Target area photometry and exposure mapping method and system in transmission tower inspection

Abstract

The present application relates to the technical field of computer vision and unmanned aerial vehicle inspection, in particular to a target area photometry and exposure mapping method and system in transmission tower inspection. The method comprises determining the target area of key components of the tower and its outer expansion ring background area, calculating the target brightness statistics and highlight risk index in the target area, and calculating the exposure compensation in combination with the background brightness statistics; under the constraints of the upper limit of shutter time and the upper limit of sensitivity, the exposure compensation is quickly mapped to the target shutter time, target sensitivity and target exposure compensation value; the mapping parameters are subjected to step limitation, exponential smoothing and hysteresis processing to obtain the exposure parameters; when the exposure is stable and the image quality meets the conditions, a photograph trigger signal is output, forming a continuous closed-loop control. The present application solves the problems of uncontrollable exposure, slow convergence and easy oscillation of the traditional method in strong reflection and high dynamic range scenes, and improves the inspection image quality and defect recognition reliability.

Description

Technical Field

[0001] This invention relates to the fields of computer vision and UAV inspection technology, specifically to a method and system for photometric and exposure mapping of target areas in power transmission tower inspection. Background Technology

[0002] Drone inspection of power transmission towers has become an important tool for power operation and maintenance, and the quality of inspection images directly affects the reliability of defect identification and maintenance decisions. However, in tower inspection scenarios, critical components such as insulators and fittings often exhibit strong reflections, and the images also contain large areas of bright sky background. The small size of the targets and their rapid scale changes pose a severe challenge to exposure control.

[0003] Existing exposure control technologies primarily rely on the camera's built-in automatic exposure mechanism, based on global brightness statistics or zone metering, making it difficult to perform precise exposure optimization for small target components. In scenes with strong reflections and high dynamic range, traditional methods are easily dominated by the sky background, leading to overexposure / saturation or underexposure and loss of detail in the target area. Furthermore, existing methods lack a mechanism to quickly map the target area's brightness statistics into executable parameters such as shutter speed and ISO, and struggle to achieve stable exposure convergence while meeting hard constraints on motion blur resistance and noise control, easily resulting in parameter oscillations and image flicker. In addition, existing technologies lack robust degradation strategies and temporally stable control under dynamic conditions such as changes in target scale, attitude disturbances, and sudden changes in illumination, making it difficult to form a reliable airborne closed loop. Summary of the Invention

[0004] This invention provides a method and system for photometry and exposure mapping of target areas in transmission tower inspection, aiming to solve the problems of uncontrollable exposure, slow convergence and easy oscillation in target areas during transmission tower inspection, and to achieve stable acquisition of high-quality inspection images.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] Methods for photometry and exposure mapping of target areas in transmission tower inspection include:

[0007] S100: Obtain camera preview frames and camera metadata, including current shutter speed, current ISO and current exposure compensation value;

[0008] S200: Determine the target region in the current frame and the ring-shaped background region formed by the expansion of the target region, wherein the target region corresponds to the key components of the tower;

[0009] S300: Calculate the target brightness statistics and highlight risk index within the target area, and calculate the background brightness statistics within the annular background area; the highlight risk index is the proportion of pixels in the target area whose brightness value exceeds the highlight threshold to the total number of pixels in the target area;

[0010] S400: Calculate the exposure compensation amount based on the difference between the target brightness statistic and the reference brightness, the highlight risk index, and the background brightness statistic; reduce the exposure compensation amount when the highlight risk index increases.

[0011] S500: Under the constraints of upper limits on shutter speed and upper limits on ISO, the exposure compensation amount is mapped to the target shutter speed, target ISO, and target exposure compensation value.

[0012] S600: Perform time-series stabilization processing on the target shutter speed, target sensitivity, and target exposure compensation value to obtain the output exposure parameters;

[0013] S700: The exposure parameters are sent to the camera via the camera control interface;

[0014] S800: Determines whether the exposure is stable and whether the image quality meets the preset conditions. If it does, it outputs a photo or video recording trigger signal.

[0015] S900: Repeat steps S100-S800 to form a continuous closed-loop control.

[0016] As a preferred embodiment of the present invention, determining the target region in the current frame includes:

[0017] The detection frame of the key components of the tower is obtained through target detection or target tracking;

[0018] The detection box can be used as the target region, or the detection box can be cropped and then used as the target region.

[0019] As a preferred embodiment of the present invention, S200 further includes:

[0020] Determine whether the target region is valid and stable. If the jump distance of the center position of the target region between consecutive frames exceeds a preset jump threshold, the target region is determined to be unstable.

[0021] When the target area is determined to be invalid or unstable, a downgraded target area is used to replace the target area. The target brightness statistics and exposure compensation are calculated and the exposure parameters are mapped. The downgraded target area is a preset center area of ​​the screen, a gimbal pointing area, or a pre-configured attention area of ​​the task.

[0022] When the target area is detected to have recovered effectively and stably, switch back to using the target area to calculate the target brightness statistics and exposure compensation and map the exposure parameters.

[0023] As a preferred embodiment of the present invention, the method for generating the annular background region formed by the outward expansion of the target region is as follows:

[0024] The target area is expanded outward by a preset pixel distance or a preset expansion ratio to obtain the expanded area;

[0025] The difference region between the extended region and the target region is taken as the annular background region.

[0026] As a preferred embodiment of the present invention, the target brightness statistics include the average brightness value, median brightness value, or quantile brightness value within the target area.

[0027] As a preferred embodiment of the present invention, calculating the exposure compensation amount based on the difference between the target brightness statistic and the reference brightness, the highlight risk index, and the background brightness statistic includes:

[0028] The initial exposure compensation is calculated based on the difference between the target brightness statistic and the reference brightness.

[0029] When the highlight risk index exceeds the preset highlight threshold, reduce the exposure increase component in the initial exposure compensation amount or adjust the initial exposure compensation amount to the exposure decrease direction to obtain the exposure compensation amount after highlight suppression.

[0030] The exposure compensation amount is obtained by adjusting the exposure compensation amount after highlight suppression based on the background brightness statistics.

[0031] As a preferred embodiment of the present invention, mapping the exposure compensation amount to a target shutter speed, target ISO, and target exposure compensation value under the constraints of upper limits on shutter speed and ISO includes:

[0032] Construct a discrete set of shutter speed and ISO sensitivity, wherein the shutter speed of each combination in the discrete set does not exceed the upper limit constraint of the shutter speed and the ISO sensitivity does not exceed the upper limit constraint of the ISO sensitivity;

[0033] Based on the exposure compensation amount, the parameter combination with the closest exposure equivalent value is selected from the discrete set as the target shutter speed and the target sensitivity;

[0034] The residual between the exposure compensation amount and the exposure equivalent value of the selected parameter combination is taken as the target exposure compensation value.

[0035] As a preferred embodiment of the present invention, the timing stabilization process includes:

[0036] The changes in the target shutter speed, target ISO, and target exposure compensation value compared to the corresponding parameters in the previous frame are limited by a step size to obtain the shutter speed, ISO, and exposure compensation value after the step size limitation.

[0037] The shutter speed, ISO, and exposure compensation value after the step limit are subjected to exponential smoothing filtering to obtain the smoothed shutter speed, ISO, and exposure compensation value.

[0038] The smoothed shutter speed, ISO and exposure compensation values ​​are subjected to hysteresis processing. When the parameter changes do not exceed the exit threshold, the exposure parameters of the previous frame are kept unchanged. When the parameter changes exceed the entry threshold, the new parameters are updated to obtain the exposure parameters.

[0039] As a preferred embodiment of the present invention, the step of determining whether the exposure is stable and whether the image quality meets the preset conditions includes:

[0040] Calculate the variation of the exposure parameters over multiple consecutive frames;

[0041] When the change is less than a preset stability threshold, the exposure is determined to be stable;

[0042] Calculate the image sharpness index of the current frame;

[0043] When the image clarity index is greater than the clarity threshold and the target brightness statistic is within the preset brightness range, the image quality is determined to meet the preset conditions.

[0044] This invention also proposes a target area photometry and exposure mapping system for transmission tower inspection, comprising:

[0045] The image acquisition module is used to acquire camera preview frames and camera metadata, including the current shutter speed, current ISO, and current exposure compensation value.

[0046] The target region determination module is used to determine the target region in the current frame and the ring-shaped background region formed by the expansion of the target region, wherein the target region corresponds to the key components of the tower.

[0047] The target area metering module is used to calculate the target brightness statistics and the highlight risk index within the target area, and to calculate the background brightness statistics within the annular background area; the highlight risk index is the proportion of pixels in the target area whose brightness values ​​exceed the highlight threshold to the total number of pixels in the target area;

[0048] The exposure compensation solution module is used to calculate the exposure compensation amount based on the difference between the target brightness statistics and the reference brightness, the highlight risk index, and the background brightness statistics. When the highlight risk index increases, the exposure compensation amount is reduced.

[0049] The parameter fast mapping module is used to map the exposure compensation amount to the target shutter speed, target ISO and target exposure compensation value under the constraints of shutter speed upper limit and ISO upper limit.

[0050] The timing stabilization control module is used to perform timing stabilization processing on the target shutter time, target sensitivity, and target exposure compensation value to obtain the output exposure parameters;

[0051] The camera control interface module is used to send the exposure parameters to the camera through the camera control interface;

[0052] The quality gating and acquisition control module is used to determine whether the exposure is stable and whether the image quality meets the preset conditions. If the conditions are met, it outputs a photo or video recording trigger signal.

[0053] The beneficial effects of this invention are:

[0054] 1. This invention uses the target area of ​​key components of a tower as the exposure optimization center for metering, and introduces a highlight risk suppression mechanism and background ring auxiliary statistics to avoid the dominant influence of bright backgrounds such as the sky on exposure decisions. By comprehensively considering target brightness, highlight clipping rate, and background brightness, precise exposure control of highly reflective components is achieved, effectively reducing the risk of overexposure and saturation, and improving the ability to retain details in the target area.

[0055] 2. This invention rapidly maps exposure compensation amounts to executable shutter speed and ISO parameters under airborne hard constraints, prioritizing the upper limit of shutter speed to suppress motion blur, and compensating for remaining exposure requirements through ISO while controlling noise. This mapping mechanism employs a discrete candidate set lookup table or nearest-neighbor projection method, resulting in low computational complexity and fast response speed. It is suitable for airborne edge computing scenarios with limited resources, providing an efficient and interpretable parameter solution for real-time exposure control.

[0056] 3. This invention employs a combination of step-limiting, exponential smoothing, and hysteresis processing to achieve time-series stabilization control of exposure parameter updates, effectively suppressing frequent switching and parameter oscillations near the threshold. By combining target area validity determination and degradation strategies, it maintains exposure control continuity even when the target is temporarily lost or detection is unstable, and outputs a trigger signal when exposure is stable and quality meets standards, forming a complete airborne closed loop from metering to triggering, thus improving system robustness and engineering usability. Attached Figure Description

[0057] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0058] Figure 1This is a flowchart illustrating the target area photometry and exposure mapping method in the transmission tower inspection of the present invention.

[0059] Figure 2 This is a schematic diagram of the target area and the annular background area of ​​the present invention;

[0060] Figure 3 This is a schematic diagram of the rapid exposure mapping of the present invention;

[0061] Figure 4 This is a schematic diagram of the timing stability control of the present invention;

[0062] Figure 5 This is a schematic diagram of the target area photometry and exposure mapping system for power transmission tower inspection according to the present invention. Detailed Implementation

[0063] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0064] Example 1: As Figure 1 As shown, the target area photometry and exposure mapping method for transmission tower inspection of the present invention includes:

[0065] S100: Obtain camera preview frames and camera metadata, including current shutter speed, current ISO and current exposure compensation value;

[0066] Specifically, the preview frame or video frame of the current frame is obtained through the camera SDK or driver interface. subscript This indicates the current frame number. Simultaneously, camera metadata is read, which includes at least: the current shutter speed. Current ISO and current exposure compensation value .

[0067] in, This refers to the preview or video frames output in real-time from the camera, which can be in color space formats such as RGB or YUV. Current shutter speed. Current ISO sensitivity, measured in seconds or milliseconds. Current exposure compensation value expressed in ISO numerical values ​​(e.g., ISO 100, ISO 400, etc.). These are expressed in EV levels (e.g., -2EV, +1EV, etc.). This metadata reflects the camera's current exposure status and provides a baseline parameter for subsequent exposure compensation calculations.

[0068] S200: Determine the target region in the current frame and the ring-shaped background region formed by the expansion of the target region, wherein the target region corresponds to the key components of the tower;

[0069] Further, determining the target region in the current frame includes:

[0070] The detection frame of the key components of the tower is obtained through target detection or target tracking;

[0071] The detection box can be used as the target region, or the detection box can be cropped and then used as the target region.

[0072] Furthermore, S200 also includes:

[0073] Determine whether the target region is valid and stable. If the jump distance of the center position of the target region between consecutive frames exceeds a preset jump threshold, the target region is determined to be unstable.

[0074] When the target area is determined to be invalid or unstable, a downgraded target area is used to replace the target area. The target brightness statistics and exposure compensation are calculated and the exposure parameters are mapped. The downgraded target area is a preset center area of ​​the screen, a gimbal pointing area, or a pre-configured attention area of ​​the task.

[0075] When the target area is detected to have recovered effectively and stably, switch back to using the target area to calculate the target brightness statistics and exposure compensation and map the exposure parameters.

[0076] Furthermore, the annular background region formed by the outward expansion of the target region is generated as follows:

[0077] The target area is expanded outward by a preset pixel distance or a preset expansion ratio to obtain the expanded area;

[0078] The difference region between the extended region and the target region is taken as the annular background region.

[0079] Specifically, determine the target region of the current frame. The target area The detection frames of key components of the tower (such as insulator strings, hardware connection areas, conductors, etc.) can be obtained through target detection or target tracking algorithms. The detection frames can be directly used as target areas. Alternatively, the detection frame can be appropriately cropped to serve as the target area. This is to remove edge interference areas or adjust the metering range.

[0080] In obtaining the target area Then, an annular background region is generated through an outward expansion operation. Specifically, the target area The outward expansion area is obtained by extending it outward by a preset pixel distance (in pixels) or by a preset expansion ratio, denoted as... , For the outward expansion width. Background area of ​​the ring band. The difference between the extended region and the target region is:

[0081] ;

[0082] like Figure 2 As shown, the target area Located in the preview frame or video frame In the middle, the outer width It can be a fixed pixel value or dynamically determined based on the proportion of the target area size. (Ring-shaped background area) Used to calculate background brightness statistics This provides auxiliary information for calculating exposure compensation.

[0083] In addition, to ensure robust exposure control, the target area needs to be... Perform validity and stability assessments. Let the center of the target region in the previous frame be... The center of the target region in the current frame is Define the jump distance:

[0084] ;

[0085] in, The jump distance;

[0086] When the target area does not exist or Time (of which) (Based on a preset jump threshold), the target region is determined to be unstable. In this case, a downgraded target region is activated. Alternative Subsequent metering and exposure control are performed. The degraded target area can be generated using one of the following methods: a fixed-size rectangular area based on the center of the image; an area projected onto the image plane by the gimbal pointing or the task's point of interest; or a pre-configured area of ​​interest by the inspection task.

[0087] When the target region is detected to be continuous The frame satisfies the stability condition (e.g.) When this happens, the system automatically switches back to the target area. Metering and exposure control are performed to restore accurate exposure optimization for critical components of the tower.

[0088] S300: Calculate the target brightness statistics and highlight risk index within the target area, and calculate the background brightness statistics within the annular background area; the highlight risk index is the proportion of pixels in the target area whose brightness value exceeds the highlight threshold to the total number of pixels in the target area;

[0089] Furthermore, the target brightness statistics include the average brightness value, median brightness value, or quantile brightness value within the target area.

[0090] Specifically, in the target area Robust photometry is performed inside the device to calculate the target brightness statistics. High-light risk indicators and in the annular background region Internal calculation of background brightness statistics .

[0091] For target brightness statistics The average luminance value, median luminance value, or quantile luminance value within the target area can be used. Preferably, the quantile luminance value is used to enhance robustness to localized luminance anomalies. The specific calculation is as follows:

[0092] ;

[0093] in, Indicates a preview frame or video frame. The brightness channel, Describes the quantile operator. This is the quantile parameter, and its value range is... Typical values ​​can be taken as follows: (Corresponding median) or (Corresponding to higher quantiles). Indicates the target area A set of brightness values ​​within the range.

[0094] For high light risk indicators It represents the proportion of pixels in the target area whose brightness value exceeds the highlight threshold, and the calculation formula is:

[0095] ;

[0096] in, For target area The total number of pixels, The saturation threshold is typically chosen to be close to the maximum grayscale value, such as within the normalized brightness range. . As an indicator function, when the brightness value If the value is 1, then the value is 0; otherwise, the value is 0. A higher value indicates a higher risk of highlight saturation in the target area.

[0097] For background brightness statistics In the annular background region The internal calculation method is the same as that used for the target brightness statistics, that is:

[0098] ;

[0099] in, Indicates the background region of the annular zone A set of brightness values ​​within the range. Used to help evaluate the impact of background brightness on target exposure, and to prevent bright backgrounds such as the sky from dominating exposure decisions.

[0100] like Figure 2 As shown, the target area Internal calculation obtained and Annular background area Internal calculation obtained These light measurements will be used to calculate subsequent exposure compensation.

[0101] S400: Calculate the exposure compensation amount based on the difference between the target brightness statistic and the reference brightness, the highlight risk index, and the background brightness statistic; reduce the exposure compensation amount when the highlight risk index increases.

[0102] Further, calculating the exposure compensation based on the difference between the target brightness statistic and the reference brightness, the highlight risk index, and the background brightness statistic includes:

[0103] The initial exposure compensation is calculated based on the difference between the target brightness statistic and the reference brightness.

[0104] When the highlight risk index exceeds the preset highlight threshold, reduce the exposure increase component in the initial exposure compensation amount or adjust the initial exposure compensation amount to the exposure decrease direction to obtain the exposure compensation amount after highlight suppression.

[0105] The exposure compensation amount is obtained by adjusting the exposure compensation amount after highlight suppression based on the background brightness statistics.

[0106] Specifically, the exposure compensation is calculated based on the photometer results. This indicates the extent to which exposure should be increased or decreased. The calculation of exposure compensation takes into account the difference between the target brightness statistic and the reference brightness, the highlight risk index, and the background brightness statistic. The calculation formula is as follows:

[0107] ;

[0108] in, The reference brightness, i.e., the desired brightness level of the target area, typically ranges from [value range missing]. (Within the normalized brightness range); These are the weighting coefficients for the target brightness deviation term, the highlight suppression term, and the background auxiliary term, respectively. To prevent division by zero of small constants, a typical value is [value to be filled in]. .

[0109] The above formula contains three terms:

[0110] First item For the target brightness deviation term, when When this item is positive, it indicates that more exposure is needed; when If this value is negative, it means that exposure needs to be reduced.

[0111] Second item For highlight suppression, when the highlight risk index When increasing the exposure, reduce the tendency to increase exposure or increase the tendency to decrease exposure to avoid further saturation of the target area.

[0112] Third item As a background auxiliary item, when the background brightness... Brightness higher than reference When the background brightness is lower than the reference brightness, this item is negative, reducing the exposure compensation amount to prevent bright backgrounds such as the sky from dominating the exposure decision; when the background brightness is lower than the reference brightness, this item is positive, increasing the exposure compensation amount.

[0113] By adjusting the weighting coefficients It can balance the needs of target exposure optimization and highlight protection in different scenarios. When the target area is too dark, the exposure is increased; when the highlight cropping rate is increased, the exposure gain is suppressed to avoid overexposure; at the same time, the background brightness is used as an auxiliary factor to suppress the dominant influence of the sky background.

[0114] S500: Under the constraints of upper limits on shutter speed and upper limits on ISO, the exposure compensation amount is mapped to the target shutter speed, target ISO, and target exposure compensation value.

[0115] Furthermore, mapping the exposure compensation amount to the target shutter speed, target ISO, and target exposure compensation value under the constraints of upper limits on shutter speed and ISO includes:

[0116] Construct a discrete set of shutter speed and ISO sensitivity, wherein the shutter speed of each combination in the discrete set does not exceed the upper limit constraint of the shutter speed and the ISO sensitivity does not exceed the upper limit constraint of the ISO sensitivity;

[0117] Based on the exposure compensation amount, the parameter combination with the closest exposure equivalent value is selected from the discrete set as the target shutter speed and the target sensitivity;

[0118] The residual between the exposure compensation amount and the exposure equivalent value of the selected parameter combination is taken as the target exposure compensation value.

[0119] Specifically, under airborne hard constraints, the exposure compensation amount... Mapped to target shutter time Target light sensitivity and target exposure compensation value The hard constraints include an upper limit constraint on shutter time. (Used to suppress motion blur) and sensitivity upper limit constraint (Used for noise control).

[0120] First, regarding the exposure compensation amount Step limiting is used to avoid screen flickering caused by excessive exposure changes within a single frame. Step limiting can be expressed as:

[0121] ;

[0122] in, For the amplitude limiting function, This represents the upper limit of the exposure change step per frame, with a typical value range of [value range missing]. EV. Obtained after step limiting. .

[0123] Then, Convert to exposure ratio :

[0124] ;

[0125] Next, under the upper limit constraint of shutter time and ISO sensitivity limit Under these conditions, the exposure magnification will be increased. Mapped to target shutter time and target sensitivity The mapping formula is:

[0126] ;

[0127] ;

[0128] in, The current shutter speed. The current ISO sensitivity. The upper limit constraint on shutter time is used to suppress motion blur and can be estimated from the gimbal angular velocity or the drone velocity; This is an upper limit constraint on photosensitivity, used to control noise.

[0129] The above mapping formula prioritizes satisfying the shutter speed limit constraint to suppress motion blur. When At that time, the shutter speed was set to At this time, the sensitivity remains at 1. ;when At that time, the shutter speed was limited to At this point, adjust the ISO. Compensate for any remaining exposure needs. If the ISO has also reached its maximum. The actual exposure ratio will be less than At this point, the target exposure compensation value can be used. Perform residual compensation.

[0130] In practical deployments, shutter speed and ISO sensitivity are usually discrete stops supported by the camera (e.g., shutter speed is 10 ... (Sensitivity levels include ISO 100, ISO 200, ISO 400, etc.). Therefore, continuous values ​​are calculated. and After that, the constraints must be satisfied. and Select the exposure equivalent value from the discrete set. closest to the target value The combination of parameters.

[0131] like Figure 3 As shown, in the discrete candidate set of shutter speed and ISO, the feasible region is formed by... and Constraints are determined. By looking up a table or using the nearest projection method, the parameter combination that satisfies the constraints and has the closest equivalent exposure value is quickly found; that is, within the feasible region, the parameter combination that minimizes error is selected. Minimum discrete point .

[0132] Finally, the residual between the exposure compensation amount and the exposure equivalent value of the selected parameter combination is used as the target exposure compensation value. . Specifically:

[0133] ;

[0134] in, To compensate for the difference between discrete parameter combinations and continuous exposure magnification, the camera's exposure compensation function can be used. If necessary, the exposure compensation level supported by the camera SDK can be rounded down to the nearest integer.

[0135] Through the above mapping process, exposure compensation can be quickly converted into executable exposure parameters while meeting the hard constraints of shutter speed and ISO limits, providing an efficient and interpretable parameter mapping scheme for airborne real-time deployment.

[0136] S600: Perform time-series stabilization processing on the target shutter speed, target sensitivity, and target exposure compensation value to obtain the output exposure parameters;

[0137] Furthermore, the timing stabilization process includes:

[0138] The changes in the target shutter speed, target ISO, and target exposure compensation value compared to the corresponding parameters in the previous frame are limited by a step size to obtain the shutter speed, ISO, and exposure compensation value after the step size limitation.

[0139] The shutter speed, ISO, and exposure compensation value after the step limit are subjected to exponential smoothing filtering to obtain the smoothed shutter speed, ISO, and exposure compensation value.

[0140] The smoothed shutter speed, ISO and exposure compensation values ​​are subjected to hysteresis processing. When the parameter changes do not exceed the exit threshold, the exposure parameters of the previous frame are kept unchanged. When the parameter changes exceed the entry threshold, the new parameters are updated to obtain the exposure parameters.

[0141] Specifically, regarding the target shutter speed Target light sensitivity and target exposure compensation value Timing stabilization is performed to suppress exposure parameter oscillations and image flicker, resulting in the final output exposure parameters. , and .

[0142] First, the step size of the change in the target exposure parameters compared to the corresponding parameters in the previous frame is limited. Taking shutter speed as an example, let the shutter speed of the previous frame be... The target shutter time in the current frame is Then the shutter time after step limitation for:

[0143] ;

[0144] in, For the amplitude limiting function, To limit the step magnification, the typical value range is: This step limit prevents excessive jumps in exposure parameters between adjacent frames. The step limit method for ISO and exposure compensation values ​​is similar.

[0145] like Figure 4 As shown in (a), the original exposure compensation amount There may be significant fluctuations, which can be obtained through step limits. To ensure that the change in a single frame does not exceed This helps to suppress drastic changes in exposure.

[0146] Then, the parameters after step amplitude limitation are subjected to exponential smoothing filtering to achieve continuous parameter updates. Taking shutter speed as an example, the exponential smoothing formula is:

[0147] ;

[0148] in, This represents the intermediate result after smoothing. This is the smoothing coefficient, and its value range is... Typical value . The larger the value, the smoother the surface, but the slower the response time. The smaller the value, the faster the response time, but the weaker the smoothing effect. The exponential smoothing method is the same for ISO and exposure compensation values.

[0149] like Figure 4 As shown in (b), the parameter changes are more continuous after exponential smoothing, avoiding abrupt changes.

[0150] Finally, hysteresis processing is applied to the parameters after exponential smoothing. Hysteresis processing is achieved by setting an entry threshold. and exit threshold ,Exceed Update to new parameters in time, below The parameters of the previous frame are kept unchanged, and the current state is maintained between the two thresholds, thereby further suppressing small jitters and avoiding frequent switching near the threshold.

[0151] Taking shutter speed as an example, the hysteresis rule is:

[0152] ;

[0153] in, To enter the threshold, The exit threshold must be met. When the parameter changes Update to new parameters in real time; when the change The parameters of the previous frame are kept unchanged; the current state is kept unchanged between two thresholds, forming a hysteresis interval to avoid frequent switching of parameters near the threshold. The hysteresis processing method for ISO and exposure compensation values ​​is the same.

[0154] like Figure 4 As shown in (b), by using hysteresis, the parameters remain stable when fluctuating slightly, avoiding the "jitter" phenomenon caused by frequent updates.

[0155] By combining the above-mentioned step limiting, exponential smoothing, and hysteresis processing, the final exposure parameters are obtained. and This achieves temporal stabilization of exposure control, effectively suppressing exposure oscillations and image flicker.

[0156] S700: The exposure parameters are sent to the camera via the camera control interface;

[0157] Specifically, the time-stabilized exposure parameters are transmitted via the camera control interface. and The data is then sent to the camera. The camera control interface can be an API interface provided by the camera SDK or a driver layer interface; the specific implementation depends on the support of the camera platform.

[0158] When distributing exposure parameters, the shutter speed is distributed according to the camera SDK interface specification. , Send ISO and issue exposure compensation value These settings should be entered into the corresponding registers or parameter configurations on the camera. Parameter formats may differ across camera platforms; for example, shutter speed may need to be converted to camera-specific shutter speed encoding, ISO sensitivity to ISO encoding, and exposure compensation to EV encoding. Before distribution, the parameters must be converted and their range checked according to the camera SDK requirements to ensure they are within the legal range supported by the camera.

[0159] After the exposure parameters are issued, the camera will apply the new exposure settings in the next frame or subsequent frames. Due to the delay in camera hardware response, the actual exposure may take effect 1 to 3 frames later. Therefore, this delay factor must be considered in the closed-loop control process to avoid oscillations caused by overly rapid adjustments.

[0160] S800: Determines whether the exposure is stable and whether the image quality meets the preset conditions. If it does, it outputs a photo or video recording trigger signal.

[0161] Furthermore, the determination of whether the exposure is stable and whether the image quality meets the preset conditions includes:

[0162] Calculate the variation of the exposure parameters over multiple consecutive frames;

[0163] When the change is less than a preset stability threshold, the exposure is determined to be stable;

[0164] Calculate the image sharpness index of the current frame;

[0165] When the image clarity index is greater than the clarity threshold and the target brightness statistic is within the preset brightness range, the image quality is determined to meet the preset conditions.

[0166] Specifically, the exposure stability and image quality are gating and judged. When the stability condition is met for multiple consecutive frames and the image quality meets the standard, a photo or video recording trigger signal is output to allow the shooting operation to be performed; otherwise, the exposure parameters continue to be iteratively updated to form a closed-loop control.

[0167] First, calculate the variation of the exposure parameters over multiple consecutive frames. Taking shutter speed as an example, let's assume continuous... The shutter speed for sending frames is Calculate its magnitude of change:

[0168] ;

[0169] Alternatively, use the standard deviation form:

[0170] ;

[0171] in, The frame number window for stability determination typically takes the value of [value missing]. frame, It is continuous The average shutter speed for each frame. The calculation method for the variation in ISO and exposure compensation values ​​is similar.

[0172] Exposure is considered stable when the change in the current exposure parameters is less than a preset stability threshold. For example, when... and and At that time, the exposure was determined to be stable. Among them, , and These are the stable thresholds for shutter speed, ISO, and exposure compensation, which can be set according to actual application needs.

[0173] Then, the image sharpness index for the current frame is calculated. Image sharpness indices can be calculated using methods such as gradient magnitude, Laplacian operator response, or frequency domain energy. A commonly used sharpness index is gradient-based sharpness evaluation:

[0174] ;

[0175] in, For the image at position gradient at, This represents the gradient magnitude. The larger the value, the clearer the image.

[0176] Finally, it is determined whether the image quality meets the preset conditions. When the image sharpness index... Greater than the resolution threshold And target brightness statistics Within the preset brightness range Within this timeframe, the image quality is determined to meet preset conditions. The preset brightness range is typically set to... (Normalized brightness range) to ensure that the target area is neither too dark nor too bright.

[0177] The comprehensive judgment condition is: when consecutive The frame satisfies the exposure stability condition ( , , And the current frame meets the image quality requirements ( , When the camera is activated, it outputs a photo or video recording trigger signal. This trigger signal can be linked with the gimbal control module or the photo module to perform the actual shooting operation.

[0178] S900: Repeat steps S100-S800 to form a continuous closed-loop control.

[0179] Specifically, steps S100 to S800 are repeated. In each frame processing cycle, the following are completed in sequence: acquiring camera preview frames and metadata, determining the target area and the ring background area, calculating brightness statistics and highlight risk index, solving for exposure compensation, mapping target exposure parameters, performing temporal stabilization processing, sending exposure parameters to the camera, and judging exposure stability and image quality.

[0180] When the S800 determines that the exposure is stable and the image quality meets the preset conditions, it outputs a photo or video recording trigger signal to allow the shooting operation to be performed. At the same time, the closed-loop control continues to run to maintain stable exposure. When the determination result is that the exposure is unstable or the image quality does not meet the standard, it does not output a trigger signal and continues to iteratively update the exposure parameters.

[0181] This closed-loop control operates continuously on a frame-by-frame basis, responding in real-time to dynamic factors such as changes in the target area's position, lighting conditions, and camera movement. Through a complete process of metering, compensation, mapping, stabilization, and data transmission, it achieves continuous exposure optimization for critical components of the tower. The closed-loop control operates at the same frequency as the camera preview frame rate, typically 15-30 frames per second, ensuring the real-time and continuous nature of exposure adjustments.

[0182] Through the aforementioned continuous closed-loop control, the system can automatically adapt to different shooting scenarios during the inspection of power transmission towers and stably output high-quality inspection images that meet the needs of defect identification and maintenance review.

[0183] Example 2: Figure 5 As shown, this embodiment provides a target area photometry and exposure mapping system for power transmission tower inspection, including:

[0184] The image acquisition module is used to acquire camera preview frames and camera metadata, including the current shutter speed, current ISO, and current exposure compensation value.

[0185] The target region determination module is used to determine the target region in the current frame and the ring-shaped background region formed by the expansion of the target region, wherein the target region corresponds to the key components of the tower.

[0186] The target area metering module is used to calculate the target brightness statistics and the highlight risk index within the target area, and to calculate the background brightness statistics within the annular background area; the highlight risk index is the proportion of pixels in the target area whose brightness values ​​exceed the highlight threshold to the total number of pixels in the target area;

[0187] The exposure compensation solution module is used to calculate the exposure compensation amount based on the difference between the target brightness statistics and the reference brightness, the highlight risk index, and the background brightness statistics. When the highlight risk index increases, the exposure compensation amount is reduced.

[0188] The parameter fast mapping module is used to map the exposure compensation amount to the target shutter speed, target ISO and target exposure compensation value under the constraints of shutter speed upper limit and ISO upper limit.

[0189] The timing stabilization control module is used to perform timing stabilization processing on the target shutter time, target sensitivity, and target exposure compensation value to obtain the output exposure parameters;

[0190] The camera control interface module is used to send the exposure parameters to the camera through the camera control interface;

[0191] The quality gating and acquisition control module is used to determine whether the exposure is stable and whether the image quality meets the preset conditions. If the conditions are met, it outputs a photo or video recording trigger signal.

[0192] It should be noted that the target area photometry and exposure mapping system for transmission tower inspection provided in this embodiment of the invention is used to execute all the process steps of the target area photometry and exposure mapping method for transmission tower inspection in the above embodiment. The working principles and beneficial effects of the two are one-to-one, so they will not be described again.

[0193] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for photometric and exposure mapping of target areas during transmission tower inspection, characterized in that, include: S100: Obtain camera preview frames and camera metadata, including current shutter speed, current ISO and current exposure compensation value; S200: Determine the target region in the current frame and the ring-shaped background region formed by the expansion of the target region, wherein the target region corresponds to the key components of the tower; Determining the target region in the current frame includes: The detection frame of the key components of the tower is obtained through target detection or target tracking; The detection box can be used as the target region, or the detection box can be cropped and used as the target region. The S200 further includes: Determine whether the target region is valid and stable. If the jump distance of the center position of the target region between consecutive frames exceeds a preset jump threshold, the target region is determined to be unstable. When the target area is determined to be invalid or unstable, a downgraded target area is used to replace the target area. The target brightness statistics and exposure compensation are calculated and the exposure parameters are mapped. The downgraded target area is a preset center area of ​​the screen, a gimbal pointing area, or a pre-configured attention area of ​​the task. When the target area is detected to have recovered effectively and stably, switch back to using the target area to calculate the target brightness statistics and exposure compensation and map the exposure parameters; The method for generating the annular background region formed by the expansion of the target region is as follows: The target area is expanded outward by a preset pixel distance or a preset expansion ratio to obtain the expanded area; The difference region between the extended region and the target region is taken as the annular background region; S300: Calculate the target brightness statistics and highlight risk index within the target area, and calculate the background brightness statistics within the annular background area; the highlight risk index is the proportion of pixels in the target area whose brightness value exceeds the highlight threshold to the total number of pixels in the target area; S400: Calculate the exposure compensation amount based on the difference between the target brightness statistic and the reference brightness, the highlight risk index, and the background brightness statistic; reduce the exposure compensation amount when the highlight risk index increases. S500: Under the constraints of upper limits on shutter speed and upper limits on ISO, the exposure compensation amount is mapped to the target shutter speed, target ISO, and target exposure compensation value. S600: Perform time-series stabilization processing on the target shutter speed, target sensitivity, and target exposure compensation value to obtain the output exposure parameters; S700: The exposure parameters are sent to the camera via the camera control interface; S800: Determines whether the exposure is stable and whether the image quality meets the preset conditions. If it does, it outputs a photo or video recording trigger signal. S900: Repeat steps S100-S800 to form a continuous closed-loop control.

2. The method for photometry and exposure mapping of target areas in transmission tower inspection according to claim 1, characterized in that, In S300, the target brightness statistics include the average brightness value, median brightness value, or quantile brightness value within the target area.

3. The method for photometry and exposure mapping of target areas in transmission tower inspection according to claim 1, characterized in that, In S400, calculating the exposure compensation amount based on the difference between the target brightness statistic and the reference brightness, the highlight risk index, and the background brightness statistic includes: The initial exposure compensation is calculated based on the difference between the target brightness statistic and the reference brightness. When the highlight risk index exceeds the preset highlight threshold, reduce the exposure increase component in the initial exposure compensation amount or adjust the initial exposure compensation amount to the exposure decrease direction to obtain the exposure compensation amount after highlight suppression. The exposure compensation amount is obtained by adjusting the exposure compensation amount after highlight suppression based on the background brightness statistics.

4. The method for photometry and exposure mapping of target areas in transmission tower inspection according to claim 1, characterized in that, In S500, mapping the exposure compensation amount to a target shutter speed, target ISO, and target exposure compensation value under the constraints of upper limits on shutter speed and ISO includes: Construct a discrete set of shutter speed and ISO sensitivity, wherein the shutter speed of each combination in the discrete set does not exceed the upper limit constraint of the shutter speed and the ISO sensitivity does not exceed the upper limit constraint of the ISO sensitivity; Based on the exposure compensation amount, the parameter combination with the closest exposure equivalent value is selected from the discrete set as the target shutter speed and the target sensitivity; The residual between the exposure compensation amount and the exposure equivalent value of the selected parameter combination is taken as the target exposure compensation value.

5. The method for photometry and exposure mapping of target areas in transmission tower inspection according to claim 1, characterized in that, In S600, the timing stabilization process includes: The changes in the target shutter speed, target ISO, and target exposure compensation value compared to the corresponding parameters in the previous frame are limited by a step size to obtain the shutter speed, ISO, and exposure compensation value after the step size limitation. The shutter speed, ISO, and exposure compensation value after the step limit are subjected to exponential smoothing filtering to obtain the smoothed shutter speed, ISO, and exposure compensation value. The smoothed shutter speed, ISO and exposure compensation values ​​are subjected to hysteresis processing. When the parameter changes do not exceed the exit threshold, the exposure parameters of the previous frame are kept unchanged. When the parameter changes exceed the entry threshold, the new parameters are updated to obtain the exposure parameters.

6. The method for photometry and exposure mapping of target areas in transmission tower inspection according to claim 1, characterized in that, In S800, determining whether the exposure is stable and whether the image quality meets the preset conditions includes: Calculate the variation of the exposure parameters over multiple consecutive frames; When the change is less than a preset stability threshold, the exposure is determined to be stable; Calculate the image sharpness index of the current frame; When the image clarity index is greater than the clarity threshold and the target brightness statistic is within the preset brightness range, the image quality is determined to meet the preset conditions.

7. A photometric and exposure mapping system for target areas during transmission tower inspection, characterized in that, The system is used to execute the target area photometry and exposure mapping method for transmission tower inspection as described in any one of claims 1-6, and the system includes: The image acquisition module is used to acquire camera preview frames and camera metadata, including the current shutter speed, current ISO, and current exposure compensation value. The target region determination module is used to determine the target region in the current frame and the ring-shaped background region formed by the expansion of the target region, wherein the target region corresponds to the key components of the tower. The target area metering module is used to calculate the target brightness statistics and the highlight risk index within the target area, and to calculate the background brightness statistics within the annular background area; the highlight risk index is the proportion of pixels in the target area whose brightness values ​​exceed the highlight threshold to the total number of pixels in the target area; The exposure compensation solution module is used to calculate the exposure compensation amount based on the difference between the target brightness statistics and the reference brightness, the highlight risk index, and the background brightness statistics. When the highlight risk index increases, the exposure compensation amount is reduced. The parameter fast mapping module is used to map the exposure compensation amount to the target shutter speed, target ISO and target exposure compensation value under the constraints of shutter speed upper limit and ISO upper limit. The timing stabilization control module is used to perform timing stabilization processing on the target shutter time, target sensitivity, and target exposure compensation value to obtain the output exposure parameters; The camera control interface module is used to send the exposure parameters to the camera through the camera control interface; The quality gating and acquisition control module is used to determine whether the exposure is stable and whether the image quality meets the preset conditions. If the conditions are met, it outputs a photo or video recording trigger signal.

Images