Unmanned aerial vehicle image processing method and system based on artificial intelligence

Abstract

The invention relates to the field of artificial intelligence, in particular to an unmanned aerial vehicle image processing method and system based on artificial intelligence, and the system comprises an image collection module, a preprocessing module, an image recognition module, a flight control cooperation module and a user interaction module. The preprocessing module is used for preprocessing a target image and obtaining a processed image, the image recognition module is used for recognizing the processed image and obtaining a resampling index according to the processed image, and the flight control cooperation module is used for adjusting the flight state of the unmanned aerial vehicle. And the user interaction module is used for displaying the acquired processed image and an identification result of the image identification module. According to the method, quantitative judgment on the effectiveness of the image is realized by setting a resampling index formula. When the image quality is poor or a high-risk area is detected, the system can automatically judge whether re-shooting is needed or not, and information omission caused by blind shooting of a traditional unmanned aerial vehicle is avoided.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence, and in particular to an artificial intelligence-based unmanned aerial vehicle image processing method and system. Background Art

[0002] With the rapid development of drone technology, drones are widely used in a variety of fields, including military reconnaissance, agricultural monitoring, power inspections, traffic control, and disaster relief. In particular, drones, with their high maneuverability and low cost, are becoming an important means of aerial data collection, particularly in image acquisition and analysis.

[0003] For example, the prior art disclosed in WO2020107487A1 discloses an image processing method and a drone. The image processing method is applied to a drone, which is equipped with imaging devices in at least two directions. The image processing method comprises: acquiring an image to be processed in each of the at least two directions; determining a first direction in the at least two directions based on the image to be processed in each of the at least two directions, and acquiring a reference value for the first direction; the reference value for the first direction is used to determine whether to update key reference frames corresponding to the at least two directions; and if the reference value for the first direction satisfies a preset condition, updating the key reference frames corresponding to the at least two directions.

[0004] Another typical example is a drone image acquisition method and a drone disclosed in the prior art of WO2019061063A1, wherein the method includes: receiving a take-off command; acquiring a target object and saving the characteristics of the target object; tracking the target object according to the characteristics of the target object to acquire a current image; wherein the current image includes the target object; analyzing the position of the target object in the current image, and acquiring the image if the position of the target object in the current image meets the image acquisition conditions.

[0005] Let's look at the image processing method disclosed in the prior art such as WO2020124355A1, which includes: downsampling the original image n-1 times to determine an image with n resolutions; performing edge-preserving filtering on each image to obtain high-frequency information and filtering results; fusing the high-frequency information of the i-th image with the high-frequency information of the i-1-th image to obtain the i-1-th high-frequency fusion result; and based on the i-1-th high-frequency fusion result, fusing the filtering result of the i-th image with the filtering result of the i-1-th image to obtain the i-1-th filtering fusion result.

[0006] At present, the existing drone image collection method usually only follows a set route and continuously captures images during movement. The shooting method is not adjusted according to the image processing results, which easily leads to omissions or large errors in the captured content. In order to solve the common problems in this field, the present invention has been made. Summary of the Invention

[0007] The purpose of the present invention is to address the current deficiencies and propose an artificial intelligence-based drone image processing method and system.

[0008] In order to overcome the deficiencies of the prior art, the present invention adopts the following technical solutions:

[0009] An artificial intelligence-based unmanned aerial vehicle (UAV) image processing system includes an image acquisition module, a preprocessing module, an image recognition module, a flight control collaboration module, and a user interaction module. The image acquisition module is used to acquire a target image during the flight of the UAV. The preprocessing module is used to preprocess the target image and obtain a processed image. The image recognition module is used to recognize the processed image and obtain a resampling index based on the processed image. The flight control collaboration module is used to adjust the flight state of the UAV based on the resampling index. The user interaction module is used to display the acquired processed image and the recognition result of the image recognition module.

[0010] Furthermore, the image acquisition module includes a visible light camera, a gimbal stabilization device and a shooting parameter adjustment unit. The visible light camera is used to capture the visible light image of the target, the gimbal stabilization device is used to stabilize the visible light camera, and the shooting parameter adjustment unit is used to automatically focus the visible light camera.

[0011] Furthermore, the preprocessing module includes an image enhancement unit, an image conversion unit and a synchronization information unit. The image enhancement unit is used to denoise and enhance the contrast of the target image. The image conversion unit is used to convert the format of the enhanced image into a format that conforms to the input of the image recognition module. The synchronization information unit is used to mark the format-converted image according to the current positioning and current time.

[0012] Furthermore, the image recognition module includes an image quality detection unit, a target detection unit, a storage unit and a calculation unit. The image quality detection unit is used to detect various parameters of the processed image, the target detection unit is used to perform image segmentation on the processed image and identify the target detection object in the processed image, the storage unit is used to store the preset parameters required for the calculation unit to perform calculations, and the calculation unit is used to calculate the resampling index of the drone at the current shooting location based on the detection results of the target detection unit.

[0013] Furthermore, the flight control collaboration module includes a judgment unit and a control command generation unit. The judgment unit is used to judge whether the drone needs to be adjusted based on the resampling index. The control command generation unit is used to generate a corresponding control command and send it to the drone when the judgment unit determines that adjustment is required.

[0014] An artificial intelligence-based drone image processing method, which is applied to an artificial intelligence-based drone image processing system, comprises the following steps:

[0015] S1, the image acquisition module collects target images during the flight of the UAV;

[0016] S2, the preprocessing module preprocesses the target image and obtains a processed image;

[0017] S3, the image recognition module recognizes the processed image and obtains the resampling index of the processed image;

[0018] S4, the user interaction module displays the acquired target image and the recognition result of the image recognition module;

[0019] S5, the flight control collaborative module determines whether the flight state of the drone needs to be adjusted based on the resampling index. If so, a control command is generated and sent to the drone. Otherwise, the process ends.

[0020] Furthermore, the image recognition module recognizes the processed image and obtains the resampling index of the processed image, including the following steps:

[0021] S31, the image quality detection unit detects various parameters of the processed image;

[0022] S32, the target detection unit performs image segmentation on the processed image and identifies the target detection object in the segmented image;

[0023] S33, the storage unit outputs the preset parameters required for calculation to the calculation unit;

[0024] S34, the calculation unit calculates the resampling index according to the detection result of the target detection unit and the obtained parameters.

[0025] The beneficial effects achieved by this invention are as follows: 1. By setting a resampling index formula, a quantitative judgment of image validity is achieved. When image quality is poor or a high-risk area is detected, the system can automatically determine whether a retake is needed, avoiding the information omission caused by traditional drone blind shooting.

[0026] 2. The system determines whether to adjust the flight status based on the resampling index, and re-shoots the same area with the drone through methods such as circling, improving the integrity and pertinence of image acquisition and reducing the impact of omissions and errors in shooting. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The present invention can be further understood from the following description in conjunction with the accompanying drawings. The components in the figures are not necessarily drawn to scale, but rather the emphasis is placed on illustrating the principles of the embodiments. In different views, the same reference numerals designate the same parts.

[0028] Figure 1 It is a structural schematic diagram of the present invention.

[0029] Figure 2 It is the workflow diagram of the present invention.

[0030] Figure 3 This is a flow chart of the image recognition module of the present invention identifying a processed image and obtaining a resampling index of the processed image.

[0031] Figure 4 FIG. 4 is a relationship diagram of the regional risk parameter, the actual number of target detection objects, and the ideal number of target detection objects according to the present invention. DETAILED DESCRIPTION

[0032] The following is an explanation of the embodiments of the present invention through specific embodiments. Those skilled in the art can understand the advantages and effects of the present invention from the contents disclosed in this specification. The present invention can be implemented or applied through other different specific embodiments, and the details in this specification can also be modified and changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. In addition, the drawings of the present invention are only simple schematic illustrations and are not depicted in actual size. It is stated in advance. The following embodiments will further explain the relevant technical content of the present invention in detail, but the disclosed content is not intended to limit the scope of protection of the present invention.

[0033] Example 1: According to Figure 1 、 Figure 2 、 Figure 3 and Figure 4This embodiment provides an artificial intelligence-based unmanned aerial vehicle image processing system, including an image acquisition module, a preprocessing module, an image recognition module, a flight control collaboration module, and a user interaction module. The image acquisition module is used to acquire a target image during the flight of the unmanned aerial vehicle; the preprocessing module is used to preprocess the target image and obtain a processed image; the image recognition module is used to recognize the processed image and obtain a resampling index based on the processed image; the flight control collaboration module is used to adjust the flight state of the unmanned aerial vehicle based on the resampling index; and the user interaction module is used to display the acquired processed image and the recognition result of the image recognition module.

[0034] Preferably, the processed image includes the positioning and time information of the image, and the image recognition module also transmits the positioning and time information corresponding to the resampling index when transmitting the resampling index to the flight control collaborative module, so that the flight control collaborative module has data reference for subsequent adjustment of the flight status of the UAV according to the resampling index. For example, the flight control collaborative module can determine whether the work task has been completed in the area corresponding to the resampling index, thereby ensuring that the flight adjustment has an accurate geographical reference basis; or the flight control collaborative module can determine whether the work task has been completed within the ideal working time.

[0035] Specifically, the user interaction module includes a display unit and a user instruction receiving unit. The display unit is used to display the target image and the recognition result of the image recognition module. The user instruction receiving unit is used to receive user instructions and send the user instructions to the drone to manually control the drone.

[0036] Furthermore, the image acquisition module includes a visible light camera, a gimbal stabilization device and a shooting parameter adjustment unit. The visible light camera is used to capture the visible light image of the target, the gimbal stabilization device is used to stabilize the visible light camera, and the shooting parameter adjustment unit is used to automatically focus the visible light camera.

[0037] Furthermore, the preprocessing module includes an image enhancement unit, an image conversion unit and a synchronization information unit. The image enhancement unit is used to denoise and enhance the contrast of the target image. The image conversion unit is used to convert the format of the enhanced image into a format that conforms to the input of the image recognition module. The synchronization information unit is used to mark the format-converted image according to the current positioning and current time.

[0038] Furthermore, the image recognition module includes an image quality detection unit, a target detection unit, a storage unit and a calculation unit. The image quality detection unit is used to detect various parameters of the processed image, the target detection unit is used to perform image segmentation on the processed image and identify the target detection object in the processed image, the storage unit is used to store the preset parameters required for the calculation unit to perform calculations, and the calculation unit is used to calculate the resampling index of the drone at the current shooting location based on the detection results of the target detection unit.

[0039] Specifically, the target detection object is set by a technician in this field according to the task that the drone wants to perform. For example, if the drone is to detect pests, the target detection object is a pest. If the drone is to detect human traffic, the target detection object is a human. The target detection unit detects the target detection object in the image through an artificial intelligence recognition model (such as the YOLO series or Faster R-CNN) and outputs the corresponding parameters (the average pest density per leaf or the human density per road, etc. The specific parameter type is set by a technician in this field according to the task). The artificial intelligence recognition model is trained by a technician in this field based on existing technologies and existing data sets.

[0040] Furthermore, the flight control collaboration module includes a judgment unit and a control command generation unit. The judgment unit is used to judge whether the drone needs to be adjusted based on the resampling index. The control command generation unit is used to generate a corresponding control command and send it to the drone when the judgment unit determines that adjustment is required.

[0041] Specifically, the judgment unit determines whether adjustment is required by judging whether the resampling index is greater than the resampling index threshold. When the resampling index is greater than the resampling index threshold, the drone is adjusted. The resampling index threshold preferably has a value range of 0 to 2*ln(3). For example, when the task type is precision detection (such as pest identification), the image quality and target quantity accuracy requirements are high, and the threshold can be set to 0.4. When the task type is general detection (such as crowd detection), the tolerable error is large, and the threshold can be set to 0.6. The specific value is set by those skilled in the art based on experience and the required accuracy. The drone is adjusted by maintaining the shooting pitch angle (the initial shooting pitch angle is preferably selected as follows: 90°, that is, vertically downward, when detecting pests, 75°-90° when detecting large-scale human flows, and 45°-60° when detecting local density, that is, biased towards vertical downward. The lower the flight altitude, the smaller the initial shooting pitch angle) and shooting distance remain unchanged. The drone circles along the vertical line between the center of the shooting location and the ground to retake multiple target images and automatically uses the obstacle avoidance algorithm to avoid obstacles during the circle.

[0042] An artificial intelligence-based drone image processing method, which is applied to an artificial intelligence-based drone image processing system, comprises the following steps:

[0043] S1, the image acquisition module collects target images during the flight of the UAV;

[0044] S2, the preprocessing module preprocesses the target image and obtains a processed image;

[0045] S3, the image recognition module recognizes the processed image and obtains the resampling index of the processed image;

[0046] S4, the user interaction module displays the acquired target image and the recognition result of the image recognition module;

[0047] S5, the flight control collaborative module determines whether the flight state of the drone needs to be adjusted based on the resampling index. If so, a control command is generated and sent to the drone. Otherwise, the process ends.

[0048] Furthermore, the image recognition module recognizes the processed image and obtains the resampling index of the processed image, including the following steps:

[0049] S31, the image quality detection unit detects various parameters of the processed image;

[0050] S32, the target detection unit performs image segmentation on the processed image and identifies the target detection object in the segmented image;

[0051] S33, the storage unit outputs the preset parameters required for calculation to the calculation unit;

[0052] Specifically, the preset parameters include various thresholds and ideal values set by those skilled in the art.

[0053] S34, the calculation unit calculates the resampling index based on the detection result of the target detection unit and the various parameters obtained. Specifically, the resampling index can be calculated according to the following formula:

[0054] CCY=k1*TX+k2*FX

[0055]

[0056]

[0057] Among them, CCY is the resampling index, which is used to measure the necessity of taking additional target images. The larger the index, the greater the necessity. k1 is the image quality weight, TX is the image quality parameter, k2 is the regional risk weight, and FX is the regional risk parameter. The value range of k1 and k2 is between 0 and 1, and k1+k2=1. The specific value is set by those skilled in the art based on the purpose of the image to be taken by the drone. If the purpose is to determine the use of target detection objects such as insect pests, which will cause adverse effects due to excessive use, the value of k2 can be appropriately increased. One optional value is k2 equal to 0.7 and k1 equal to 0.3. If the purpose is to monitor the flow of people and other target detection objects, which have less adverse effects, the value of k1 can be appropriately increased. One optional value is k1 equal to 0.6 and k2 equal to 0.4.

[0058] e is a natural constant, L is the standard deviation of the Laplace gradient of the processed image, which is obtained by the image quality detection unit through OpenCV, l is the lower limit threshold of the gradient difference, which is set by those skilled in the art according to the required image accuracy. The higher the required accuracy, the larger the threshold. M is the number of pixels in each row of the processed image, N is the number of pixels in each column of the processed image, XS a is the gray value of the a-th pixel, xs a is the pixel value after convolution smoothing corresponding to the a-th pixel;

[0059] K is the amplitude parameter, which is set by those skilled in the art between 0.5 and 1.5 based on the conventional number of target detection objects in the image to be captured by the drone. Generally, the larger the number, the smaller the amplitude parameter. The purpose of the amplitude parameter design is to control the impact of the difference between AU and SET on the final result through the slope. The larger the number, the larger the difference value. By reducing K, the impact of large values on the final result is reduced, and the error is reduced. Preferably, the amplitude parameter is reduced by 0.05 for each additional target detection object. When more than 30 target detection objects are reached, the amplitude parameter is constant at the minimum value of 0.5. AU is the actual number of target detection objects, and SET is the ideal number of target detection objects. The ideal number is set by those skilled in the art based on actual conditions. is the tolerance threshold, which is set by those skilled in the art based on the allowable error range (between the ideal value and the actual value of the target detection object). If K is too small (<0.5), FX changes very slowly regardless of the difference between AU and SET. This sluggish response is inappropriate for high-risk tasks such as those with excessive pest density. When K is very large (>1.5), even if the difference between AU and SET is small, FX is already close to 1. The system may frequently misjudge high-risk situations, resulting in excessive resampling or false triggering of circling shots. This can seriously affect system efficiency and drone resource utilization. The range of 0.5 to 1.5 is the optimal compromise in actual tasks. In common applications, the typical difference between AU and SET is between 0 and 20. Setting K = 0.5 to 1.5 within this range ensures that the response is neither sluggish nor excessive.

[0060] The following is the procedure required to calculate the resampling metrics:

[0061] import torch

[0062] import cv2

[0063] import numpy as np

[0064] from PILimportImage

[0065] defload_image(image_path:str)->np.ndarray:

[0066] """Load the image as a grayscale image in OpenCV format"""

[0067] image=cv2.imread(image_path)

[0068] gray_image=cv2.cvtColor(image,cv2.COLOR_BGR2GRAY)

[0069] return gray_image

[0070] def detect_targets_yolov5(image_path:str,target_class_names=['person'])->int:

[0071] """

[0072] Use the YOLOv5 model to detect targets in images and count the number of detected targets

[0073] target_class_names: specified target categories, such as ['person'], ['insect'], etc.

[0074] """

[0075] #Load the pre-trained model, which can be replaced with custom weights

[0076] model=torch.hub.load('ultralytics / yolov5','yolov5s',pretrained=True)

[0077] model.classes=None#Detect all categories

[0078] results = model(image_path)

[0079] detected=results.pandas().xyxy[0]

[0080] #Filter by target category

[0081] count=0

[0082] for_,row in detected.iterrows():

[0083] if row['name']in target_class_names:

[0084] count+=1

[0085] return countdef calculate_CCY_updated(image:np.ndarray,AU:float,SET:float,

[0086] k1:float=0.7,k2:float=0.3,K:float=1.0,

[0087] L:float=1.0,delta:float=5.0,e:float=0.01)->float:

[0088] """

[0089] Calculates the image resampling metric CCY.

[0090] image: input grayscale image

[0091] AU: Actual number of target detection objects

[0092] SET: Ideal number of target detectors

[0093] For other parameters, refer to patent documents

[0094] """

[0095] M,N = image.shape

[0096] total_pixels = M*N

[0097] laplacian=cv2.Laplacian(image,cv2.CV_64F)

[0098] XS = np.abs(laplacian)

[0099] XS_flat = XS.flatten()

[0100] XS_avg = np.mean(XS_flat)

[0101] #TX part

[0102] diff_sum=np.sum(np.abs(XS_flat-XS_avg))

[0103] norm_diff=diff_sum / (total_pixels*255)

[0104] TX_part = 1 / (1+np.exp(1 / L))

[0105] TX_inner=TX_part+norm_diff+e

[0106] TX=np.log(TX_inner)-1if TX_inner>0else-1.0

[0107] #FX section

[0108] FX_exp=-K*abs(AU-SET) / delta

[0109] FX=1 / (1+np.exp(FX_exp))

[0110] #Comprehensive indicators

[0111] CCY=k1*TX+k2*FX

[0112] return CCY

[0113] #Main process call example

[0114] if__name__ == '__main__':

[0115] image_path = 'drone_image.jpg' #Replace with the actual drone image path

[0116] ideal_count = 10 # Ideal number of detections (e.g. 10 pest targets per frame)

[0117] #Step 1: Image loading

[0118] image_gray=load_image(image_path)

[0119] #Step 2: Object Detection

[0120] actual_count = detect_targets_yolov5(image_path, target_class_names = ['person']) #Target categories can be customized

[0121] #Step 3: Indicator calculation

[0122] ccy_value=calculate_CCY_updated(image_gray,AU=actual_count,SET=ideal_count)

[0123] print(f'Number of targets detected: {actual_count}')

[0124] print(f'Calculated resampling index CCY:{ccy_value:.4f}')

[0125] Specifically, such as Figure 4 As shown, Figure 4 Assume that K is 1 and When FX is 5, the relationship between the regional risk parameter FX, the actual number of target detection objects AU, and the ideal number of target detection objects SET.

[0126] The benefits of this solution include: 1. By setting a resampling indicator formula, the system can quantitatively determine the effectiveness of images. When image quality is poor or high-risk areas are detected, the system can automatically determine whether reshooting is necessary, avoiding the information omissions caused by traditional drone blind shooting.

[0127] 2. The system determines whether to adjust the flight status based on the resampling index, and re-shoots the same area with the drone through methods such as circling, improving the integrity and pertinence of image acquisition and reducing the impact of omissions and errors in shooting.

[0128] Embodiment 2: This embodiment should be understood to include all the features of any of the aforementioned embodiments, and further improve upon them, and also include a method for adjusting the flight state of the drone according to the resampling index. When the current shooting distance is greater than the recommended shooting distance (the recommended shooting distance is set by those skilled in the art based on the camera performance), it is believed that shortening the shooting distance can effectively improve the accuracy of the captured image. The shooting distance is adjusted using this method. When the absolute value of the difference between the actual number of target detection objects and the ideal number is still greater than the absolute value of the difference between the actual number of target detection objects and the ideal number after the resampling index and the distance are adjusted, the difference between the actual number of target detection objects and the ideal number is still greater than the absolute value of the difference between the actual number of target detection objects and the ideal number. (tolerance threshold), it is considered that there may be omissions in the number of shots due to the shooting pitch angle, and this method is used to adjust the pitch angle. Otherwise, it continues to adjust according to the resampling index and distance.

[0129] The shooting distance is adjusted according to the following formula:

[0130]

[0131] Among them, D is the shooting distance after adjustment, D0 is the shooting distance before adjustment, e is a natural constant, D set is the recommended shooting distance, CCY is the resampling index, CCY set is the resampling indicator threshold.

[0132] The shooting pitch angle is adjusted according to the following formula:

[0133]

[0134] Among them, C1 is the adjusted first pitch angle, C2 is the adjusted second pitch angle, C0 is the current pitch angle, AU is the actual number of target detection objects, and SET is the ideal number of target detection objects. is the tolerance threshold, CCY is the resampling index, CCY set is the resampling indicator threshold, ver is the averaging function, and 90 and 45 in the formula correspond to 90 and 45 degree pitch angles respectively.

[0135] After adjustment, the drone performs two rounds of shooting at the first and second pitch angles according to the adjusted shooting distance. If the current pitch angle of the drone is in the range of [0°, 45°), it is considered to be a high pitch angle, and shooting is performed at the first pitch angle first (first lower and then raise). If it is in the range of [45°, 90°], it is considered to be a low pitch angle, and shooting is performed at the second pitch angle first (first raise and then lower). In each round, the drone circles along the vertical line between the center of the shooting location and the ground to re-shoot multiple target images.

[0136] The beneficial effects of this embodiment are as follows: when the resampling index is high, the system automatically reduces the shooting distance to increase the proportion of the target area in the imaging, and at the same time increases the images shot from different pitch angles, so as to obtain a more accurate actual number of target detection objects, thereby improving the image clarity and the perceptual resolution of the recognition model, and ensuring the accuracy of subsequent recognition tasks.

[0137] The above disclosure is only a preferred embodiment of the present invention and does not limit the scope of protection of the present invention. Therefore, any equivalent technical changes made by using the contents of the present invention and the drawings are included in the scope of protection of the present invention. In addition, as technology develops, the elements therein may be updated. The above units are only examples. Those skilled in the art can make different designs and adopt corresponding units according to actual needs when implementing this solution.

Claims

1. An artificial intelligence-based drone image processing system, characterized in that: It includes an image acquisition module, a preprocessing module, an image recognition module, a flight control collaboration module and a user interaction module. The image acquisition module is used to acquire target images during the flight of the UAV. The preprocessing module is used to preprocess the target images and obtain processed images. The image recognition module is used to recognize the processed images and obtain resampling indicators based on the processed images. The flight control collaboration module is used to adjust the flight status of the UAV according to the resampling indicators. The user interaction module is used to display the acquired processed images and the recognition results of the image recognition module.

2. The artificial intelligence-based drone image processing system according to claim 1, characterized in that: The image acquisition module includes a visible light camera, a pan-tilt stabilization device and a shooting parameter adjustment unit. The visible light camera is used to capture a visible light image of a target, the pan-tilt stabilization device is used to stabilize the visible light camera, and the shooting parameter adjustment unit is used to automatically focus the visible light camera.

3. The artificial intelligence-based drone image processing system according to claim 1, characterized in that: The preprocessing module includes an image enhancement unit, an image conversion unit and a synchronization information unit. The image enhancement unit is used to denoise and enhance the contrast of the target image. The image conversion unit is used to convert the format of the enhanced image into a format that conforms to the input of the image recognition module. The synchronization information unit is used to mark the format-converted image according to the current positioning and the current time.

4. The artificial intelligence-based drone image processing system according to claim 1, characterized in that: The image recognition module includes an image quality detection unit, a target detection unit, a storage unit and a calculation unit. The image quality detection unit is used to detect various parameters of the processed image, the target detection unit is used to perform image segmentation on the processed image and identify the target detection object in the processed image, the storage unit is used to store the preset parameters required for the calculation unit to perform calculations, and the calculation unit is used to calculate the resampling index of the drone at the current shooting location based on the detection results of the target detection unit.

5. The artificial intelligence-based drone image processing system according to claim 1, characterized in that: The flight control collaboration module includes a judgment unit and a control command generation unit. The judgment unit is used to judge whether the drone needs to be adjusted based on the resampling index. The control command generation unit is used to generate a corresponding control command and send it to the drone when the judgment result of the judgment unit is that adjustment is required.

6. An artificial intelligence-based drone image processing method, applied to the artificial intelligence-based drone image processing system according to claim 1, characterized in that: The method comprises the following steps: S1, the image acquisition module collects target images during the flight of the UAV; S2, the preprocessing module preprocesses the target image and obtains a processed image; S3, the image recognition module recognizes the processed image and obtains the resampling index of the processed image; S4, the user interaction module displays the acquired target image and the recognition result of the image recognition module; S5, the flight control collaborative module determines whether the flight state of the drone needs to be adjusted based on the resampling index. If so, a control command is generated and sent to the drone. Otherwise, the process ends.

7. The method for processing drone images based on artificial intelligence according to claim 6, characterized in that: The image recognition module recognizes the processed image and obtains the resampling index of the processed image, including the following steps: S31, the image quality detection unit detects various parameters of the processed image; S32, the target detection unit performs image segmentation on the processed image and identifies the target detection object in the segmented image; S33, the storage unit outputs the preset parameters required for calculation to the calculation unit; S34, the calculation unit calculates the resampling index according to the detection result of the target detection unit and the obtained parameters.

Images