Method and device for adjusting exposure of camera of unmanned aerial vehicle

Abstract

The invention discloses an unmanned aerial vehicle camera exposure adjusting method and device, and relates to the technical field of image processing. Acquiring a real-time image and real-time illumination intensity shot by a camera of the unmanned aerial vehicle; calculating the pixel brightness value of the real-time image to obtain a brightness standard deviation; calculating the real-time illumination intensity to obtain an illumination intensity ratio; when the brightness standard deviation is greater than a brightness threshold value or the illumination intensity ratio is greater than an illumination threshold value, determining that the current shooting environment of the unmanned aerial vehicle is a strong light source, calling a first exposure adjustment strategy, and adjusting exposure parameters of the camera of the unmanned aerial vehicle based on the first exposure adjustment strategy; and when the brightness standard deviation is smaller than or equal to the brightness threshold value and the illumination intensity ratio is smaller than or equal to the illumination threshold value, determining that the current shooting environment of the unmanned aerial vehicle is a non-strong light source, calling a second exposure adjustment strategy, and adjusting the exposure parameter of the camera of the unmanned aerial vehicle based on the second exposure adjustment strategy. By implementing the provided technical scheme, the overall quality of the image is improved.

Description

Technical Field

[0001] This application relates to the field of image processing technology, specifically to a method and apparatus for adjusting the exposure of a drone camera. Background Technology

[0002] With the rapid development of drone technology, its application in fields such as aerial photography, inspection and mapping, and agricultural plant protection is becoming increasingly widespread. High-quality aerial images are the foundation for realizing the value of these applications, and exposure, as a key parameter that determines the brightness and detail of an image, is crucial for ensuring image quality through automatic and precise adjustment.

[0003] To achieve automatic exposure control, existing drone camera systems typically employ image analysis-based automatic exposure (AE) algorithms. Mainstream AE algorithms include global average metering, center-weighted metering, and zone-based evaluative metering.

[0004] However, existing automatic exposure algorithms still have many limitations when facing complex lighting environments. Traditional metering methods mainly rely on the average brightness value of the image for exposure control, lacking quantitative analysis of the dispersion of brightness distribution in the image. In scenes with strong light sources such as direct sunlight, water surface reflections, and glass curtain wall reflections, because the brightness of some areas in the image is much higher than other areas, even with zonal evaluative metering, the overall exposure will be significantly reduced to avoid excessive highlight clipping. Although this can protect highlight details to some extent, it will cause a large loss of details in midtones and shadows. Moreover, using a uniform metering and adjustment strategy for all scenes fails to design differentiated control schemes for the different lighting characteristics of strong light source scenes and normal scenes, thus affecting image quality. Summary of the Invention

[0005] This application provides a method and apparatus for adjusting the exposure of a drone camera. This method breaks through the limitations of traditional automatic exposure algorithms that rely solely on image analysis and adopt a uniform adjustment strategy. It achieves precise exposure control in strong light source scenarios and significantly improves the image acquisition quality of drones in complex lighting environments.

[0006] Firstly, this application provides a method for adjusting the exposure of a drone camera, the method comprising: The system acquires real-time images captured by a drone camera and reads the real-time illumination intensity from multiple ambient light sensors deployed in different directions on the drone body. It calculates the brightness standard deviation for each pixel in the real-time image, obtains the maximum illumination intensity from the multiple real-time illumination intensities, and averages these intensities to obtain the average illumination intensity. The maximum illumination intensity is divided by the average illumination intensity to obtain the illumination intensity ratio. The system then determines whether the brightness standard deviation or the illumination intensity ratio is greater than a brightness threshold. If the brightness standard deviation or the illumination intensity ratio is greater than the brightness threshold, the drone is confirmed to be in a strong light source environment. A first exposure adjustment strategy is invoked based on the strong light source, and the drone camera's exposure parameters are adjusted accordingly. If the brightness standard deviation is less than or equal to the brightness threshold, and the illumination intensity ratio is less than or equal to the illumination threshold, the drone is confirmed to be in a non-strong light source environment. A second exposure adjustment strategy is invoked based on the non-strong light source environment, and the drone camera's exposure parameters are adjusted accordingly.

[0007] By employing the above technical solution, the standard deviation of brightness in real-time images is calculated, and real-time illumination intensity data collected by multiple ambient light sensors is acquired. This allows for the acquisition of real-time image data from the UAV camera and real-time illumination intensity data from multiple ambient light sensors, establishing a dual judgment mechanism of image analysis and environmental perception. Firstly, by calculating the standard deviation of image pixel brightness, the dispersion of brightness distribution in the image is quantitatively analyzed, effectively identifying locally bright areas caused by strong light sources. Secondly, by calculating the ambient light intensity ratio, the relative differences in illumination intensity from different directions are accurately reflected, reliably detecting the presence of strong light sources. Based on the dual judgment of brightness standard deviation and light intensity ratio, it can accurately distinguish between strong light source scenes and normal scenes, and then adopt targeted exposure adjustment strategies for each. For strong light source scenes, the first exposure adjustment strategy is adopted, which preserves highlight details while fully retaining midtone and shadow details through optimized parameter configuration. For non-strong light source scenes, the second exposure adjustment strategy is adopted to ensure that the overall brightness and darkness of the image are rich and the details are complete. This breaks through the limitations of traditional automatic exposure algorithms that rely solely on image analysis and adopt a uniform adjustment strategy, and achieves precise exposure control in strong light source scenes, significantly improving the image acquisition quality of drones in complex lighting environments.

[0008] Optionally, a first exposure adjustment strategy is retrieved based on a strong light source, and the exposure parameters of the drone camera are adjusted based on the first exposure adjustment strategy. Specifically, this includes: selecting saturated pixels with brightness values ​​greater than or equal to a saturation brightness threshold from the real-time image and counting the target number of saturated pixels; calculating the proportion of the target number in the total number of pixels in the real-time image to obtain the current saturated region proportion; comparing the current saturated region proportion with a preset saturated region proportion and determining a first exposure adjustment amount based on the comparison result; converting the first exposure adjustment amount into a first change in the exposure parameters and updating the exposure parameters of the drone camera based on the first change; acquiring a target image captured with the updated exposure parameters, calculating the saturated pixels in the target image to obtain the target saturated region proportion, until the target saturated region proportion meets a preset condition, at which point the updating of the drone camera's exposure parameters is stopped.

[0009] By employing the aforementioned technical solution, the number of pixels exceeding the saturation brightness threshold is identified and counted from real-time images. By calculating the proportion of saturated pixels in the total number of pixels and dynamically comparing the current saturated area proportion with the preset saturated area proportion, the degree of overexposure can be accurately determined. Based on the comparison result, the required initial exposure adjustment is calculated. This exposure parameter adjustment method, based on saturation ratio, overcomes the limitation of traditional exposure control relying solely on average brightness, enabling more targeted responses to strong light source scenes. After converting the exposure adjustment amount into specific parameter changes, the exposure parameters are iteratively adjusted until the saturated area proportion of the image reaches the expected range. This ensures that the exposure parameter adjustment effectively suppresses overexposure without excessively darkening the overall image, thus preserving details in strong light sources while maintaining normal display effects in other areas.

[0010] Optionally, the current saturated region proportion is compared with the preset saturated region proportion, and the first exposure adjustment amount is determined based on the comparison result. Specifically, this includes: calculating the difference between the current saturated region proportion and the preset saturated region proportion to obtain the saturation error; when the absolute value of the saturation error is less than the preset stable threshold, confirming that the first exposure adjustment amount is 0; when the absolute value of the saturation error is greater than or equal to the preset stable threshold, acquiring the current flight status data of the UAV, extracting the angular velocity value from the flight status data, comparing the angular velocity value with the low-speed motion threshold and the high-speed motion threshold respectively, and determining the motion adjustment factor based on the motion comparison result; comparing the brightness standard deviation with the low contrast threshold and the high contrast threshold respectively, and determining the image content adjustment factor based on the brightness comparison result; multiplying the baseline scaling factor, the motion adjustment factor, and the image content adjustment factor to obtain the first scaling factor, and multiplying the saturation error by the first scaling factor to obtain the first exposure adjustment amount.

[0011] By adopting the above technical solution, the current exposure parameters are maintained when the saturation error is less than the preset threshold, avoiding image jitter caused by unnecessary minor adjustments. When exposure adjustment is required, the angular velocity value is detected and compared with the preset motion threshold to dynamically determine the motion adjustment factor, effectively addressing the impact of different motion states on exposure control during flight. Furthermore, the image content features are evaluated by comparing the standard deviation of the image brightness with the contrast threshold to determine the image content adjustment factor, achieving adaptive adjustment for different scene complexities. The first proportional coefficient is obtained by multiplying the baseline proportional coefficient with the motion adjustment factor and the image content adjustment factor. Multiple influencing factors are fully considered when calculating the first exposure adjustment amount, ensuring the accuracy and stability of exposure parameter adjustment.

[0012] Optionally, the angular velocity value is compared with the low-speed motion threshold and the high-speed motion threshold respectively, and a motion adjustment factor is determined based on the motion comparison results. Specifically, this includes: when the angular velocity value is less than or equal to the low-speed motion threshold, the UAV is confirmed to be in a stable motion state, and the motion adjustment factor is determined as the first adjustment value; when the angular velocity value is greater than the high-speed motion threshold, the UAV is confirmed to be in a high-speed motion state, and the motion adjustment factor is determined as the second adjustment value, where the first adjustment value is greater than the second adjustment value, and the low-speed motion threshold is less than the high-speed motion threshold; when the angular velocity value is greater than the low-speed motion threshold and less than or equal to the high-speed motion threshold, the UAV is confirmed to be in a transitional motion state; based on the transitional motion state, the difference between the angular velocity value and the low-speed motion threshold is calculated to obtain the current speed increment; the difference between the high-speed motion threshold and the low-speed motion threshold is calculated to obtain the total speed range; the current speed increment is divided by the total speed range to obtain the speed range ratio; the difference between the first adjustment value and the second adjustment value is calculated to obtain the adjustment difference; the speed range ratio is multiplied by the adjustment difference to obtain the adjustment value attenuation amount; and the adjustment value attenuation amount is subtracted from the first adjustment value to obtain the motion adjustment factor.

[0013] By adopting the above technical solution, the motion state of the UAV is divided into three states: stable motion, high-speed motion, and transitional motion based on the angular velocity value. Corresponding motion adjustment factors are configured for different states. In the stable motion state, the first adjustment value is used to ensure sensitive adjustment of exposure parameters. In the high-speed motion state, the second adjustment value is used to effectively suppress exposure fluctuations caused by rapid motion. In the transitional motion state between stable and high-speed motion, by calculating the proportion of the current speed increment in the total speed range and multiplying it by the difference in adjustment value, a continuous and smooth change of the motion adjustment factor is achieved. This not only avoids the jump in exposure parameters caused by abrupt changes in motion state, but also realizes the dynamic adaptation of exposure control to motion speed, significantly improving the image stability and clarity of the UAV during variable-speed flight.

[0014] Optionally, the brightness standard deviation of each pixel in the real-time image is calculated. Specifically, this includes: converting the real-time image to grayscale to obtain a grayscale image, where the grayscale value of each pixel is the brightness value of the pixel; calculating the arithmetic mean of the brightness values ​​of all pixels in the grayscale image to obtain the average brightness value; calculating the square of the difference between the brightness value of each pixel in the grayscale image and the average brightness value, and summing the calculation results of all pixels to obtain the sum of squared differences; dividing the sum of squared differences by the total number of pixels in the grayscale image to obtain the brightness variance; and taking the square root of the brightness variance to obtain the brightness standard deviation.

[0015] By adopting the above technical solution, the color image is converted into a grayscale image, which simplifies the computational complexity while retaining the key brightness information. The arithmetic mean of the brightness values ​​of all pixels in the grayscale image is calculated. The sum of the squares of the differences between each pixel brightness value and the average brightness value is calculated and averaged to obtain the brightness variance. Finally, the standard deviation of brightness is obtained by taking the square root, so that the calculation result has a reliable mathematical basis. The obtained standard deviation of brightness can accurately reflect the dispersion of brightness distribution in the image.

[0016] Optionally, a second exposure adjustment strategy is retrieved based on the non-strong light source, and the exposure parameters of the drone camera are adjusted based on the second exposure adjustment strategy. Specifically, this includes: calculating the difference between the average brightness value and the preset brightness value to obtain the brightness error; when the absolute value of the brightness error is less than the brightness stabilization threshold, confirming that the second exposure adjustment amount is 0 and terminating the current adjustment; when the absolute value of the brightness error is greater than or equal to the brightness stabilization threshold, inputting the absolute value of the brightness error into a pre-stored nonlinear gain scheduling table to obtain the second proportional coefficient; multiplying the brightness error by the second proportional coefficient to obtain the second exposure adjustment amount; converting the second exposure adjustment amount into a second change in the exposure parameters, and updating the exposure parameters of the drone camera based on the second change amount.

[0017] By employing the above technical solution, the deviation between the average brightness value and the preset target value is calculated to assess the current exposure state. A reasonable stability threshold is set to filter out minor fluctuations and avoid over-responding to subtle brightness changes. When a significant brightness error is detected, a nonlinear gain scheduling table is used to determine the adjustment of the second proportional coefficient. This nonlinear mapping relationship can adaptively adjust the response intensity according to the degree of brightness error. The final second exposure adjustment amount is obtained by multiplying the brightness error by the dynamically determined second proportional coefficient, ensuring timely response to changes in illumination while avoiding the oscillation and over-adjustment problems that may occur with traditional linear control.

[0018] Optionally, the absolute value of the brightness error is input into a pre-stored nonlinear gain scheduling table to obtain a second proportional coefficient. Specifically, this includes: when the absolute value of the brightness error completely matches the target error sampling point recorded in the nonlinear gain scheduling table, the second proportional coefficient corresponding to the target error sampling point is directly called. The nonlinear gain scheduling table stores a one-to-one correspondence between multiple sampling points including the absolute value of the brightness error and the second proportional coefficient; when the absolute value of the brightness error falls between the first error sampling point and the second error sampling point in the gain scheduling table, linear interpolation is performed on the second proportional coefficients corresponding to the first error sampling point and the second error sampling point to obtain the second proportional coefficient corresponding to the current absolute value of the brightness error. The first error sampling point and the second error sampling point are two adjacent error sampling points in the nonlinear gain scheduling table.

[0019] By adopting the above technical solution, the problem of balancing adjustment speed and stability in traditional fixed-gain control methods is solved. When the detected brightness error precisely matches the sampling point in the nonlinear gain scheduling table, the corresponding second proportional coefficient can be directly invoked, ensuring timely response. For error values ​​falling between adjacent sampling points, a smooth transition of the second proportional coefficient is achieved by introducing a linear interpolation calculation method. Based on the hybrid control strategy of table lookup and interpolation, not only is the rapid adjustment capability under large error conditions guaranteed, but also the stable convergence characteristics under small error conditions are ensured, effectively suppressing overshoot during the adjustment process.

[0020] A second aspect of this application provides an exposure adjustment device for a drone camera. The device includes an acquisition unit, a processing unit, a first adjustment unit, and a second adjustment unit. The acquisition unit acquires a real-time image captured by the drone camera and reads the real-time illumination intensity from multiple ambient light sensors, wherein the multiple ambient light sensors are deployed in different directions on the drone body. The processing unit calculates the brightness value of each pixel in the real-time image to obtain the brightness standard deviation; obtains the maximum illumination intensity from the multiple real-time illumination intensities; averages the multiple real-time illumination intensities to obtain the average illumination intensity; divides the maximum illumination intensity by the average illumination intensity to obtain the illumination intensity ratio; and determines... The system checks whether the standard deviation of brightness is greater than the brightness threshold or whether the light intensity ratio is greater than the light threshold. The first adjustment unit, when the standard deviation of brightness is greater than the brightness threshold or the light intensity ratio is greater than the light threshold, confirms that the current shooting environment of the drone is under strong light. It then retrieves a first exposure adjustment strategy based on the strong light source and adjusts the exposure parameters of the drone camera accordingly. The second adjustment unit, when the standard deviation of brightness is less than or equal to the brightness threshold and the light intensity ratio is less than or equal to the light threshold, confirms that the current shooting environment of the drone is under non-strong light. It then retrieves a second exposure adjustment strategy based on the non-strong light source and adjusts the exposure parameters of the drone camera accordingly.

[0021] In a third aspect, this application provides an electronic device including a processor, a memory, a user interface, and a network interface. The memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory, causing the electronic device to perform the method as described in any of the above-described methods of this application.

[0022] In a fourth aspect, this application provides a computer-readable storage medium storing instructions that, when executed, perform any of the methods described above in this application.

[0023] In summary, one or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: 1. The system calculates the standard deviation of brightness in real-time images and simultaneously acquires real-time illumination intensity data from multiple ambient light sensors. This establishes a dual judgment mechanism combining image analysis and environmental perception. Firstly, by calculating the standard deviation of image pixel brightness, the system quantifies the dispersion of brightness distribution, effectively identifying localized bright areas caused by strong light sources. Secondly, by calculating the ambient light intensity ratio, the system accurately reflects the relative differences in illumination intensity from different directions, reliably detecting the presence of strong light sources. Based on this dual judgment of brightness standard deviation and illumination intensity ratio, the system can accurately distinguish between strong light source scenes and normal scenes, and then employ targeted exposure adjustment strategies accordingly. For strong light source scenes, a first exposure adjustment strategy is used to preserve highlight details while fully retaining midtone and shadow details through optimized parameter configuration. For non-strong light source scenes, a second exposure adjustment strategy is used to ensure rich overall brightness and detail in the image. This overcomes the limitations of traditional automatic exposure algorithms that rely solely on image analysis and employ a uniform adjustment strategy, achieving precise exposure control in strong light source scenes and significantly improving the image acquisition quality of drones in complex lighting environments. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the first process of a method for adjusting the exposure of a drone camera provided in an embodiment of this application; Figure 2 This is a schematic diagram of the second process of a method for adjusting the exposure of a drone camera provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of a drone camera exposure adjustment device provided in an embodiment of this application; Figure 4 This is a schematic diagram of the structure of an electronic device disclosed in an embodiment of this application.

[0025] Explanation of reference numerals in the attached drawings: 301, acquisition unit; 302, processing unit; 303, first adjustment unit; 304, second adjustment unit; 400, electronic device; 401, processor; 402, memory; 403, user interface; 404, network interface; 405, communication bus. Detailed Implementation

[0026] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification 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.

[0027] In the description of the embodiments of this application, the words "for example" or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design that is described as "for example" or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design options. Rather, the use of the words "for example" or "for instance" is intended to present the relevant concepts in a specific manner.

[0028] In the description of the embodiments of this application, the term "multiple" means two or more. For example, multiple systems means two or more systems, and multiple screen terminals means two or more screen terminals. Furthermore, 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 indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of that feature. The terms "comprising," "including," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0029] Global average metering measures the light of all pixels or most of the image and adjusts exposure parameters based on the difference between the average brightness and the preset brightness. Center-weighted metering assigns higher weight to the center area of ​​the image, suitable for shooting scenarios where the subject is in the center of the frame. Zonal evaluative metering divides the image into multiple areas, measures the light of each area separately, and then calculates the overall brightness. These algorithms analyze the brightness distribution of the real-time image and automatically adjust exposure parameters such as shutter speed, aperture, or ISO to make the overall brightness of the image approach the preset target. Some drone camera systems are also equipped with ambient light sensors for auxiliary metering, but these are usually deployed in a single location on the drone body and lack comprehensive perception of light sources from different directions.

[0030] Therefore, how to overcome the limitations of traditional automatic exposure algorithms that rely solely on image analysis and employ a uniform adjustment strategy, and improve the image acquisition quality of drones in complex lighting environments, is a problem that urgently needs to be solved. This application provides a method for adjusting the exposure of a drone camera, applied in a server. The server in this application can be a platform that provides exposure adjustment services for drone cameras. Figure 1 This is a schematic diagram of the first process of a method for adjusting the exposure of a drone camera according to an embodiment of this application. (Refer to...) Figure 1 The method includes the following steps S101-S108.

[0031] S101: Acquire real-time images captured by the drone camera and read the real-time light intensity from multiple ambient light sensors.

[0032] In step S101 above, real-time images are acquired through the drone's camera, and these images are used for subsequent image analysis and exposure control. To comprehensively perceive the lighting conditions in the drone's flight environment, multiple ambient light sensors are deployed at different locations on the drone's body.

[0033] Specifically, an ambient light sensor is installed on the upper surface of the drone to detect the intensity of light from above, facilitating the detection of situations such as direct sunlight at midday. An ambient light sensor is installed on the lower surface of the drone to detect the intensity of light sources below, such as ground reflections and water reflections. An ambient light sensor is installed on each of the left and right sides of the drone to detect the intensity of lateral light sources, facilitating the detection of strong lateral light scenes such as sunrise and sunset. These ambient light sensors all utilize photoelectric conversion, converting received light signals into electrical signals, which are then read in real-time by the drone's data acquisition system. This multi-directional light intensity acquisition provides comprehensive data support for subsequently determining the presence of strong light sources, overcoming the misjudgment problems easily caused by traditional single-direction light measurement.

[0034] For example, when a drone performs a backlighting flight mission, it can promptly identify strong backlighting scenes by comparing the difference in light intensity in the front and rear directions; when flying above the water surface, it can effectively detect strong light interference caused by water surface reflection through data from the sensors below.

[0035] S102: Calculate the brightness value of each pixel in the real-time image to obtain the brightness standard deviation.

[0036] In step S102 above, the brightness value of each pixel in the real-time image is calculated to obtain the brightness standard deviation. Specifically, this includes: converting the real-time image to grayscale to obtain a grayscale image, where the grayscale value of each pixel is the brightness value of the pixel; calculating the arithmetic mean of the brightness values ​​of all pixels in the grayscale image to obtain the average brightness value; calculating the square of the difference between the brightness value of each pixel in the grayscale image and the average brightness value, and summing the calculation results of all pixels to obtain the sum of squared differences; dividing the sum of squared differences by the total number of pixels in the grayscale image to obtain the brightness variance; and taking the square root of the brightness variance to obtain the brightness standard deviation.

[0037] Specifically, to simplify computational complexity and extract pure brightness information, the acquired real-time image is converted to grayscale. For each pixel in the RGB color image, a weighted average method is used to convert it into a grayscale value. For example, the calculation process for converting a pixel with an RGB value of (255, 128, 0) to grayscale is: 255 × 0.299 + 128 × 0.587 + 0 × 0.114 = 151.7, which is rounded to obtain a grayscale value of 152. This conversion method fully considers the different sensitivities of the human eye to red, green, and blue colors, ensuring that the converted grayscale value accurately reflects the actual brightness of the pixel. Through this processing, a 1920 × 1080 resolution color image is converted into a grayscale image containing 2,073,400 grayscale pixels.

[0038] After obtaining the grayscale image, the arithmetic mean of the brightness values ​​of all pixels in the grayscale image is calculated. First, all pixels in the grayscale image are traversed, and the brightness value of each pixel is accumulated. Then, this sum is divided by the total number of pixels to obtain the arithmetic mean. Taking a simplified 4×4 pixel grayscale image as an example, if the brightness values ​​of the 16 pixels are: 100, 120, 150, 160, 130, 140, 170, 180, 110, 125, 145, 155, 135, 145, 165, 175, then the average brightness value is calculated as: (100 + 120 + ... + 175) ÷ 16 = 144.06. The average brightness value reflects the overall brightness of the entire image, providing a benchmark value for subsequent dispersion calculations.

[0039] After obtaining the average brightness value, the deviation between the brightness value of each pixel and the average brightness value is calculated. Each pixel in the grayscale image is iterated again, and the previously calculated average brightness value is subtracted from the brightness value of each pixel to obtain the brightness deviation for that pixel. To highlight larger brightness deviations, each brightness deviation is squared. This not only eliminates the influence of positive and negative deviations but also increases the weight of large deviations. Continuing with the example above, the calculation process for the first pixel is: (100-144.06) 2=1944.36, and so on for all pixels, finally summing these squared values ​​to get the sum of squared differences.

[0040] To eliminate the influence of image size on the calculation results, the sum of squared differences is divided by the total number of pixels in the grayscale image to obtain the luminance variance. This standardization process makes the calculation results for images of different sizes comparable. The luminance variance reflects the dispersion of the luminance distribution in the image; a larger value indicates a more significant difference in luminance within the image. Taking the example above, assuming the sum of squared differences is 28000, the luminance variance is: 28000 ÷ 16 = 1750.

[0041] Finally, the square root of the calculated luminance variance is taken to obtain the luminance standard deviation. Standard deviation is used instead of variance as the final indicator because it has the same dimensions as the original data, making it easier to understand and use. This luminance standard deviation provides a reliable quantitative basis for subsequent judgments about the presence of strong light sources. For example, when a strong light source appears in the image, the luminance standard deviation will be larger because the brightness of local areas is significantly higher than other areas; while in a scene with uniform lighting, the luminance standard deviation will be relatively smaller.

[0042] S103: Obtain the maximum light intensity from multiple real-time light intensities, average the multiple real-time light intensities to obtain the average light intensity, and divide the maximum light intensity by the average light intensity to obtain the light intensity ratio.

[0043] In step S103 above, the real-time illumination intensity is read from the ambient light sensors deployed above, below, to the left, and to the right of the drone. The maximum illumination intensity is then determined from the multiple real-time illumination intensities. The average illumination intensity is then calculated by averaging the multiple real-time illumination intensities. Finally, the maximum illumination intensity is divided by the average illumination intensity to obtain the illumination intensity ratio. A large illumination intensity ratio indicates the presence of a significant strong light source in a certain direction, which may lead to localized overexposure of the image; a ratio close to 1 indicates relatively uniform ambient lighting.

[0044] For example, at a certain moment, the upper sensor detects a light intensity of 100,000 lux (typical midday direct sunlight intensity), the lower sensor detects 10,000 lux (ground reflected light intensity), the left sensor detects 8,000 lux, and the right sensor detects 7,500 lux. By comparing these four light intensity values, the maximum light intensity is determined to be 100,000 lux detected by the upper sensor. The arithmetic mean of the light intensity values ​​from these four directions is calculated as: (100,000 + 10,000 + 8,000 + 7,500) ÷ 4 = 31,375 lux, yielding the average light intensity. Dividing the maximum light intensity by the average light intensity: 100,000 ÷ 31,375 ≈ 3.19, gives the light intensity ratio. This ratio reflects the relative intensity of the strongest light source in the environment relative to the overall lighting environment.

[0045] S104: Determine whether the standard deviation of brightness is greater than the brightness threshold, or whether the ratio of light intensity is greater than the light threshold.

[0046] In S104 above, in order to accurately determine whether there is a strong light source in the current scene, it is necessary to compare the calculated brightness standard deviation and the ratio of light intensity with a pre-calibrated threshold. The pre-calibrated threshold includes a brightness threshold and a light intensity threshold.

[0047] The brightness and illumination thresholds are adaptively adjusted based on historical flight data. Image samples classified as strong light sources and non-strong light sources from the most recent N flights are statistically analyzed. The mean standard deviations μ1 and μ2 of brightness and the variances σ1 and σ2 of brightness for each class of samples are calculated. According to the Bayesian optimal classification principle, when the variances σ1 and σ2 of the two classes of samples are known, the optimal classification threshold is: Brightness threshold = (μ1·σ2 + μ2·σ1) / (σ1 + σ2), where μ1 and μ2 are the mean standard deviations of brightness for strong and non-strong light source samples, respectively. The illumination threshold is updated similarly. During the statistical phase, typical experimental scenarios are selected: midday (sunny) and cloudy or evening (no sun). In sunny scenarios, the drone camera collects a large number of images at different altitudes and angles, such as at altitudes of 100 meters, 200 meters, and 300 meters, capturing images from the front, 45-degree downward, and eye-level directions, respectively. Similarly, corresponding image collection is performed in non-sunny scenarios. Analysis of numerous images collected in sunny scenes revealed that the standard deviation of brightness was typically above 80, and the light intensity ratio was usually above 2.5. Conversely, in sunless scenes, the standard deviation of brightness was typically below 40, and the light intensity ratio was usually below 1.5. By calculating the mean and variance of the standard deviation of brightness from these statistical data, and based on the Bayesian optimal classification principle, a brightness threshold of 60 and a light intensity threshold of 2.0 were ultimately determined. These thresholds effectively distinguish between scenes with and without strong light sources. This dual-criteria recognition mechanism significantly improves the accuracy and reliability of the system in identifying strong light source scenes, providing a precise basis for selecting subsequent adaptive exposure control strategies.

[0048] When the historical flight data is less than 100 times, a conservative preset threshold is adopted: brightness threshold 60, illumination threshold 2.0.

[0049] S105: When the standard deviation of brightness is greater than the brightness threshold, or the ratio of light intensity is greater than the light threshold, it is confirmed that the current shooting environment of the drone is in a strong light source.

[0050] In S105 above, based on the judgment result of the brightness standard deviation and the ratio of light intensity, it is determined whether there is a strong light source in the current shooting environment. A strong light source refers to a directional light source that produces a local bright area in the image, including but not limited to direct sunlight, specular reflection light, strong artificial light source, etc.

[0051] Specifically, when the detected brightness standard deviation is greater than the brightness threshold, it indicates that there is a significant brightness difference in the image, which is usually caused by direct sunlight or reflection. For example, when a drone is performing an aerial photography mission at noon on a sunny day, if the detected brightness standard deviation is 85, exceeding the brightness threshold of 60, the current scene is judged as a strong light source scene.

[0052] Similarly, when the detected light intensity ratio exceeds the light threshold, it indicates the presence of a strong directional light source in the environment. For example, in the morning or evening, when the sun is near the horizon, the ambient light sensor in one direction may detect significantly higher light intensity than in other directions, causing the light intensity ratio to reach 2.8, exceeding the preset threshold of 2.0. This will also indicate that the scene is in a strong light source environment. This determination is used to decide on the appropriate exposure adjustment strategy, because in the presence of a strong light source like the sun, traditional average metering strategies may lead to localized overexposure or underexposure of the image. By promptly identifying strong light source scenes, the exposure control strategy can be adjusted accordingly, such as using more conservative exposure parameters or activating a special exposure compensation mechanism. This improves the system's ability to cope with complex lighting conditions and provides a strong guarantee for obtaining high-quality aerial images.

[0053] S106: Retrieve the first exposure adjustment strategy based on the strong light source, and adjust the exposure parameters of the drone camera based on the first exposure adjustment strategy.

[0054] In S106 above, the first exposure adjustment strategy is retrieved based on the strong light source, and the exposure parameters of the UAV camera are adjusted based on the first exposure adjustment strategy. Specifically, this includes: selecting saturated pixels with brightness values ​​greater than or equal to the saturation brightness threshold from the real-time image and counting the target number of saturated pixels; calculating the proportion of the target number in the total number of pixels in the real-time image to obtain the current saturated region proportion; comparing the current saturated region proportion with the preset saturated region proportion and determining the first exposure adjustment amount based on the comparison result; converting the first exposure adjustment amount into a first change in the exposure parameters and updating the exposure parameters of the UAV camera based on the first change; acquiring the target image captured by the updated exposure parameters, calculating the saturated pixels in the target image to obtain the target saturated region proportion, until the target saturated region proportion meets the preset condition, so as to stop updating the exposure parameters of the UAV camera.

[0055] Specifically, precise image analysis and parameter adjustments are performed to optimize exposure, including the selection and counting of saturated pixels in real-time images. By setting a saturation brightness threshold, each pixel in the image is traversed, and pixels with brightness values ​​greater than or equal to the threshold are marked as saturated pixels. A counter is used to record the number of these saturated pixels, yielding the target number. The core purpose of this step is to accurately identify and quantify overexposed areas in the image. For example, for an 8-bit image, the saturation brightness threshold is set to 235; for a 10-bit image, it is set to 980; for a 12-bit Raw format, it is set to 95%-98% of the full scale based on the sensor's dynamic range. Each pixel in the image is traversed, and pixels with brightness values ​​greater than or equal to the threshold are selected and marked as saturated pixels. The number of saturated pixels in the image is then counted, with a target number of 207,360.

[0056] After determining the target number of saturated pixels in the image, the total number of pixels in the real-time image is obtained. Then, the target number of saturated pixels is divided by the total number of pixels in the image to obtain the current saturated area percentage. This ratio directly reflects the relative size of the overexposed area in the image, providing an important basis for subsequent exposure parameter adjustments. For example, for a 1920×1080 resolution image with a total of 2,073,400 pixels, if the number of saturated pixels is 207,360, then the current saturated area percentage is 10%. This ratio directly reflects the severity of image overexposure.

[0057] Furthermore, the calculated current saturation region percentage is compared with a pre-set preset saturation region percentage. A first exposure adjustment is determined based on the comparison result. The difference between the current saturation region percentage and the preset saturation region percentage is calculated to obtain the saturation error. When the absolute value of the saturation error is less than a preset stability threshold, the first exposure adjustment is confirmed to be 0. When the absolute value of the saturation error is greater than or equal to the preset stability threshold, the current flight status data of the UAV is acquired, and the angular velocity value is extracted from the flight status data. The angular velocity value is compared with the low-speed motion threshold and the high-speed motion threshold respectively, and a motion adjustment factor is determined based on the motion comparison result. The brightness standard deviation is compared with the low-contrast threshold and the high-contrast threshold respectively, and an image content adjustment factor is determined based on the brightness comparison result. The reference scaling factor, the motion adjustment factor, and the image content adjustment factor are multiplied together to obtain the first scaling factor. The saturation error is multiplied by the first scaling factor to obtain the first exposure adjustment.

[0058] Specifically, the saturation error is obtained by subtracting the current saturation region percentage from the preset saturation region percentage. The saturation error value directly reflects the degree of deviation between the current image exposure state and the target state. The absolute value of the saturation error is compared with a preset stability threshold. When the absolute value of the error is less than the preset stability threshold, it indicates that the current exposure state is close enough to the target state. At this time, the first exposure adjustment is set to 0, keeping the current exposure parameters unchanged. For example, if the preset stability threshold is set to 0.2, and the saturation error is 2, the absolute value of the saturation error is less than 0.2, indicating that the current exposure parameters have reached the ideal state, so the first exposure adjustment is set to 0.

[0059] Furthermore, when the absolute value of the saturation error is greater than or equal to the preset stability threshold, the flight status data of the UAV is read and the angular velocity value is extracted. The angular velocity value is compared with low-speed motion thresholds and high-speed motion thresholds respectively. Based on the motion comparison results, a motion adjustment factor is determined. Specifically, this includes: when the angular velocity value is less than or equal to the low-speed motion threshold, the UAV is confirmed to be in a stable motion state, and the motion adjustment factor is determined as the first adjustment value; when the angular velocity value is greater than the high-speed motion threshold, the UAV is confirmed to be in a high-speed motion state, and the motion adjustment factor is determined as the second adjustment value, where the first adjustment value is greater than the second adjustment value, and the low-speed motion threshold is less than the high-speed motion threshold; when the angular velocity value is greater than the low-speed motion threshold and less than or equal to the high-speed motion threshold, the UAV is confirmed to be in a transitional motion state; based on the transitional motion state, the difference between the angular velocity value and the low-speed motion threshold is calculated to obtain the current speed increment; the difference between the high-speed motion threshold and the low-speed motion threshold is calculated to obtain the total speed range; the current speed increment is divided by the total speed range to obtain the speed range ratio; the difference between the first adjustment value and the second adjustment value is calculated to obtain the adjustment difference; the speed range ratio is multiplied by the adjustment difference to obtain the adjustment value attenuation; the adjustment value attenuation is subtracted from the first adjustment value to obtain the motion adjustment factor.

[0060] Specifically, the motion adjustment factor is determined by analyzing the angular velocity value of the drone to achieve dynamic optimization of exposure parameter adjustment. First, the acquired angular velocity value is compared with preset low-speed motion thresholds and high-speed motion thresholds, set at 15 degrees / second and 45 degrees / second respectively, to classify the drone's motion state. When the detected angular velocity value is less than or equal to the low-speed motion threshold, the drone is determined to be in a stable motion state. In this case, image jitter is minimal, allowing for faster exposure adjustment; therefore, the motion adjustment factor is set to the first adjustment value, such as 1.0. When the angular velocity value is greater than the high-speed motion threshold, the drone is determined to be in a high-speed motion state. In this case, image jitter is severe, requiring a reduction in exposure adjustment speed to avoid exposure oscillations; therefore, the motion adjustment factor is set to the second adjustment value, such as 0.3. After determining that the drone is in a high-speed motion state, a multi-frame smoothing filtering mechanism can be enabled to obtain a second adjustment value by weighted averaging of the exposure adjustment of N consecutive frames (N is between 3 and 10). The weighting coefficient increases with the frame number to suppress single-frame abrupt changes. The weighting coefficient wi = i / N (i is the frame number, i = 1, 2, ..., N) makes the latest frame have the largest weight.

[0061] When the detected angular velocity value falls between the low-speed motion threshold and the high-speed motion threshold, the drone is determined to be in a transitional motion state, requiring precise calculation to determine the appropriate motion adjustment factor. First, the difference between the angular velocity value and the low-speed motion threshold is calculated to obtain the current velocity increment, which reflects the degree of deviation of the current motion state from the stable state. Next, the difference between the high-speed motion threshold and the low-speed motion threshold is calculated to obtain a total velocity range of 30 degrees / second, representing the complete range of the transitional state. Then, the current velocity increment is divided by the total velocity range to obtain the velocity range ratio, which reflects the relative position of the current motion state within the transitional range. Next, the adjustment difference between the first and second adjustment values ​​is calculated; for example, subtracting the second adjustment value of 0.3 from the first adjustment value of 1.0 yields 0.7. Multiplying the velocity range ratio by the adjustment difference gives the adjustment value attenuation, which increases linearly with the angular velocity value. Finally, the calculated adjustment value attenuation is subtracted from the first adjustment value to obtain the motion adjustment factor for the current state. This dynamic adjustment mechanism based on the motion state achieves a smooth transition of the motion adjustment factor.

[0062] For example, when the angular velocity is 30 degrees / second, the calculated velocity range ratio is 0.5, the adjustment value attenuation is 0.35, and the final motion adjustment factor is 0.65. This precise mathematical calculation ensures that the motion adjustment factor can be smoothly adjusted according to the changes in the drone's motion state, avoiding abrupt changes in the adjustment factor, thus providing a more stable and reliable exposure control effect. This adaptive adjustment mechanism can effectively avoid exposure instability caused by violent motion while ensuring timely exposure adjustment, significantly improving the image quality of the drone under various motion states. It should be noted that the purpose of reducing the adjustment factor under high-speed motion is: images have motion blur during high-speed motion, and over-adjustment will amplify the impact of the blurred area; lighting conditions change drastically when the gimbal rotates rapidly, requiring a reduced response speed to avoid parameter oscillation; by extending the adjustment period, a smooth transition can be achieved between multiple frames. Experimental data shows that in fast rotation scenarios with angular velocities greater than 45° / s, using a motion adjustment factor of 0.3 compared to 1.0 can reduce the exposure parameter variance by 60%, significantly improving image stability.

[0063] Simultaneously, the brightness distribution characteristics of the current image are analyzed. The standard deviation of brightness is compared with both the low contrast threshold and the high contrast threshold. Based on the brightness comparison results, the image content adjustment factor is determined. Specifically, when the standard deviation of brightness is less than or equal to the low contrast threshold, the current image is confirmed as a low contrast scene, and the image content adjustment factor is determined as the third adjustment value. When the standard deviation of brightness is greater than the high contrast threshold, the current image is confirmed as a high contrast scene, and the image content adjustment factor is determined as the fourth adjustment value. The third adjustment value is greater than the fourth adjustment value, and the low contrast threshold is less than the high contrast threshold. When the standard deviation of brightness is greater than the low contrast threshold and less than or equal to the high contrast threshold, the current image is confirmed as a transitional contrast scene. Based on the transitional contrast scene, the difference between the standard deviation of brightness and the low contrast threshold is calculated to obtain the current contrast increment. The difference between the high contrast threshold and the low contrast threshold is calculated to obtain the total contrast range. The current contrast increment is divided by the total contrast range to obtain the contrast range ratio. The difference between the third adjustment value and the fourth adjustment value is calculated to obtain the adjustment difference. The contrast range ratio is multiplied by the adjustment difference to obtain the adjustment factor attenuation. The adjustment factor attenuation is subtracted from the third adjustment value to obtain the image content adjustment factor.

[0064] Specifically, the image content adjustment factor is determined by analyzing the standard deviation of the image's brightness, enabling precise exposure control for different scene features. First, the standard deviation of the real-time image's brightness is calculated and compared with preset low-contrast and high-contrast thresholds, such as 30 and 90, to distinguish the image's contrast characteristics. When the brightness standard deviation is less than or equal to the low-contrast threshold, the current image is determined to be a low-contrast scene. In this case, the overall image contrast is low, requiring rapid adjustment of exposure parameters to improve image quality. Therefore, the image content adjustment factor is set to the third adjustment value, such as 1.2. When the brightness standard deviation is greater than the high-contrast threshold, the current image is determined to be a high-contrast scene. In this case, careful adjustment of exposure parameters is needed to protect image details, especially dark details. Therefore, the image content adjustment factor is set to the fourth adjustment value, such as 0.8.

[0065] When the detected brightness standard deviation falls between the low and high contrast thresholds, the current image is determined to be in a transitional contrast scene. First, the difference between the brightness standard deviation and the low contrast threshold is calculated to obtain the current contrast increment, reflecting the degree of contrast change in the current scene relative to the low contrast scene. Next, the difference between the high and low contrast thresholds is calculated to obtain the total contrast interval, representing the complete range of transitional contrast. The current contrast increment is then divided by the total contrast interval to obtain the contrast interval ratio, reflecting the relative position of the current scene within the transition interval. Next, the adjustment difference between the third and fourth adjustment values ​​is calculated, i.e., the third adjustment value 1.2 minus the fourth adjustment value 0.8, resulting in an adjustment difference of 0.4. Multiplying the contrast interval ratio by the adjustment difference yields the adjustment factor attenuation, which increases linearly with the brightness standard deviation. Finally, the calculated adjustment factor attenuation is subtracted from the third adjustment value to obtain the image content adjustment factor for the current scene. This scene-feature-based dynamic adjustment mechanism achieves a smooth transition of the image content adjustment factor. In low-contrast scenes, a larger third adjustment value speeds up the adjustment of exposure parameters, quickly suppressing saturated areas and achieving rapid exposure convergence. In high-contrast scenes, a smaller fourth adjustment value ensures smooth exposure adjustment, suppressing highlights while preserving shadow details to the maximum extent, avoiding detail loss due to overly rapid adjustments. In transitional contrast scenes, precise mathematical calculations achieve smooth changes in the adjustment factor, ensuring the continuity and stability of exposure adjustments.

[0066] After obtaining the motion adjustment factor and image content adjustment factor, a large amount of image data under different scenes and lighting conditions was collected and stored, recording the changes in saturation error before and after each exposure parameter adjustment. By statistically analyzing this historical data, a correspondence between saturation error and the optimal exposure adjustment amount was established, thereby determining the most suitable baseline scaling factor. This baseline scaling factor ensures the effectiveness of the adjustment while avoiding system instability caused by excessive adjustment. The baseline scaling factor, motion adjustment factor, and image content adjustment factor are then multiplied to obtain a first scaling factor that comprehensively considers multiple influencing factors. Multiplying this first scaling factor by the saturation error yields the final first exposure adjustment amount.

[0067] Then the first exposure adjustment amount is converted into a specific change in exposure parameters. In this conversion process, shutter speed is adjusted first because this adjustment method has the least impact on image quality. When the shutter speed reaches its limit, the aperture value or ISO value will be adjusted. The specific method for converting the first exposure adjustment amount into the first change in exposure parameters is as follows: According to the principle of exposure triangle, the exposure adjustment amount is converted into the change in EV value. The calculation formula is: ΔEV = log2(current saturation area percentage / preset saturation area percentage). The parameters are adjusted according to the following priority order: (1) If the current shutter speed is between 1 / 8000s and 1 / 30s, the shutter speed is adjusted first. The change amount is: new shutter speed = current shutter speed × 2 (-ΔEV) (2) If the shutter speed has reached its limit, adjust the aperture value (within the available aperture range); (3) If both the aperture and shutter speed have reached their limits, finally adjust the ISO sensitivity. This parameter adjustment priority strategy ensures that image quality is maintained to the greatest extent while achieving exposure control.

[0068] Finally, the target image is captured using the updated exposure parameters, and the saturation pixel count and proportion calculation are repeated on the newly acquired image. The calculated target saturation area proportion is compared with the preset allowable range. When the target saturation area proportion meets the preset conditions, these conditions include: the target saturation area proportion of three consecutive frames falling within the preset allowable range; or the iteration count reaching 10; or the change in exposure parameters being less than the minimum adjustment step size. If the target saturation area proportion meets one of these conditions, it indicates that the ideal exposure effect has been achieved; if the target saturation area proportion still does not meet one of these conditions, a new round of parameter adjustment will continue. This iterative optimization mechanism ensures that the final image effect can avoid overexposure while retaining sufficient detail.

[0069] S107: When the standard deviation of brightness is less than or equal to the brightness threshold and the ratio of light intensity is less than or equal to the light threshold, it is confirmed that the current shooting environment of the drone is in a non-strong light source.

[0070] In step S107 above, when the brightness standard deviation is less than or equal to the brightness threshold, it indicates that the brightness distribution of the image is relatively uniform, and there are no significant local bright areas. Simultaneously, the illuminance ratio is compared with the illuminance threshold; when the illuminance ratio is less than or equal to the brightness threshold, it indicates that there are no significant strong light source areas in the image. Only when both the brightness standard deviation and the illuminance ratio are satisfied can it be confirmed that the drone's current shooting environment is in a non-strong light source state. This dual-judgment mechanism can more accurately identify the lighting characteristics of the shooting environment, avoiding misjudgments that may result from judging a single parameter.

[0071] S108: Based on the second exposure adjustment strategy retrieved from the non-strong light source, the exposure parameters of the drone camera are adjusted according to the second exposure adjustment strategy.

[0072] In S108 above, the second exposure adjustment strategy is retrieved based on the non-strong light source, and the exposure parameters of the drone camera are adjusted based on the second exposure adjustment strategy. Specifically, this includes: calculating the difference between the average brightness value and the preset brightness value to obtain the brightness error; when the absolute value of the brightness error is less than the brightness stability threshold, confirming that the second exposure adjustment amount is 0 and terminating the current adjustment; when the absolute value of the brightness error is greater than or equal to the brightness stability threshold, inputting the absolute value of the brightness error into a pre-stored nonlinear gain scheduling table to obtain the second proportional coefficient; multiplying the brightness error by the second proportional coefficient to obtain the second exposure adjustment amount; converting the second exposure adjustment amount into a second change in the exposure parameters, and updating the exposure parameters of the drone camera according to the second change amount.

[0073] Specifically, when the drone's current shooting environment is determined to be a non-strong light source, the average brightness value of the image is first calculated. This average brightness value is then subtracted from a preset brightness value to obtain the brightness error. The brightness error reflects the degree of deviation between the overall brightness of the current image and the target brightness. For example, if the calculated average brightness value is 80, and the preset brightness value is set to 128, the brightness error is -48. The absolute value of the brightness error is then compared to a brightness stabilization threshold. When the absolute value of the error is less than the brightness stabilization threshold, it indicates that the overall brightness of the current image has reached the ideal state. At this point, the second exposure adjustment is set to 0, and the adjustment process is terminated to avoid unnecessary parameter adjustments affecting image quality.

[0074] Furthermore, when the absolute value of the detected brightness error is greater than or equal to the brightness stability threshold, the exposure parameters need to be adjusted. First, the absolute value of the brightness error is input into a pre-stored nonlinear gain scheduling table to obtain a second proportional coefficient. Specifically, when the absolute value of the brightness error perfectly matches the target error sampling point recorded in the nonlinear gain scheduling table, the second proportional coefficient corresponding to the target error sampling point is directly called. The nonlinear gain scheduling table stores a one-to-one correspondence between multiple sampling points, including the absolute value of the brightness error, and the second proportional coefficient. When the absolute value of the brightness error falls between the first error sampling point and the second error sampling point in the gain scheduling table, linear interpolation is performed on the second proportional coefficients corresponding to the first and second error sampling points to obtain the second proportional coefficient corresponding to the current absolute value of the brightness error. The first and second error sampling points are two adjacent error sampling points in the nonlinear gain scheduling table.

[0075] Specifically, the nonlinear gain scheduling table pre-stores a series of sampling points for the absolute value of brightness error and their corresponding second proportional coefficients. These correspondences are derived from statistical analysis of a large amount of experimental data. The nonlinear gain scheduling table is generated by first setting 100 sampling points for brightness errors ranging from 0 to 100; for each sampling point, at least 20 adjustment experiments are conducted under a standard test scenario; the number of adjustments required to reach a stable state and the final effect are recorded for each experiment; a quadratic polynomial function k(e) = a·e² + b·e + c is used for fitting, where e is the absolute value of brightness error, k is the second proportional coefficient, and a, b, and c are the parameters to be fitted. The fitting residual is minimized using the least squares method: min Σ[k_measured(ei) - k(ei)]², to obtain the optimal parameters. Sampling points are extracted at the key inflection points of the curve and stored in the scheduling table. Key inflection points are defined as: points where the rate of change of the second proportional coefficient between adjacent sampling points exceeds 0.05, or fixed nodes with absolute error values ​​of 10, 30, 50, and 80. In these correspondences, larger absolute values ​​of brightness error correspond to larger second proportional coefficients. This design enables rapid adjustment when image brightness deviations are large. Conversely, smaller absolute values ​​of brightness error correspond to smaller second proportional coefficients, achieving smooth adjustment when brightness approaches the target value and effectively suppressing overshoot during adjustment. For example, when the absolute value of brightness error is 80, the corresponding second proportional coefficient is 1.5; when the absolute value is 50, the corresponding second proportional coefficient is 1.2; when the absolute value is 30, the corresponding second proportional coefficient is 0.9; and when the absolute value is 10, the corresponding second proportional coefficient is 0.6. This non-linear correspondence ensures rapid adjustment when brightness deviations are large, while employing a gentle adjustment strategy when brightness deviations are small, effectively preventing oscillations during adjustment. When the calculated absolute value of brightness error perfectly matches a target error sampling point in the non-linear gain scheduling table, the second proportional coefficient value corresponding to that sampling point is directly read from the table. This direct mapping method eliminates the need for interpolation calculations, improving response speed while ensuring the accuracy of the adjustment parameters. For example, when the calculated absolute value of the brightness error is exactly 50, the corresponding second proportional coefficient 1.2 can be directly called without interpolation calculation. This direct mapping method improves the response speed.

[0076] Furthermore, when the calculated absolute value of the brightness error falls between two adjacent sampling points in the nonlinear gain scheduling table, a more accurate second proportional coefficient needs to be obtained through linear interpolation. First, the interval containing the current absolute value of the brightness error is determined, i.e., the adjacent first and second error sampling points are identified. Then, the relative position of the current absolute value of the brightness error between these two sampling points is calculated, and based on this relative position, linear interpolation is performed on the second proportional coefficients corresponding to the two sampling points. The specific calculation process includes: calculating the positional ratio of the current absolute value of the brightness error between the two sampling points; applying this ratio to the difference between the second proportional coefficients corresponding to the two sampling points; and adding the calculated increment value to the second proportional coefficient corresponding to the first error sampling point, thereby obtaining the accurate second proportional coefficient corresponding to the current absolute value of the brightness error.

[0077] For example, when the absolute value of the brightness error is 40, it is determined that it falls between the first error sampling point 30 and the second error sampling point 50. Subsequently, linear interpolation is performed based on the second proportional coefficients (0.9 and 1.2) corresponding to these two sampling points. First, the relative position ratio of the current absolute value of the brightness error between the two sampling points is calculated, i.e., (40-30) / (50-30)=0.5. This ratio is applied to the difference of the second proportional coefficients corresponding to the two sampling points, (1.2-0.9)=0.3, to obtain the increment to be interpolated, 0.3×0.5=0.15. This increment is added to the second proportional coefficient corresponding to the first error sampling point, i.e., 0.9+0.15=1.05, to obtain the second proportional coefficient corresponding to the current absolute value of the brightness error.

[0078] The second proportional coefficient obtained from the lookup table is multiplied by the brightness error to obtain the second exposure adjustment amount. This second exposure adjustment amount comprehensively considers the direction and magnitude of the brightness deviation, as well as the influence of non-linear gain, enabling more precise exposure control. The second exposure adjustment amount is then converted into a specific change in exposure parameters, i.e., the second change amount, and the drone camera's exposure parameters are updated based on this second change amount. During the parameter update process, shutter speed is adjusted first, as this adjustment method has the least impact on image quality. Only when the shutter speed reaches its limit will other parameters, such as aperture value or ISO sensitivity, be considered for adjustment.

[0079] This adaptive adjustment mechanism based on nonlinear gain can employ different adjustment strategies according to varying degrees of brightness error. It uses a gentle adjustment range when the brightness error is small, and a more aggressive adjustment range when the brightness error is large. This ensures timely exposure adjustment while avoiding image quality problems caused by over-adjustment. Through this precise exposure control strategy, rapid convergence and stable maintenance of image brightness can be achieved in non-strong light source environments, significantly improving the imaging quality of UAVs in complex lighting conditions.

[0080] like Figure 2 As shown, the system acquires real-time images and multiple real-time illumination intensities. It calculates the brightness value of each pixel in the real-time image to obtain the brightness standard deviation, and calculates the illumination ratio for multiple real-time illumination intensities. It then determines whether the brightness standard deviation or the illumination ratio is greater than a brightness threshold. When the brightness standard deviation or the illumination ratio is greater than the illumination threshold, the system is considered to be in a strong light source environment. Based on the strong light source environment, it selects saturated pixels from the real-time image and calculates the current saturated region proportion, then calculates the saturation error between the current saturated region proportion and the preset saturated region proportion. It then determines whether the saturation error is less than or equal to a preset stable threshold. When the saturation error is less than or equal to the preset stable threshold, it confirms that the first exposure adjustment is 0. When the saturation error is greater than the preset stable threshold, it determines the motion adjustment factor based on flight status data and the image content adjustment factor based on the brightness standard deviation. Finally, it adjusts the motion adjustment factor and image content... The first proportional coefficient is obtained by multiplying the section factor and the baseline proportional coefficient. This first proportional coefficient is then multiplied by the saturation error to obtain the first exposure adjustment amount. The exposure parameters of the drone camera are updated based on this first exposure adjustment amount. When the brightness standard deviation is less than or equal to the brightness threshold, and the light intensity ratio is less than or equal to the illumination threshold, the environment is considered a non-strong light source. The brightness error between the average brightness value and the preset brightness value is calculated based on the non-strong light source environment. It is then determined whether the brightness error is less than the brightness stability threshold. If the brightness error is less than the brightness stability threshold, the second exposure adjustment amount is confirmed to be 0. If the brightness error is greater than or equal to the brightness stability threshold, the absolute value of the brightness error is input into the nonlinear gain scheduling table to obtain the second proportional coefficient. This second proportional coefficient is then multiplied by the brightness error to obtain the second exposure adjustment amount. The exposure parameters of the drone camera are updated based on this second exposure adjustment amount. After updating the exposure parameters, the updated exposure parameters are applied for image capture. Depending on whether the drone's current shooting environment is a strong or non-strong light source scene, the corresponding exposure adjustment strategy is selected to adjust the drone camera's exposure parameters. After adjusting the exposure parameters, the adjusted parameters are immediately used for image capture, and the captured image is uploaded. With rapid and accurate parameter adjustments and instant image capture, it can effectively meet the imaging needs in both strong and weak light source environments, significantly improve the overall quality of captured images, and ensure balanced image brightness.

[0081] This application also provides a device for adjusting the exposure of a drone camera. Figure 3 This is a schematic diagram of the structure of a drone camera exposure adjustment device provided in an embodiment of this application, with reference to... Figure 3 The device includes an acquisition unit 301, a processing unit 302, a first adjustment unit 303, and a second adjustment unit 304. The acquisition unit 301 acquires real-time images captured by the drone camera and reads the real-time light intensity of multiple ambient light sensors, wherein the multiple ambient light sensors are deployed in different directions on the drone body. The processing unit 302 calculates the brightness value of each pixel in the real-time image to obtain the brightness standard deviation; obtains the maximum light intensity from multiple real-time light intensities, averages the multiple real-time light intensities to obtain the average light intensity; divides the maximum light intensity by the average light intensity to obtain the light intensity ratio; and determines whether the brightness standard deviation is greater than the brightness threshold or whether the light intensity ratio is greater than the light threshold. The first adjustment unit 303 confirms that the current shooting environment of the drone is under strong light source when the standard deviation of brightness is greater than the brightness threshold or the ratio of light intensity is greater than the light threshold; it retrieves the first exposure adjustment strategy according to the strong light source and adjusts the exposure parameters of the drone camera based on the first exposure adjustment strategy. The second adjustment unit 304 confirms that the current shooting environment of the drone is in a non-strong light source when the standard deviation of brightness is less than or equal to the brightness threshold and the ratio of light intensity is less than or equal to the light threshold; it then retrieves the second exposure adjustment strategy based on the non-strong light source and adjusts the exposure parameters of the drone camera based on the second exposure adjustment strategy.

[0082] In one possible implementation, the processing unit 302 is used to filter out saturated pixels with brightness values ​​greater than or equal to a saturation brightness threshold from the real-time image and count the target number of saturated pixels; calculate the proportion of the target number in the total number of pixels in the real-time image to obtain the current saturated region proportion; compare the current saturated region proportion with a preset saturated region proportion, and determine a first exposure adjustment amount based on the comparison result; the first adjustment unit 303 is used to convert the first exposure adjustment amount into a first change in the exposure parameters, and update the exposure parameters of the drone camera based on the first change; the acquisition unit is used to acquire the target image captured by the updated exposure parameters, calculate the saturated pixels in the target image to obtain the target saturated region proportion, and stop updating the drone camera exposure parameters until the target saturated region proportion meets a preset condition.

[0083] In one possible implementation, processing unit 302 is used to calculate the difference between the current saturated region ratio and the preset saturated region ratio to obtain the saturation error; when the absolute value of the saturation error is less than the preset stable threshold, the first exposure adjustment amount is confirmed to be 0; acquisition unit 301 is used to acquire the current flight status data of the UAV when the absolute value of the saturation error is greater than or equal to the preset stable threshold, extract the angular velocity value from the flight status data, compare the angular velocity value with the low-speed motion threshold and the high-speed motion threshold respectively, and determine the motion adjustment factor based on the motion comparison result; processing unit 302 is used to compare the brightness standard deviation with the low contrast threshold and the high contrast threshold respectively, and determine the image content adjustment factor based on the brightness comparison result; multiply the reference scaling factor, the motion adjustment factor and the image content adjustment factor to obtain the first scaling factor, and multiply the saturation error by the first scaling factor to obtain the first exposure adjustment amount.

[0084] In one possible implementation, the processing unit 302 is configured to: confirm that the UAV is in a stable motion state when the angular velocity value is less than or equal to a low-speed motion threshold, and determine the motion adjustment factor as a first adjustment value; confirm that the UAV is in a high-speed motion state when the angular velocity value is greater than a high-speed motion threshold, and determine the motion adjustment factor as a second adjustment value, wherein the first adjustment value is greater than the second adjustment value, and the low-speed motion threshold is less than the high-speed motion threshold; confirm that the UAV is in a transitional motion state when the angular velocity value is greater than the low-speed motion threshold and less than or equal to the high-speed motion threshold; calculate the difference between the angular velocity value and the low-speed motion threshold according to the transitional motion state to obtain the current speed increment; calculate the difference between the high-speed motion threshold and the low-speed motion threshold to obtain the total speed range; divide the current speed increment by the total speed range to obtain the speed range ratio; calculate the difference between the first adjustment value and the second adjustment value to obtain the adjustment difference; multiply the speed range ratio by the adjustment difference to obtain the adjustment value attenuation amount; and subtract the adjustment value attenuation amount from the first adjustment value to obtain the motion adjustment factor.

[0085] In one possible implementation, the processing unit 302 is used to perform grayscale processing on the real-time image to obtain a grayscale image, wherein the grayscale value of each pixel is the brightness value of the pixel; calculate the arithmetic mean of the brightness values ​​of all pixels in the grayscale image to obtain the average brightness value; calculate the square of the difference between the brightness value of each pixel in the grayscale image and the average brightness value, and sum the calculation results of all pixels to obtain the sum of squared differences; divide the sum of squared differences by the total number of pixels in the grayscale image to obtain the brightness variance; and take the square root of the brightness variance to obtain the brightness standard deviation.

[0086] In one possible implementation, the processing unit 302 is used to calculate the difference between the average brightness value and the preset brightness value to obtain the brightness error; when the absolute value of the brightness error is less than the brightness stabilization threshold, the second exposure adjustment amount is confirmed to be 0, and the current adjustment is terminated; when the absolute value of the brightness error is greater than or equal to the brightness stabilization threshold, the absolute value of the brightness error is input into a pre-stored nonlinear gain scheduling table to obtain the second proportional coefficient; the brightness error is multiplied by the second proportional coefficient to obtain the second exposure adjustment amount; the second adjustment unit 304 is used to convert the second exposure adjustment amount into a second change in the exposure parameters, and update the exposure parameters of the UAV camera according to the second change amount.

[0087] In one possible implementation, the processing unit 302 is used to directly call the second proportional coefficient corresponding to the target error sampling point when the absolute value of the brightness error completely matches the target error sampling point recorded in the nonlinear gain scheduling table. The nonlinear gain scheduling table stores a one-to-one correspondence between multiple sampling points including the absolute value of the brightness error and the second proportional coefficient. When the absolute value of the brightness error falls between the first error sampling point and the second error sampling point in the gain scheduling table, linear interpolation is performed on the second proportional coefficient corresponding to the first error sampling point and the second error sampling point to obtain the second proportional coefficient corresponding to the current absolute value of the brightness error. The first error sampling point and the second error sampling point are two adjacent error sampling points in the nonlinear gain scheduling table.

[0088] It should be noted that the above embodiments of the apparatus are only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the apparatus and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0089] This application also discloses an electronic device. (See reference...) Figure 4 , Figure 4 This application provides a schematic diagram of the structure of an electronic device. The electronic device 400 may include: at least one processor 401, at least one network interface 404, a user interface 403, a memory 402, and at least one communication bus 405.

[0090] The communication bus 405 is used to enable communication between these components.

[0091] The user interface 403 may include a display screen and a camera. Optionally, the user interface 403 may also include a standard wired interface and a wireless interface.

[0092] The network interface 404 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0093] The processor 401 may include one or more processing cores. The processor 401 connects to various parts of the server using various interfaces and lines, and performs various server functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in memory 402, and by calling data stored in memory 402. Optionally, the processor 401 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 401 may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and application requests; the GPU is responsible for rendering and drawing the content required for display; and the modem handles wireless communication. It is understood that the modem may also not be integrated into the processor 401 and may be implemented as a separate chip.

[0094] The memory 402 may include random access memory (RAM) or read-only memory. Optionally, the memory 402 may include a non-transitory computer-readable storage medium. The memory 402 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 402 may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch functionality, sound playback functionality, image playback functionality, etc.), instructions for implementing the various method embodiments described above, etc. The data storage area may store data involved in the various method embodiments described above. Optionally, the memory 402 may also be at least one storage device located remotely from the aforementioned processor 401.

[0095] like Figure 4As shown, the memory 402, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and an application for adjusting the exposure of the drone camera.

[0096] exist Figure 4 In the electronic device 400 shown, the user interface 403 is mainly used to provide an input interface for the user and to obtain the user input data; while the processor 401 can be used to call the application stored in the memory 402 for adjusting the exposure of the drone camera. When executed by one or more processors, the electronic device performs one or more of the methods described in the above embodiments.

[0097] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0098] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0099] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some service interfaces; indirect couplings or communication connections between devices or units may be electrical or other forms.

[0100] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0101] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0102] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0103] The foregoing description is merely an exemplary embodiment of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. Other embodiments of this disclosure will be readily apparent to those skilled in the art upon consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described herein.

Claims

1. A method for adjusting the exposure of a drone camera, characterized in that, The method comprises: acquiring a real-time image captured by a drone camera, and reading real-time light intensities of a plurality of ambient light sensors, wherein the plurality of ambient light sensors are respectively arranged in different directions of a drone body; calculating brightness values of each pixel in the real-time image to obtain a brightness standard deviation; acquiring a maximum light intensity from the plurality of real-time light intensities, and performing average calculation on the plurality of real-time light intensities to obtain an average light intensity; dividing the maximum light intensity by the average light intensity to obtain a light intensity ratio; determining whether the brightness standard deviation is greater than a brightness threshold value or whether the light intensity ratio is greater than a light threshold value; when the brightness standard deviation is greater than the brightness threshold value or the light intensity ratio is greater than the light threshold value, confirming that the current shooting environment of the drone is in a strong light source; according to the strong light source, calling a first exposure adjustment strategy, and adjusting the exposure parameter of the drone camera based on the first exposure adjustment strategy; when the brightness standard deviation is less than or equal to the brightness threshold value and the light intensity ratio is less than or equal to the light threshold value, confirming that the current shooting environment of the drone is in a non-strong light source; according to the non-strong light source, calling a second exposure adjustment strategy, and adjusting the exposure parameter of the drone camera based on the second exposure adjustment strategy.

2. The method of claim 1, wherein, The method according to the strong light source, calling a first exposure adjustment strategy, and adjusting the exposure parameter of the drone camera based on the first exposure adjustment strategy, specifically comprises: selecting saturated pixels with brightness values greater than or equal to a saturated brightness threshold value from the real-time image, and counting a target number of the saturated pixels; calculating a proportion of the target number in a total number of pixels in the real-time image to obtain a current saturated area proportion; comparing the current saturated area proportion with a preset saturated area proportion, and determining a first exposure adjustment amount according to a comparison result; converting the first exposure adjustment amount into a first change amount of the exposure parameter, and updating the exposure parameter of the drone camera according to the first change amount; acquiring a target image captured by the updated exposure parameter, calculating saturated pixels in the target image to obtain a target saturated area proportion, and stopping updating the exposure parameter of the drone camera until the target saturated area proportion meets a preset condition.

3. The method of claim 2, wherein, The method of comparing the current saturated area proportion with the preset saturated area proportion, and determining the first exposure adjustment amount according to a comparison result, specifically comprises: calculating a difference between the current saturated area proportion and the preset saturated area proportion to obtain a saturation error; when an absolute value of the saturation error is less than a preset stability threshold value, confirming that the first exposure adjustment amount is 0; when the absolute value of the saturation error is greater than or equal to the preset stability threshold value, acquiring current flight state data of the drone, extracting an angular velocity value from the flight state data, comparing the angular velocity value with a low-speed motion threshold value and a high-speed motion threshold value respectively, and determining a motion adjustment factor based on a motion comparison result. comparing the luminance standard deviation with a low contrast threshold and a high contrast threshold respectively, determining an image content adjustment factor based on the luminance comparison result; multiplying the reference proportionality coefficient, the motion adjustment factor and the image content adjustment factor to obtain a first proportionality coefficient, multiplying the saturation error by the first proportionality coefficient to obtain the first exposure adjustment amount.

4. The method of claim 3, wherein, The comparison of the angular velocity value with a low-speed motion threshold and a high-speed motion threshold respectively, and determining a motion adjustment factor based on the motion comparison result, specifically includes: when the angular velocity value is less than or equal to the low-speed motion threshold, it is confirmed that the unmanned aerial vehicle is in a stable motion state, and the motion adjustment factor is determined as a first adjustment value; when the angular velocity value is greater than the high-speed motion threshold, it is confirmed that the unmanned aerial vehicle is in a high-speed motion state, and the motion adjustment factor is determined as a second adjustment value, the first adjustment value is greater than the second adjustment value, and the low-speed motion threshold is less than the high-speed motion threshold; when the angular velocity value is greater than the low-speed motion threshold and less than or equal to the high-speed motion threshold, it is confirmed that the unmanned aerial vehicle is in a transition motion state; according to the transition motion state, the difference between the angular velocity value and the low-speed motion threshold is calculated to obtain a current speed increment; the difference between the high-speed motion threshold and the low-speed motion threshold is calculated to obtain a total speed interval; the current speed increment is divided by the total speed interval to obtain a speed interval proportion; the first adjustment value and the second adjustment value are calculated to obtain an adjustment difference value; the speed interval proportion is multiplied by the adjustment difference value to obtain an adjustment value attenuation amount, and the adjustment value attenuation amount is subtracted from the first adjustment value to obtain the motion adjustment factor.

5. The method of claim 1, wherein, The calculation of the luminance value of each pixel in the real-time image to obtain a luminance standard deviation, specifically includes: gray processing the real-time image to obtain a gray image; calculating the arithmetic mean of the luminance values of all pixels in the gray image to obtain an average luminance value; calculating the square of the difference between the luminance value of each pixel in the gray image and the average luminance value, and summing the calculation results of all pixels to obtain a difference square sum; dividing the difference square sum by the total number of pixels in the gray image to obtain a luminance variance; taking the square root of the luminance variance to obtain the luminance standard deviation.

6. The method of claim 5, wherein, According to the second exposure adjustment strategy, the exposure parameters of the unmanned aerial vehicle camera are adjusted based on the second exposure adjustment strategy, specifically including: calculating the difference between the average luminance value and a preset luminance value to obtain a luminance error; when the absolute value of the luminance error is less than a luminance stability threshold, confirming that the second exposure adjustment amount is 0, and terminating this adjustment; when the absolute value of the luminance error is greater than or equal to the luminance stability threshold, inputting the absolute value of the luminance error into a pre-stored nonlinear gain scheduling table to obtain a second proportionality coefficient; multiplying the luminance error by the second proportionality coefficient to obtain a second exposure adjustment amount; The second exposure adjustment amount is converted into a second change amount of an exposure parameter, and an exposure parameter of the UAV camera is updated according to the second change amount.

7. The method of claim 6, wherein, The absolute value of the luminance error is input into a pre-stored nonlinear gain scheduling table to obtain a second proportional coefficient, and the method specifically comprises the following steps: When the absolute value of the luminance error completely matches a target error sampling point recorded in the nonlinear gain scheduling table, a second proportional coefficient corresponding to the target error sampling point is directly called, and the nonlinear gain scheduling table stores a one-to-one correspondence between a plurality of error sampling points including the absolute value of the luminance error and the second proportional coefficient; When the absolute value of the luminance error falls between a first error sampling point and a second error sampling point in the gain scheduling table, a second proportional coefficient corresponding to the first error sampling point and the second error sampling point is calculated by linear interpolation to obtain a second proportional coefficient corresponding to the current absolute value of the luminance error, and the first error sampling point and the second error sampling point are two adjacent error sampling points in the nonlinear gain scheduling table.

8. A device for adjusting the exposure of a drone camera, characterized in that, The device comprises an acquisition unit, a processing unit, a first adjustment unit and a second adjustment unit, The acquisition unit acquires a real-time image photographed by a UAV camera and reads real-time illumination intensities of a plurality of ambient light sensors, wherein the plurality of ambient light sensors are respectively arranged in different directions of a UAV body. The processing unit calculates luminance values of each pixel in the real-time image to obtain a luminance standard deviation, acquires a maximum illumination intensity from the plurality of real-time illumination intensities, performs average calculation on the plurality of real-time illumination intensities to obtain an average illumination intensity, divides the maximum illumination intensity by the average illumination intensity to obtain an illumination intensity ratio, and judges whether the luminance standard deviation is greater than a luminance threshold value or whether the illumination intensity ratio is greater than an illumination threshold value. The first adjustment unit confirms that a current shooting environment of the UAV is in a strong light source when the luminance standard deviation is greater than the luminance threshold value or the illumination intensity ratio is greater than the illumination threshold value, retrieves a first exposure adjustment strategy according to the strong light source, and adjusts an exposure parameter of the UAV camera based on the first exposure adjustment strategy. The second adjustment unit confirms that the current shooting environment of the UAV is in a non-strong light source when the luminance standard deviation is less than or equal to the luminance threshold value and the illumination intensity ratio is less than or equal to the illumination threshold value, retrieves a second exposure adjustment strategy according to the non-strong light source, and adjusts the exposure parameter of the UAV camera based on the second exposure adjustment strategy.

9. An electronic device, comprising: The electronic device comprises a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory to enable the electronic device to perform the method of any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions, and when the instructions are executed, the method of any one of claims 1-7 is performed.

Images