to claim 8, characterized in that, The formula for calculating the stress index of the pressure ring is: ,in, This represents the stress index of the pressure ring at time t. Let be the change in amplitude of the second characteristic peak at time t. The amplitude of the second characteristic peak under the baseline condition. This is the strain amplification factor. The maximum absolute value of the strain change of the pressure ring collected at time t. This represents the maximum strain of the compression ring.

10. A system for detecting the optical axis stability of a thermal imaging device, applied to the method for detecting the optical axis stability of a thermal imaging device as described in any one of claims 1-9, characterized in that, The system includes an image acquisition module, a calculation module, and a stability judgment module; The image acquisition module is used to control the thermal imaging device under test through the controller, observe the multi-layer irregular target and acquire the initial image and the secondary image after the preset evaluation operation; The calculation module is used to calculate multi-dimensional offsets such as translation offset, rotation offset, torsion offset, and axial offset; The stability judgment module is used to judge the distortion type using preset judgment logic, and at the same time, calculate the comprehensive offset and judge whether the optical axis stability is qualified according to the preset comprehensive threshold. The image acquisition module is connected to the calculation module, and the calculation module is connected to the stability judgment module.

Images