CN118728654B - A wind turbine tower blade detection system based on drone - Google Patents

Abstract

Translated fromChinese

本发明提供了一种基于无人机的风电塔筒叶片检测系统，所述系统包括机体控制模块、图像采集模块、通信模块、检测诊断模块和能源管理模块；所述机体控制模块用于控制实现无人机的稳定飞行和精确导航；所述通信模块用于完成无人机与地面控制站的数据传输；所述图像采集模块用于在无人机飞行过程中采集风电塔筒叶片图像；所述检测诊断模块用于完成对于风电塔筒叶片的检测诊断；所述能源管理模块用于监控管理无人机飞行时的能源使用情况；本发明通过整合先进的机体控制、高效的图像采集处理和详尽的检测诊断方案，实现对风电塔筒叶片的全面监测和维护，显著提升了风电设备的运行安全性和维护效率。

The present invention provides a wind turbine tower blade detection system based on an unmanned aerial vehicle, the system comprising an airframe control module, an image acquisition module, a communication module, a detection and diagnosis module and an energy management module; the airframe control module is used for controlling the stable flight and precise navigation of the unmanned aerial vehicle; the communication module is used for completing data transmission between the unmanned aerial vehicle and a ground control station; the image acquisition module is used for collecting images of wind turbine tower blades during the flight of the unmanned aerial vehicle; the detection and diagnosis module is used for completing the detection and diagnosis of the wind turbine tower blades; the energy management module is used for monitoring and managing the energy usage of the unmanned aerial vehicle during flight; the present invention realizes comprehensive monitoring and maintenance of wind turbine tower blades by integrating advanced airframe control, efficient image acquisition and processing and detailed detection and diagnosis schemes, thereby significantly improving the operating safety and maintenance efficiency of wind power equipment.

Description

Wind power tower barrel blade detection system based on unmanned aerial vehicle

Technical Field

The invention relates to the field of fan blade detection systems, in particular to a wind power tower barrel blade detection system based on an unmanned aerial vehicle.

Background

With the rapid development of renewable energy industry, wind power is taken as an important component of renewable energy, maintenance and efficiency optimization of related equipment become particularly important, particularly, the wind power tower blades are taken as a key part of a wind power generator, the health state of the wind power tower blades directly influences the performance and safety of the whole power generation system, the wind power blades are often exposed to complex and variable external environments, so that the problems of abrasion, damage or aging and the like are easy to occur, timely and effective detection and maintenance are particularly critical, the traditional wind power blade detection method is often dependent on manual detection or ground infrastructure, the cost is high, the efficiency is low, and the requirements of the rapidly developed wind power industry are difficult to meet, so that searching for a wind power blade detection technology with higher efficiency and higher cost effectiveness becomes an important requirement of industry development.

Referring to related published technical schemes, the technology with the publication number of CN117386567A provides a fan blade detection method and system, the method comprises the steps of obtaining a real-time aerial view image of a flying robot system, extracting data from the real-time aerial view image through a central voting method to construct a blade simplified model of a fan blade to be detected, generating a safe flight reference track according to the blade simplified model and a safe flight offset distance, calculating an optimal placement position based on the safe flight reference track and the blade simplified model, controlling the flying robot system to place a crawling robot on the fan blade at a hovering position corresponding to the optimal placement position, controlling the crawling robot to detect the fan blade to obtain detection result information, and detecting the fan blade through the fan blade detection by combining pose information, the safe flight reference track and the optimal placement position.

Disclosure of Invention

The invention aims to provide a wind power tower barrel blade detection system based on an unmanned aerial vehicle aiming at the defects existing at present.

The invention adopts the following technical scheme:

A wind power tower barrel blade detection system based on an unmanned aerial vehicle comprises a machine body control module, an image acquisition module, a communication module, a detection diagnosis module and an energy management module;

The system comprises an unmanned aerial vehicle, a machine body control module, a communication module, an image acquisition module, a detection diagnosis module, an energy management module and a control module, wherein the machine body control module is used for controlling and realizing stable flight and accurate navigation of the unmanned aerial vehicle;

The machine body control module comprises a flight control unit and an automatic obstacle avoidance unit, wherein the flight control unit is used for controlling the unmanned aerial vehicle to fly along a set detection route, and the automatic obstacle avoidance unit is used for controlling the unmanned aerial vehicle to detect obstacles in a flight path in real time and adjust the flight path of the unmanned aerial vehicle to realize real-time obstacle avoidance in the flight process;

The detection and diagnosis module comprises an image preprocessing unit, a real-time detection unit and an early warning unit, wherein the image preprocessing unit is used for preprocessing and storing wind power tower blade images acquired in each flight task of the unmanned aerial vehicle, the real-time detection unit is used for detecting wind power tower blade conditions when the unmanned aerial vehicle executes the flight tasks, and the early warning unit is used for completing early warning of abnormal conditions of wind power tower blades in the future according to detection results of the real-time detection unit in a history record;

The image preprocessing unit further comprises an image acquisition subunit, an image splicing subunit and an image storage subunit, wherein the image acquisition subunit is used for acquiring wind power tower blade images continuously shot by the image acquisition module, and the image splicing subunit is used for correcting and splicing the continuously shot wind power tower blade images;

further, the workflow of the image stitching subunit is as follows:

s11, acquiring continuously shot wind power tower barrel blade images;

S12, correcting the brightness, contrast and color of all the blade images;

s13, identifying and extracting key feature points of all corrected blade images, and generating feature descriptors for the key feature points, wherein the feature descriptors are acquired based on gradient information of areas around the corresponding feature points;

S14, matching the extracted key feature points in each image according to the distance measurement between feature descriptors of the key feature points in each image;

s15, performing geometric transformation and alignment operation on each image according to the matched key feature points to splice, and performing fusion treatment on spliced areas of the images;

S16, generating and outputting a spliced blade image;

Furthermore, the real-time detection unit completes the detection of the blade condition by inputting the spliced blade image into a pre-trained blade defect recognition model, wherein the output of the blade defect recognition model comprises various defect types and defect severity degrees corresponding to the defect types;

Further, the real-time detection unit sets an alarm threshold for each defect type, and in the detection process of the blade condition, when the defect severity of a certain defect type exceeds the set alarm threshold, the real-time detection unit automatically triggers an alarm and sends alarm information to staff, wherein the alarm information comprises the defect type, the defect severity and the defect specific position acquired according to the blade image;

furthermore, the early warning unit analyzes the severity degree of the defects corresponding to each defect type in the comprehensive historical detection results to complete the prediction alarm of the possible defect type of the blade, and sends early warning information to staff when predicting that the future blade is likely to have abnormal situation defects.

The beneficial effects obtained by the invention are as follows:

The invention realizes a high-efficiency image processing flow through the combination of the image acquisition subunit, the image splicing subunit and the image storage subunit, generates and outputs complete blade images through integrating, correcting and splicing continuously shot partial blade images, provides a more comprehensive blade view for subsequent analysis and identification, can more accurately identify and analyze blade surface defects, identifies the blade defect condition in real time through the real-time detection unit, predicts the possible blade defects in the future through the early warning unit, and can take measures in advance to prevent the defects from further developing.

Drawings

The invention will be further understood from the following description taken in conjunction with the accompanying drawings. The components in the figures are not necessarily to scale, emphasis instead being placed upon illustrating the principles of the embodiments. Like reference numerals designate corresponding parts throughout the different views.

FIG. 1 is a schematic diagram of the overall module of the present invention.

FIG. 2 is a schematic workflow diagram of an image stitching subunit of the present invention.

Fig. 3 is a schematic flow chart of a fusion processing method for a splicing area according to the present invention.

Detailed Description

To make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below in connection with the embodiments thereof, it should be understood that the specific embodiments described herein are for explanation and not limitation of the present invention, other systems, methods and/or features of the present embodiments will become apparent to those skilled in the art after reviewing the following detailed description, it is intended that all such additional systems, methods, features and advantages be included in the present specification, be within the scope of the present invention, and be protected by the appended claims, additional features of the disclosed embodiments be described in the following detailed description, and these features will be apparent from the following detailed description.

In the description of the present invention, it should be understood that, if there is an orientation or positional relationship indicated by terms such as "upper", "lower", "left", "right", etc. based on the orientation or positional relationship shown in the drawings, it is merely for convenience of describing the present invention and simplifying the description, rather than indicating or suggesting that the apparatus or component referred to must have a specific orientation, construction and operation in the specific orientation, and thus the terms describing the positional relationship in the drawings are merely for exemplary illustration and are not to be construed as limitations of the present patent, and that the specific meaning of the terms may be understood by those skilled in the art according to circumstances.

Embodiment one.

As shown in fig. 1, the embodiment provides a wind power tower blade detection system based on an unmanned aerial vehicle, which comprises a machine body control module, an image acquisition module, a communication module, a detection diagnosis module and an energy management module;

The system comprises an unmanned aerial vehicle, a machine body control module, a communication module, an image acquisition module, a detection diagnosis module, an energy management module and a control module, wherein the machine body control module is used for controlling and realizing stable flight and accurate navigation of the unmanned aerial vehicle;

The machine body control module comprises a flight control unit and an automatic obstacle avoidance unit, wherein the flight control unit is used for controlling the unmanned aerial vehicle to fly along a set detection route, and the automatic obstacle avoidance unit is used for controlling the unmanned aerial vehicle to detect obstacles in a flight path in real time and adjust the flight path of the unmanned aerial vehicle to realize real-time obstacle avoidance in the flight process;

The detection and diagnosis module comprises an image preprocessing unit, a real-time detection unit and an early warning unit, wherein the image preprocessing unit is used for preprocessing and storing wind power tower blade images acquired in each flight task of the unmanned aerial vehicle, the real-time detection unit is used for detecting wind power tower blade conditions when the unmanned aerial vehicle executes the flight tasks, and the early warning unit is used for completing early warning of abnormal conditions of wind power tower blades in the future according to detection results of the real-time detection unit in a history record;

The image preprocessing unit further comprises an image acquisition subunit, an image splicing subunit and an image storage subunit, wherein the image acquisition subunit is used for acquiring wind power tower blade images continuously shot by the image acquisition module, and the image splicing subunit is used for correcting and splicing the continuously shot wind power tower blade images;

further, as shown in fig. 2, the workflow of the image stitching subunit is as follows:

s11, acquiring continuously shot wind power tower barrel blade images;

S12, correcting the brightness, contrast and color of all the blade images;

s13, identifying and extracting key feature points of all corrected blade images, and generating feature descriptors for the key feature points, wherein the feature descriptors are acquired based on gradient information of areas around the corresponding feature points;

S14, matching the extracted key feature points in each image according to the distance measurement between feature descriptors of the key feature points in each image;

s15, performing geometric transformation and alignment operation on each image according to the matched key feature points to splice, and performing fusion treatment on spliced areas of the images;

S16, generating and outputting a spliced blade image;

further, in the step S13, key feature points of each blade image are identified and extracted by using a SIFT algorithm;

further, as shown in fig. 3, in the step S15, the fusion process for the splicing region is completed by:

s151, decomposing two images to be spliced into a plurality of levels containing different detail features through Gaussian filtering and downsampling, wherein the uppermost level contains the most detail features and the lowermost level contains the least detail features;

S152, calculating and obtaining fusion weights on each level of the two spliced images according to the gradient amplitude of pixels of each level;

S153, effectively fusing corresponding levels of the two spliced images by using the fusion weight obtained in the previous step, and reconstructing each fused level layer by layer to generate a complete spliced blade image;

Specifically, in the step S151, gaussian filtering is first applied to the two images to smooth the images to reduce high frequency noise, and the main structure and low frequency information are retained; after Gaussian filtering is applied, downsampling is carried out on the two images so as to reduce the resolution ratio of the images, one hierarchy of the two images is obtained after the Gaussian filtering and downsampling operation is carried out once, the Gaussian filtering and downsampling operation is repeatedly carried out, so that a plurality of hierarchies of the two images containing different detail characteristics are obtained, and the detail characteristics of each hierarchy are image characteristics such as edges, textures, structures and the like contained in the hierarchy;

Further, in the step S152, the acquisition of the fusion weight for each layer is completed by:

S1521, for two spliced images, calculating gradient amplitude of pixels of each image on each level:

wherein F_1,i (x, y) is the gradient amplitude of the first spliced image at the ith level, x represents the transverse direction, y represents the longitudinal direction, I_1,x,i is the gradient of the first spliced image at the ith level in the transverse direction, I_1,y,i is the gradient of the first spliced image at the ith level in the longitudinal direction, F_2,i (x, y) is the gradient amplitude of the second spliced image at the ith level, I_2,x,i is the gradient of the second spliced image at the ith level in the transverse direction, I_2,y,i is the gradient of the second spliced image at the ith level in the longitudinal direction, F_1,,i(x,y)≥F_2,,i (x, y) is set;

S1522, calculating an initial weight of each image at each level:

Wherein, W_1,i (x, y) is the initial weight of the first spliced image at the ith level, and W_2,i (x, y) is the initial weight of the second spliced image at the ith level;

S1523, calculating fusion weight of each image on each level:

W′_1,i(x,y)＝W_1,i(x,y)+ΔW_i(x,y)×W_2,i(x,y);

W′_2,i(x,y)＝1-W′_1,i(x,y);

Wherein, W '_1,i (x, y) is the fusion weight of the first stitched image at the ith level, W'_2,i (x, y) is the fusion weight of the second stitched image at the ith level, Δw_i (x, y) is a weight adjustment factor, which satisfies the following conditions:

Wherein k is an adjustment coefficient, and the value range is [1,10];

further, in the step S53, the complete spliced blade image is generated by fusion reconstruction in the following manner:

S1531 for each level, generating a fused output for that level:

L_out,i＝W′_1,i(x,y)×L_1,i+W′_2,i(x,y)×L_2,i

wherein L_out,i is the fusion output of the ith level, L_1,i is the detail characteristic of the first image at the ith level, L_2,i is the detail characteristic of the second image at the ith level;

S1532, starting from the lowest level, reconstructing the images layer by using the L_out,i of each layer obtained by fusion until the images are reconstructed to the uppermost level, so as to generate spliced blade images;

The wind power tower has the advantages that the blade size of the wind power tower is huge, a single image is difficult to cover all areas, an effective splicing method is adopted to reconstruct a complete blade view from a plurality of images, the complete blade view is important for maintenance and safety inspection, the embodiment can effectively capture detailed information of the wind power blade on different levels through decomposing the images to different levels for image fusion, the detailed information from the whole outline to micro cracks or abrasion is included, dynamic weight adjustment is carried out on two spliced images on each level during fusion, more weights are set on the spliced images with more details according to gradient amplitude analysis calculation of the images on each level, the detail degree during image fusion is enhanced, the detail degree and definition of the spliced blade images are improved, and a more accurate visual basis is provided for subsequent blade condition analysis.

Embodiment two:

This embodiment should be understood to include at least all of the features of any one of the foregoing embodiments, and be further modified based thereon;

The embodiment provides a wind power tower barrel blade detection system based on an unmanned aerial vehicle, which comprises a machine body control module, an image acquisition module, a communication module, a detection diagnosis module and an energy management module;

The system comprises an unmanned aerial vehicle, a machine body control module, a communication module, an image acquisition module, a detection diagnosis module, an energy management module and a control module, wherein the machine body control module is used for controlling and realizing stable flight and accurate navigation of the unmanned aerial vehicle;

The machine body control module comprises a flight control unit and an automatic obstacle avoidance unit, wherein the flight control unit is used for controlling the unmanned aerial vehicle to fly along a set detection route, and the automatic obstacle avoidance unit is used for controlling the unmanned aerial vehicle to detect obstacles in a flight path in real time and adjust the flight path of the unmanned aerial vehicle to realize real-time obstacle avoidance in the flight process;

The detection and diagnosis module comprises an image preprocessing unit, a real-time detection unit and an early warning unit, wherein the image preprocessing unit is used for preprocessing and storing wind power tower blade images acquired in each flight task of the unmanned aerial vehicle, the real-time detection unit is used for detecting wind power tower blade conditions when the unmanned aerial vehicle executes the flight tasks, and the early warning unit is used for completing early warning of abnormal conditions of wind power tower blades in the future according to detection results of the real-time detection unit in a history record;

Furthermore, the real-time detection unit completes the detection of the blade condition by inputting the spliced blade image into a pre-trained blade defect recognition model, wherein the output of the blade defect recognition model comprises various defect types and defect severity degrees corresponding to the defect types;

Further, the real-time detection unit sets an alarm threshold for each defect type, and in the detection process of the blade condition, when the defect severity of a certain defect type exceeds the set alarm threshold, the real-time detection unit automatically triggers an alarm and sends alarm information to staff, wherein the alarm information comprises the defect type, the defect severity and the defect specific position acquired according to the blade image;

Furthermore, after the analysis of the detection results of the multiple real-time detection units in the history record is completed, the early warning unit sends early warning information to a user to finish early warning of abnormal conditions of the wind power tower blades in the future, wherein the early warning information comprises defect type repair early warning priority information of each wind power tower blade;

Further, the specific workflow of the early warning unit is as follows:

s21, acquiring n detection results closest to the current time for all detected wind power tower blades, wherein the detection results are various defect types output by a real-time detection unit and defect severity degrees corresponding to the defect types;

S22, for each defect type of each wind power tower blade, calculating a corresponding defect repair priority coefficient:

Wherein K is a defect repair priority coefficient of a certain defect type, Q_j is an alarm accumulation coefficient of a j-th detection result, W_avg is an average value of defect severity of all detection results which do not reach a set alarm threshold, max (E_j) is a maximum value of defect severity of all detection results which do not reach the set alarm threshold, and gamma is the set alarm threshold;

For Q_j:

S23, sequencing all defect types of each wind power tower blade according to the defect repair priority coefficient from high to low, wherein the repair early warning priority corresponding to the defect type with higher defect repair priority coefficient is higher;

The wind power tower barrel blade has higher operation and maintenance cost, is more critical to timely find and prevent blade defects in an early stage, analyzes the defect development trend of the blade according to the severity of various defect types by integrating multiple historical detection results, thereby providing early warning and repairing priorities for various defect types of each blade, and can better control the operation and maintenance cost, respond to potential risks in advance, improve the reliability and safety of equipment, finally reduce the repairing cost and prolong the service life of the equipment by integrating monitoring and early warning.

The foregoing disclosure is only a preferred embodiment of the present invention and is not intended to limit the scope of the invention, so that all equivalent technical changes made by applying the description of the present invention and the accompanying drawings are included in the scope of the present invention, and in addition, elements in the present invention can be updated as the technology develops.

Claims (6)

Translated fromChinese

1.一种基于无人机的风电塔筒叶片检测系统，其特征在于，所述系统包括机体控制模块、图像采集模块、通信模块、检测诊断模块和能源管理模块；所述机体控制模块用于控制实现无人机的稳定飞行和精确导航；所述通信模块用于完成无人机与地面控制站的数据传输；所述图像采集模块用于在无人机飞行过程中采集风电塔筒叶片图像；所述检测诊断模块用于完成对于风电塔筒叶片的检测诊断；所述能源管理模块用于监控管理无人机飞行时的能源使用情况；所述机体控制模块包括飞行控制单元和自动避障单元；所述飞行控制单元用于控制无人机沿设定的检测航线飞行，所述自动避障单元用于控制无人机在飞行过程中实时检测飞行路径中障碍物并调整无人机飞行路径实现实时避障；检测诊断模块包括图像预处理单元、实时检测单元和预警单元；所述图像预处理单元用于预处理并存储无人机每次飞行任务中采集的风电塔筒叶片图像；所述实时检测单元用于检测当次无人机执行飞行任务时的风电塔筒叶片情况；所述预警单元用于根据历史记录中多次实时检测单元的检测结果分析完成对于未来风电塔筒叶片异常情况的预警；1. A wind turbine tower blade detection system based on an unmanned aerial vehicle, characterized in that the system comprises an airframe control module, an image acquisition module, a communication module, a detection and diagnosis module and an energy management module; the airframe control module is used to control the stable flight and precise navigation of the unmanned aerial vehicle; the communication module is used to complete the data transmission between the unmanned aerial vehicle and the ground control station; the image acquisition module is used to collect wind turbine tower blade images during the flight of the unmanned aerial vehicle; the detection and diagnosis module is used to complete the detection and diagnosis of wind turbine tower blades; the energy management module is used to monitor and manage the energy usage of the unmanned aerial vehicle during flight; the airframe control module comprises a flight control unit and an automatic obstacle avoidance unit; The flight control unit is used to control the UAV to fly along the set detection route, and the automatic obstacle avoidance unit is used to control the UAV to detect obstacles in the flight path in real time during the flight and adjust the flight path of the UAV to achieve real-time obstacle avoidance; the detection and diagnosis module includes an image preprocessing unit, a real-time detection unit and an early warning unit; the image preprocessing unit is used to preprocess and store the wind turbine tower blade images collected in each flight mission of the UAV; the real-time detection unit is used to detect the condition of the wind turbine tower blades when the UAV performs the flight mission; the early warning unit is used to complete the early warning of abnormal conditions of the wind turbine tower blades in the future according to the analysis of the detection results of multiple real-time detection units in the historical records;图像预处理单元包括用于对连续拍摄的风电塔筒叶片图像进行校正和拼接处理的图像拼接子单元；所述图像拼接子单元的工作流程包括：The image preprocessing unit includes an image stitching subunit for correcting and stitching the continuously shot wind turbine tower blade images; the workflow of the image stitching subunit includes:S1521：对于两张拼接图像，计算每张图像在每一层级上像素的梯度幅度：S1521: For two stitched images, calculate the gradient amplitude of the pixels of each image at each level:；；其中，为第一张拼接图像在第层级的梯度幅度，表示横向方向，表示纵向方向，为第一张拼接图像在第层级的横向方向梯度，为第一张拼接图像在第层级的纵向方向梯度；为第二张拼接图像在第层级的梯度幅度，为第二张拼接图像在第层级的横向方向梯度，为第二张拼接图像在第层级的纵向方向梯度；设； ; ;in, The first stitched image is The gradient magnitude of the level, Indicates the horizontal direction, Indicates the vertical direction, The first stitched image is The lateral gradient of the level, The first stitched image is The longitudinal directional gradient of the hierarchy; For the second stitched image in The gradient magnitude of the level, For the second stitched image in The lateral gradient of the level, For the second stitched image in The vertical gradient of the level; ;S1522：计算每张图像在每一层级上的初始权重：S1522: Calculate the initial weight of each image at each level:；；其中，为第一张拼接图像在第层级的初始权重，为第二张拼接图像在第层级的初始权重； ; ;in, The first stitched image is The initial weights of the layers, For the second stitched image in The initial weight of the layer;S1523：计算每张图像在每一层级上的融合权重：S1523: Calculate the fusion weight of each image at each level:；； ; ;其中，为第一张拼接图像在第层级的融合权重，为第二张拼接图像在第层级的融合权重，为权重调整因子，满足：in, The first stitched image is The fusion weight of the layer, For the second stitched image in The fusion weight of the layer, is the weight adjustment factor, satisfying:；其中，为调整系数，取值范围。 ;in, is the adjustment coefficient, the value range is .2.根据权利要求1所述的一种基于无人机的风电塔筒叶片检测系统，其特征在于，所述图像预处理单元还包括图像获取子单元、图像拼接子单元和图像存储子单元；所述图像获取子单元用于获取图像采集模块连续拍摄的风电塔筒叶片图像，所述图像拼接子单元用于对连续拍摄的风电塔筒叶片图像进行校正和拼接处理；所述图像存储子单元用于完成对于拼接后叶片图像的存储。2. According to claim 1, a wind turbine tower blade detection system based on a drone is characterized in that the image preprocessing unit also includes an image acquisition subunit, an image stitching subunit and an image storage subunit; the image acquisition subunit is used to acquire the wind turbine tower blade images continuously taken by the image acquisition module, and the image stitching subunit is used to correct and stitch the continuously taken wind turbine tower blade images; the image storage subunit is used to complete the storage of the spliced blade images.3.根据权利要求1所述的一种基于无人机的风电塔筒叶片检测系统，其特征在于，所述图像拼接子单元的工作流程如下：3. The wind turbine tower blade detection system based on a drone according to claim 1, characterized in that the workflow of the image stitching subunit is as follows:S11：获取连续拍摄的风电塔筒叶片图像；S11: Acquire continuously photographed wind turbine tower blade images;S12：对所有叶片图像的亮度、对比度和色彩进行校正；S12: Correct the brightness, contrast and color of all leaf images;S13：对于所有校正后的叶片图像，识别并提取出各叶片图像的关键特征点，并为各个关键特征点生成特征描述符；所述特征描述符基于对应特征点周围区域的梯度信息获取；S13: for all corrected leaf images, identify and extract key feature points of each leaf image, and generate a feature descriptor for each key feature point; the feature descriptor is obtained based on the gradient information of the area around the corresponding feature point;S14：根据各个图像中关键特征点的特征描述符之间的距离度量来对各图像中提取的关键特征点进行匹配；S14: matching the key feature points extracted from each image according to the distance measurement between the feature descriptors of the key feature points in each image;S15：根据匹配的关键特征点对各个图像执行几何变换和对齐操作进行拼接，并对图像的拼接区域进行融合处理；S15: performing geometric transformation and alignment operations on each image to stitch them together according to the matched key feature points, and performing fusion processing on the stitching area of the image;S16：生成并输出拼接后的叶片图像。S16: Generate and output the stitched leaf image.4.根据权利要求1所述的一种基于无人机的风电塔筒叶片检测系统，其特征在于，所述实时检测单元通过将拼接后的叶片图像输入至预先训练好的叶片缺陷识别模型中从而完成对于叶片情况的检测；所述叶片缺陷识别模型的输出包括各种缺陷类型和缺陷类型对应的缺陷严重程度。4. According to claim 1, a wind turbine tower blade inspection system based on a drone is characterized in that the real-time inspection unit completes the inspection of the blade condition by inputting the spliced blade image into a pre-trained blade defect recognition model; the output of the blade defect recognition model includes various defect types and the defect severity corresponding to the defect type.5.根据权利要求1所述的一种基于无人机的风电塔筒叶片检测系统，其特征在于，所述实时检测单元为每种缺陷类型设定警报阈值，在对于叶片情况的检测过程中，当某种缺陷类型的缺陷严重程度超过设定的警报阈值时，实时检测单元将自动触发警报，并将警报信息发送至工作人员；所述警报信息包括缺陷类型、缺陷严重程度和根据叶片图像获取的缺陷具体位置。5. According to claim 1, a wind turbine tower blade inspection system based on a drone is characterized in that the real-time inspection unit sets an alarm threshold for each defect type. During the inspection of the blade condition, when the defect severity of a certain defect type exceeds the set alarm threshold, the real-time inspection unit will automatically trigger an alarm and send the alarm information to the staff; the alarm information includes the defect type, the defect severity and the specific location of the defect obtained based on the blade image.6.根据权利要求1所述的一种基于无人机的风电塔筒叶片检测系统，其特征在于，所述预警单元通过综合多次历史检测结果中各缺陷类型对应的缺陷严重程度分析完成对于叶片可能出现缺陷类型的预测警报，并在预测出未来叶片可能出现异常情况缺陷时，向工作人员发出预警信息。6. According to the UAV-based wind turbine tower blade inspection system of claim 1, the characteristic is that the early warning unit completes the predictive alarm for the possible defect types of the blades by comprehensively analyzing the defect severity corresponding to each defect type in multiple historical inspection results, and sends a warning message to the staff when it is predicted that abnormal defects may occur in the blades in the future.