UAV inspection equipment and image fusion method

Abstract

The present invention relates to the technical field of unmanned aerial vehicle image processing, and provides an unmanned aerial vehicle inspection device, an image fusion method, and an unmanned aerial vehicle inspection device. The unmanned aerial vehicle inspection device includes: a dual-light temperature measurement camera, a pan-tilt, an unmanned aerial vehicle body, and an optical system, wherein the dual-light temperature measurement camera is fixedly connected to the pan-tilt; a pan-tilt interface is provided on the pan-tilt; the pan-tilt interface is mechanically mounted on the unmanned aerial vehicle body; the optical system is mounted on the dual-light temperature measurement camera and is electrically connected to the dual-light temperature measurement camera. The unmanned aerial vehicle inspection device provided by the present invention can realize accurate imaging processing of the inspection environment image by adopting a dual-light temperature measurement camera and an optical system, and the unmanned aerial vehicle inspection device can be installed on any type of unmanned aerial vehicle body by adopting a pan-tilt, thereby increasing the application scope of unmanned aerial vehicle inspection. In addition, by adopting a dual-light temperature measurement camera, multiple imaging processing fusions can be realized, thereby improving imaging accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle image processing, and in particular to an unmanned aerial vehicle inspection device and an image fusion method. Background Art

[0002] At present, in the field of data collection used in the field of drone inspection, infrared cameras and visible light cameras are generally used to obtain infrared images and visible light images respectively. Infrared cameras are generally designed to seal their optical systems, and the filters they are equipped with cannot meet the needs of full-scene use. For different detection targets, it is often necessary to replace the entire detector to achieve it, which greatly reduces the ease of use of infrared cameras.

[0003] In the visible light and infrared image fusion method used in this field, the data is generally collected using infrared lenses and visible light lenses to obtain infrared images and visible light images respectively. The image fusion algorithm used on drones is mainly based on the multi-scale decomposition algorithm. The features extracted by the multi-scale decomposition algorithm are manual features, while the features extracted by the deep learning algorithm are deep features. Deep features can better express image texture information. Therefore, the fusion effect is better than the multi-scale decomposition algorithm. However, the deep learning algorithm has high requirements for processor performance. Summary of the invention

[0004] In order to solve the above problems existing in the prior art, the present invention provides an unmanned aerial vehicle inspection device and an image fusion method.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A drone inspection device, comprising: a dual-light temperature measurement camera, a gimbal, a drone body and an optical system;

[0007] The dual-light temperature measurement camera is fixedly connected to the gimbal; a gimbal interface is provided on the gimbal; the gimbal interface is mechanically mounted on the drone body; the optical system is mounted on the dual-light temperature measurement camera and is electrically connected to the dual-light temperature measurement camera.

[0008] Preferably, the dual-light temperature measurement camera comprises: a housing, a data acquisition module, a standard processing module and an image fusion module;

[0009] The housing is fixedly connected to the pan / tilt; a hole for installing the optical system is provided on the housing; the data acquisition module is electrically connected to the optical system and the standard processing module respectively; the standard processing module is electrically connected to the image fusion module;

[0010] The data acquisition module is used to acquire optical data in the optical system; the optical data includes visible light image data and infrared image data;

[0011] The standard processing module is used to pre-process the optical data; the pre-processing includes: image cropping and image registration;

[0012] The image fusion module is used to fuse the pre-processed optical data using the CPCT fusion algorithm to obtain a fused image; the fused image is the inspection shot image.

[0013] Preferably, the housing comprises: the first sub-housing, the second sub-housing and the third sub-housing;

[0014] The first sub-shell is mounted on one end of the second sub-shell; the third sub-shell is mounted on the other end of the second sub-shell; the first sub-shell is fixedly connected to the pan / tilt; the first sub-shell, the second sub-shell and the third sub-shell form a storage space; the data acquisition module, the standard processing module and the image fusion module are all arranged in the storage space;

[0015] The third sub-shell is provided with a first hole, a second hole and a third hole.

[0016] Preferably, the optical system comprises: a visible light imager and an infrared imager;

[0017] The visible light imager and the infrared imager are both electrically connected to the data acquisition module.

[0018] Preferably, the visible light imager comprises: a visible light lens and a visible light imaging circuit;

[0019] The visible light lens is fixedly installed in the third hole; the visible light lens is electrically connected to the visible light imaging circuit; and the visible light imaging circuit is electrically connected to the data acquisition module.

[0020] Preferably, the infrared imager comprises an infrared lens, a chopper, an infrared detector and an infrared imaging circuit which are electrically connected in sequence;

[0021] The infrared lens is installed in the second hole; the infrared imaging circuit is electrically connected to the data acquisition module.

[0022] Preferably, the chopper is a plug-in chopper.

[0023] Preferably, the plug-in chopper comprises: a chopper seat and a filter;

[0024] The chopper seat is a clip-type structure for clamping the optical filter.

[0025] Preferably, it also includes an LED fill light;

[0026] The LED fill light is installed in the first hole.

[0027] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0028] The drone inspection device provided by the present invention can realize accurate imaging processing of the inspection environment image by using a dual-light temperature measurement camera and an optical system, and the drone inspection device can be installed on any type of drone body by using a pan / tilt platform, thereby increasing the application scope of drone inspection. In addition, by using a dual-light temperature measurement camera, multiple imaging processing fusion can be realized to improve imaging accuracy.

[0029] The present invention also provides an image fusion method to be applied to the above-mentioned unmanned aerial vehicle inspection device; the image fusion method comprises:

[0030] Acquiring optical data; the optical data includes: visible light image data and infrared image data;

[0031] Based on the resolution of the infrared image in the infrared image data, cropping the visible light image in the visible light image data;

[0032] Extracting a first feature point and a second feature point; the first feature point is a feature point of a visible light image in the visible light image data; the second feature point is a feature point of an infrared image in the infrared image data;

[0033] Using a feature point registration algorithm to obtain a first image coordinate transformation parameter according to the first feature point, and to obtain a second image coordinate transformation parameter according to the second feature point;

[0034] Using the first image coordinate transformation parameter to register the cropped visible light image to obtain a first registered image, and using the second image coordinate transformation parameter to register the infrared image to obtain a second registered image;

[0035] The first registered image and the second registered image are fused using a CPCT fusion algorithm based on a reference template to obtain an inspection image. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0037] Figure 1 A schematic diagram of the structure of the drone inspection equipment provided by the present invention;

[0038] Figure 2 A schematic diagram of the structure of a dual-light temperature measurement camera provided by an embodiment of the present invention;

[0039] Figure 3 A schematic diagram of the structure of a plug-in chopper provided in an embodiment of the present invention;

[0040] Figure 4 A flow chart of the image fusion method provided by the present invention;

[0041] Figure 5 A CPCT image fusion flow chart based on YUV space provided by an embodiment of the present invention;

[0042] Figure 6 A schematic diagram of a fused image provided by an embodiment of the present invention.

[0043] Explanation of symbols:

[0044] 1-pan head, 2-first sub-housing, 3-second sub-housing, 4-third sub-housing, 5-first hole position, 6-second hole position, 7-third hole position, 8-pan head interface position, 11-chopper seat, 12-optical filter. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] The purpose of the present invention is to provide an unmanned aerial vehicle inspection device and an image fusion method, so as to improve the accuracy of inspection image shooting and increase the wide application of unmanned aerial vehicles in the inspection field.

[0047] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0048] The unmanned aerial vehicle inspection equipment provided by the present invention comprises: a dual-light temperature measurement camera, a pan / tilt platform 1, an unmanned aerial vehicle body and an optical system.

[0049] The dual-light temperature measurement camera is fixedly connected to the gimbal 1. The gimbal 1 is provided with a gimbal interface position 8. The gimbal interface position 8 is mechanically mounted on the drone body. The optical system is mounted on the dual-light temperature measurement camera and is electrically connected to the dual-light temperature measurement camera.

[0050] Among them, in order to further improve the accuracy of inspection image imaging, such as Figure 1 and Figure 2As shown, the dual-light temperature measurement camera used above includes: a housing, a data acquisition module, a standard processing module and an image fusion module.

[0051] The housing is fixedly connected to the gimbal 1. The housing is provided with holes for installing the optical system. The data acquisition module is electrically connected to the optical system and the standard processing module respectively. The standard processing module is electrically connected to the image fusion module. The dual-light temperature measurement camera can also be installed at the drone load through the gimbal 1 connecting frame and the stable damping, and its specific installation position is set according to actual needs.

[0052] The data acquisition module is used to acquire optical data in the optical system. The optical data includes visible light image data and infrared image data. In actual use, the data acquisition module is used to acquire the original data set from the optical system, and the original data set includes the visible light image data set and the infrared image data set.

[0053] The standard processing module is used to preprocess the optical data. Preprocessing includes: image cropping and image registration. In the specific use process, the standard processing module performs image matching and image cropping on the visible light image dataset and the infrared image dataset to obtain a standard dataset after image registration and size standardization.

[0054] The image fusion module is used to fuse the pre-processed optical data using the CPCT fusion algorithm to obtain a fused image. The fused image is the inspection image. The image fusion module is designed based on the YUV color space. It uses an efficient color transfer method for visible light and infrared images to achieve end-to-end infrared conversion and fusion. It is convenient and fast. It adds visible light information to infrared images that lack texture to obtain a fused image set, which can enhance the observability of the image and facilitate operators to accumulate temperature data and grasp the temperature changes of the inspected equipment.

[0055] In order to improve the convenience of maintenance and installation, the housing used above includes: a first sub-housing 2 , a second sub-housing 3 and a third sub-housing 4 .

[0056] The first sub-shell 2 is mounted on one end of the second sub-shell 3. The third sub-shell 4 is mounted on the other end of the second sub-shell 3. The first sub-shell 2 is fixedly connected to the pan / tilt head 1. The first sub-shell 2, the second sub-shell 3 and the third sub-shell 4 form a receiving space. The data acquisition module, the standard processing module and the image fusion module are all arranged in the receiving space. The third sub-shell 4 is provided with a first hole 5, a second hole 6 and a third hole 7.

[0057] Furthermore, in order to improve the image quality of the inspection equipment, the optical system used above includes: a visible light imager and an infrared imager. For example, the infrared imager is a 640-resolution staring type uncooled infrared detector, and the visible light imager is a 1080 high-definition camera. Combining the advantages of infrared in finding heat sources and visible light in observing details, a high-resolution module is integrated to make the infrared image clearer and accurately locate the heat point, so that the device can also use the fusion function in dark environments such as in pipelines.

[0058] The visible light imager and the infrared imager are both electrically connected to the data acquisition module.

[0059] The visible light imager includes: a visible light lens and a visible light imaging circuit. The visible light lens is fixedly installed in the third hole 7. The visible light lens is electrically connected to the visible light imaging circuit. The visible light imaging circuit is electrically connected to the data acquisition module. In addition, other components of the infrared imager are all placed in the accommodation space to reduce the volume of the entire drone inspection equipment.

[0060] The infrared imager includes an infrared lens, a chopper, an infrared detector and an infrared imaging circuit which are electrically connected in sequence. The infrared radiation of the target object is focused by the lens and the band range is selected by the chopper before imaging on the focal plane of the detector.

[0061] The infrared lens is installed in the second hole 6. The infrared imaging circuit is electrically connected to the data acquisition module. In order to make the installation of the infrared imager easier, an infrared imager installation structure can also be set, and its main function is to install and fix the lens, color filter switching device, infrared detector, infrared imaging circuit, display, power supply and other components. The structure includes an internal support structure and a shell, and the internal support structure is mainly used to support and fix each module.

[0062] In the actual installation process, the visible light lens and the infrared lens are fixed in the horizontal and vertical directions, and the visible light and infrared lenses are installed vertically and coaxially side by side. After being fixed, there is no angular deviation in the XYZ axis, and there is no need for a large amount of manual adjustment to correct the angular deviation between the optical axes of the two cameras, thereby greatly reducing the combination cost of the dual-light fusion camera. This solves the technical problem that the dual-light fusion camera in the prior art is difficult to ensure that the center positions of the thermal infrared camera and the visible light camera are consistent during installation, so a large amount of manual adjustment is required to correct the angle and distance errors between the optical axes of the two cameras, resulting in high combination costs for the dual-light fusion camera.

[0063] In order to adapt to the selection of different bands, the chopper used in the present invention is preferably a plug-in chopper. Figure 3 As shown, the plug-in chopper includes a chopper seat 11 and a filter 12 .

[0064] The chopper seat 11 is a clip-type structure, and a through hole is left at the lower end for clamping the filter 12. When it is necessary to replace the filter 12 with a different absorption wavelength range, it is only necessary to open the chopper seat 11, which can significantly increase the use range of band imaging and simplify the operation steps. For example, a through hole with a diameter of 26mm is left at the lower part of the chopper seat 11, and the thickness of the filter that can be installed is less than 1mm and the diameter is 25.4 / 23.0mm.

[0065] In addition, the unmanned aerial vehicle inspection device provided by the present invention is also provided with an LED fill light, which is installed in the first hole 5 to provide light compensation when the ambient light is dark.

[0066] Based on this, the image fusion technology embedded in the dual-light temperature measurement camera adopts a technical solution that combines infrared thermal imaging, visible light camera and edge computing. The infrared imaging technology is combined with the fill light to enable the drone inspection equipment to use the fusion function even in dark environments.

[0067] The following takes the example of connecting and installing the UAV inspection equipment with DJI M300 and the same series of UAVs through the gimbal 1 as an example to explain in detail the technical solution provided by the present invention.

[0068] The dual-light temperature measurement camera is installed on the load directly below the drone through an external gimbal.

[0069] Furthermore, the dual-light temperature measurement camera mainly includes a camera housing, a visible light lens, an infrared lens, an LED fill light and a microprocessor. The camera housing adopts an integrated aluminum alloy housing for installing various components in the dual-light temperature measurement camera. The front of the camera front housing 4 is respectively provided with a visible light lens mounting position 7 and an infrared lens mounting position 6, and the visible light and infrared camera lenses are coaxially installed side by side. The side of the camera housing 3 is a pan / tilt interface position 8.

[0070] The LED fill light is installed on the front of the shell through the internal support structure of the shell. When the environment is dark, turn on the fill light to make the environment where the device is located brighter, thereby making the high-definition images taken by the camera clearer, further improving the accuracy of heat source judgment, and enabling the device to perform stable inspection work in dark environments such as inside pipelines.

[0071] The visible light lens and the infrared lens are fixed in the horizontal and vertical directions, and the visible light lens and the infrared lens are installed vertically and coaxially side by side. After fixing, there is no angular deviation in the XYZ axis. Furthermore, the infrared lens and the visible light lens are fixedly connected between the shell with air damping, and the damping isolation frequency is 70-200Hz. The front of the shell is the installation position of the dual-light camera lens, and the visible light and infrared camera lenses are installed coaxially side by side. The side of the shell is the position of the pan / tilt interface.

[0072] The LED fill light is installed on the front side of the shell through the internal support structure of the shell. Together with the LED fill light, the device can perform stable inspection work in dark environments such as inside pipelines.

[0073] The standard processing module is installed inside the shell through the internal support structure of the shell. The standard processing module is designed based on a deep learning network and uses the feature conversion method of visible light and infrared images. It is convenient and fast. At the same time, it uses deep learning and big data algorithms to add visible light information to infrared images that lack texture to obtain a fused image set, which can enhance the observability of the image and facilitate operating personnel to accumulate temperature data and grasp the temperature changes of the inspected equipment.

[0074] The infrared lens and chopper form an infrared imager. The infrared radiation of the inspection target object is focused by the lens and imaged after the chopper selects the wavelength range. The infrared lens is designed with germanium material and anti-reflection film. Its wavelength covers the range of 3μm to 5μm. It is a transmission optical system. The chopper component adopts a plug-in structure, and the corresponding filter can be replaced according to the different wavelength ranges of the gas to be measured.

[0075] Based on the above description, the drone inspection equipment provided by the present invention is based on infrared temperature measurement technology, combining the advantages of infrared to find heat sources and the advantages of visible light to observe details, and developing a dual-light fusion temperature measurement detector, integrating dual-lens high resolution, making the infrared image clearer, accurately locating the hot spot, and cooperating with the LED fill light, so that the drone can also output a stable fusion image when inspecting in dark environments such as pipelines. It is convenient for operators to accumulate temperature data, and prompt operators with specific image data and location status information to track fault points, confirm alarm conditions and troubleshoot, and can adapt to most drone inspection application scenarios.

[0076] The visible light and infrared image fusion technology adopts the constant parameter color transfer, namely CPCT (constant parameter color transfer) method, to analyze the mean and variance of the reference image, thereby deriving an image fusion method. This method is generally applicable to different scenes, and when the parameters change to a certain extent, it will not have a great impact on the naturalness of the fused image. In addition, it can also reduce the computational complexity of the algorithm and improve the real-time performance of the calculation. The hardware platform mounted on the drone can be used for real-time image fusion processing.

[0077] Based on this processing concept, the present invention also provides an image fusion method to be applied to the above-mentioned drone inspection equipment. Figure 4 As shown, the image fusion method includes:

[0078] Step 100: Obtain optical data. The optical data includes: visible light image data and infrared image data. In the specific application process, the visible light image and infrared image are collected by the visible light lens and infrared lens in the dual-light temperature measurement camera carried by the drone. The field of view of the visible light image is larger than that of the infrared image.

[0079] Step 101: based on the resolution of the infrared image in the infrared image data, crop the visible light image in the visible light image data.

[0080] Step 102: extracting a first feature point and a second feature point. The first feature point is a feature point of a visible light image in the visible light image data. The second feature point is a feature point of an infrared image in the infrared image data.

[0081] Step 103: using a feature point registration algorithm to obtain a first image coordinate transformation parameter according to the first feature point, and to obtain a second image coordinate transformation parameter according to the second feature point.

[0082] Step 104: The cropped visible light image is registered using the first image coordinate transformation parameters to obtain a first registered image, and the infrared image is registered using the second image coordinate transformation parameters to obtain a second registered image. For example, the resolution size of the acquired infrared image is used as a reference, and the visible light image is cropped according to the resolution size of the infrared image. Representative parts of the image are extracted as feature points, and then the matching feature point pairs are found by obtaining the similarity through a registration algorithm based on feature point matching, and then the image coordinate transformation parameters are obtained through the matching feature point pairs. Finally, the cropped visible light image and the infrared image are registered by the coordinate transformation parameters, so as to obtain the visible light image and the infrared image after pixel registration.

[0083] Step 105: Based on the reference template, the CPCT fusion algorithm is used to fuse the first registered image and the second registered image to obtain an inspection image.

[0084] The specific implementation process of step 105 is:

[0085] The CPCT transfer method is used for the first registered image and the second registered image, and the mean and variance of the reference image are analyzed.

[0086] Furthermore, the pre-processed first registered image and the second registered image are obtained, and the image to be processed S after preliminary fusion is obtained by the following formula.

[0087] P s (Y)=(P Vis (Y)+P IR (Y)) / 2,

[0088] P s(U)=P Vis (Y)-P IR (Y),

[0089] P s (V) = P IR (Y)-P Vis (Y),

[0090] Where P Vis is the visible light image, P IR For infrared images.

[0091] Furthermore, here we mainly focus on the YUV color space of the first registered image and the second registered image. The transformation between this space and the RGB color space is a linear transformation. In the color transformation process, the complexity of the calculation can be greatly reduced. In the YUV color space, brightness is independent of color difference, that is, scene details are independent of color. The application example process of the CPCT algorithm in the YUV color space is as follows: Figure 5 shown.

[0092] Among them, IR is the preprocessed second registration image, and Vis is the first registration image. The "Y" in YUV represents brightness, that is, the grayscale value. "U" and "V" represent the color and saturation of the image, which are used to specify the color of the pixel. The RGB color mode is a color standard in the industry. RGB, namely red, green and blue, is also known as the value of the three primary color light channels. A variety of colors are obtained by changing the three color channels of red (R), green (G), and blue (B) and superimposing them on each other.

[0093] Different from other color transfer image fusion algorithms, the CPCT fusion algorithm is based on the analysis of the color transfer process of the reference image.

[0094] The specific color transfer formula is shown below.

[0095]

[0096] Where P* is the YUV three channel value of the fused image. is the mean of the reference image, is the mean of the image S to be processed, P s are the three YUV channel values ​​of the image S to be processed, is the variance of the image S to be processed, is the variance of the reference image, that is, the subscript s is the image to be processed S, and the subscript t is the reference image.

[0097] Furthermore, changing μ Y , μ U , μ V , σ Y, σ U , σ V Set one or two of the parameters, set the others as constants, and then observe the impact of the parameters on the fusion results. This example analysis summarizes the value rules of the mean and variance of the three channels Y, U, and V as follows.

[0098] 1) When μ Y =80~110, the brightness of the fused image is better. Y When =800~1200, the boundary information of the fused image is better preserved.

[0099] 2) μ U , μ V Determines the color of the fused image. When |μ U |,|μ V |<6, the color of the fused image is not rich enough. U |,|μ V When |>20, the naturalness of the fused image is low. In order to ensure the naturalness of the image, ||μ U |-|μ V ||<8. As |μ U |,|μ V |Increase,||μ U |-|μ V ||Should be reduced to ensure the naturalness of the image.

[0100] 3)σ U , σ V = 600~1000, which can ensure the fusion of image color and naturalness. U >σ V When , the fused image is green, otherwise it is red. In order to ensure naturalness, |σ U -σ V |<400.

[0101] In summary, when μ Y =80~110, the fused image brightness is better, and σ Y When =800~1200, the boundary information of the fused image is better preserved.

[0102] Color transfer is actually to replace the mean and variance of the image to be processed S with the mean and variance of the reference image, so that the color change of the image to be processed is similar to that of the reference image, and the final fusion result is obtained and stored. Figure 6 shown.

[0103] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0104] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A drone inspection device, characterized in that: include: Dual-light temperature measurement camera, gimbal, drone body and optical system; The dual-light temperature measurement camera is fixedly connected to the pan-tilt platform; a pan-tilt platform interface is provided on the pan-tilt platform; The pan / tilt interface is mechanically mounted on the drone body; the optical system is mounted on the dual-light temperature measurement camera and is electrically connected to the dual-light temperature measurement camera; The dual-light temperature measurement camera includes: a housing, a data acquisition module, a standard processing module and an image fusion module; The housing is fixedly connected to the pan / tilt; a hole for installing the optical system is provided on the housing; the data acquisition module is electrically connected to the optical system and the standard processing module respectively; the standard processing module is electrically connected to the image fusion module; The data acquisition module is used to acquire optical data in the optical system; the optical data includes visible light image data and infrared image data; The standard processing module is used to preprocess the optical data; the preprocessing includes: image cropping and image registration; wherein the process of preprocessing the optical data in the standard processing module includes: cropping the visible light image in the visible light image data based on the resolution of the infrared image in the infrared image data; extracting a first feature point and a second feature point; the first feature point is a feature point of the visible light image in the visible light image data; the second feature point is a feature point of the infrared image in the infrared image data; using a feature point registration algorithm to obtain a first image coordinate transformation parameter according to the first feature point, and to obtain a second image coordinate transformation parameter according to the second feature point; using the first image coordinate transformation parameter to register the cropped visible light image to obtain a first registered image, and using the second image coordinate transformation parameter to register the infrared image to obtain a second registered image; The image fusion module is used to use the CPCT fusion algorithm to perform image fusion on the first registered image and the second registered image to obtain an inspection image.

2. The drone inspection device according to claim 1, characterized in that: The housing comprises: a first sub-housing, a second sub-housing and a third sub-housing; The first sub-shell is mounted on one end of the second sub-shell; the third sub-shell is mounted on the other end of the second sub-shell; the first sub-shell is fixedly connected to the pan / tilt; the first sub-shell, the second sub-shell and the third sub-shell form a storage space; the data acquisition module, the standard processing module and the image fusion module are all arranged in the storage space; The third sub-shell is provided with a first hole, a second hole and a third hole.

3. The unmanned aerial vehicle inspection equipment according to claim 2, characterized in that: The optical system includes: a visible light imager and an infrared imager; The visible light imager and the infrared imager are both electrically connected to the data acquisition module.

4. The drone inspection device according to claim 3, characterized in that: The visible light imager comprises: a visible light lens and a visible light imaging circuit; The visible light lens is fixedly installed in the third hole; the visible light lens is electrically connected to the visible light imaging circuit; and the visible light imaging circuit is electrically connected to the data acquisition module.

5. The unmanned aerial vehicle inspection equipment according to claim 3, characterized in that: The infrared imager comprises an infrared lens, a chopper, an infrared detector and an infrared imaging circuit which are electrically connected in sequence; The infrared lens is installed in the second hole; the infrared imaging circuit is electrically connected to the data acquisition module.

6. The unmanned aerial vehicle inspection equipment according to claim 5, characterized in that: The chopper is a plug-in chopper.

7. The unmanned aerial vehicle inspection equipment according to claim 6, characterized in that: The plug-in chopper comprises: a chopper seat and a filter; The chopper seat is a clip-type structure for clamping the optical filter.

8. The unmanned aerial vehicle inspection equipment according to claim 2, characterized in that: Also includes LED fill light; The LED fill light is installed in the first hole.

9. An image fusion method, characterized in that: Applicable to the drone inspection equipment as described in any one of claims 1 to 8; The image fusion method comprises: Acquiring optical data; the optical data includes: visible light image data and infrared image data; Based on the resolution of the infrared image in the infrared image data, cropping the visible light image in the visible light image data; Extracting a first feature point and a second feature point; the first feature point is a feature point of a visible light image in the visible light image data; the second feature point is a feature point of an infrared image in the infrared image data; Using a feature point registration algorithm to obtain a first image coordinate transformation parameter according to the first feature point, and to obtain a second image coordinate transformation parameter according to the second feature point; Using the first image coordinate transformation parameter to register the cropped visible light image to obtain a first registered image, and using the second image coordinate transformation parameter to register the infrared image to obtain a second registered image; The first registered image and the second registered image are fused using a CPCT fusion algorithm based on a reference template to obtain an inspection image.

Images