WO2020087831A1 - 基于光场拍摄技术的电力设备巡检图影像采集处理系统 - Google Patents

基于光场拍摄技术的电力设备巡检图影像采集处理系统 Download PDF

Info

Publication number
WO2020087831A1
WO2020087831A1 PCT/CN2019/077607 CN2019077607W WO2020087831A1 WO 2020087831 A1 WO2020087831 A1 WO 2020087831A1 CN 2019077607 W CN2019077607 W CN 2019077607W WO 2020087831 A1 WO2020087831 A1 WO 2020087831A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
light field
data
module
communication module
Prior art date
Application number
PCT/CN2019/077607
Other languages
English (en)
French (fr)
Inventor
李煜煌
何程
金伟
孙潍
Original Assignee
云南兆讯科技有限责任公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 云南兆讯科技有限责任公司 filed Critical 云南兆讯科技有限责任公司
Publication of WO2020087831A1 publication Critical patent/WO2020087831A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D3/00Control of position or direction
    • G05D3/12Control of position or direction using feedback

Definitions

  • the invention belongs to the technical field of power equipment inspection robots, and particularly relates to an image acquisition and processing system for power equipment inspection charts based on light field shooting technology.
  • Power equipment inspection robots have been widely used in various production environments (such as substations and cable trenches) in the power industry. They are mainly used to replace human power to collect data such as images, images, sounds, and temperatures of power equipment. Among them, image and image data are collected.
  • the existing technology mainly uses high-definition cameras for shooting.
  • the inspection robot needs to accurately move to the vicinity of the data collection device, adjust the camera position, height and angle, and use the device as the camera lens focus to adjust the camera optics Parameters to ensure the clarity of the captured device images and images. Every time you collect data from a device, you need to move, focus, and shoot once. Because the distance, angle, and focus of each device to the camera are different, you cannot take pictures and images of multiple devices at once. The data collection time is long and the efficiency is low.
  • the invention provides an image acquisition and processing system for electric equipment inspection charts based on light field shooting technology, to solve the problem that the electric equipment inspection robot can only shoot and acquire one device image and image data at a time.
  • the image acquisition and processing system of electric equipment inspection chart based on light field shooting technology including electric power inspection robot platform, light field shooting equipment and light field data processing system;
  • the electric power inspection robot platform includes an inspection robot, a motion system, a gimbal and a first communication module;
  • the inspection robot is connected to the motion system, the gimbal, the first communication module, and the light field shooting equipment;
  • the gimbal is also connected to the light field shooting equipment;
  • the motion system is used to control the walking route of the inspection robot
  • the gimbal is used to carry light field shooting equipment, and to adjust the height and angle of the light field shooting equipment;
  • the inspection robot is used to control the gimbal to adjust the height and angle of the light field shooting equipment, and to control the work of the light field shooting equipment, and also to transmit the images and image data shot by the light field shooting equipment to the light through the first communication module Field data processing system;
  • the first communication module is used to communicate with the light field data processing system and transmit the captured image and image data to the light field data processing system;
  • the walking route of the inspection robot pre-stored in the motion system
  • the height and angle of the light field shooting equipment at each stop on the walking path of the inspection robot are pre-stored in the inspection robot;
  • the light field data processing system includes a second communication module, a light field map image processing module, a map image recognition module, and a data storage module;
  • the first communication module, the second communication module, the light field image processing module, the image recognition module and the data storage module are sequentially connected in sequence;
  • the second communication module is used for data communication with the electric power inspection robot platform through the first communication module, so as to obtain the captured image and image data;
  • the light field image image processing module is used to process the captured image and image data transmitted from the first communication module to obtain the image and image data of each device;
  • the image recognition module is used to match the image and image data of each device from the light field image processing module with the device parameters pre-stored in the image recognition module to identify the device to which the image and image data of each device belongs Name and mark on the image and image data;
  • the data storage module is also connected to the second communication module and the light field image processing module respectively, and is used to store the captured images and image data transmitted by the second communication module, and store each device's data obtained by the light field image processing module
  • the image and image data also store the recognition and annotation results of the image recognition module.
  • it further includes a display module, which is connected to the data storage module and used to display the captured images and image data stored in the storage module, and is also used to display the image of each device stored in the storage module , Image data and recognition and annotation results.
  • a display module which is connected to the data storage module and used to display the captured images and image data stored in the storage module, and is also used to display the image of each device stored in the storage module , Image data and recognition and annotation results.
  • first communication module and the second communication module are connected by wire or wirelessly.
  • the light field shooting device includes a main lens and a micro lens array, and the light passing through the main lens passes through each micro lens of the micro lens array again, so as to obtain the direction and position information of the light.
  • the present invention also provides an image acquisition and processing method for electric equipment inspection maps based on light field shooting technology.
  • the above acquisition and processing system includes the following steps:
  • Step S1 According to the pre-stored walking route, move the inspection robot to the dock of the equipment to be inspected via the motion system;
  • Step S2 According to the height and angle of the light field shooting equipment at each stop on the walking path of the pre-stored inspection robot, the inspection robot controls the gimbal to adjust the height and angle of the light field shooting equipment, and then the light field shooting equipment shoots, Obtain the image and image data of the stop;
  • Step S3 The image and image data of the stop obtained in step S2 are transmitted to the light field image image processing module through the first communication module and the second communication module, and the light field image image processing module processes the image and image of the stop
  • the data obtains the complete outline boundary of each device and its position in the image and image data of the docking point, and records the correspondence between each device and position; then, with each device position as the focus, adjust the optical parameters, re-imaging, Improve the sharpness of the image and image of the device corresponding to the focus to a level that can be recognized, and intercept the image and image data of each device from the image and image data of the docking point;
  • Step S4 Recognize the image and image data of each device obtained in step S3 through the image and image recognition module, identify the device name to which the image and image data of each device belongs, and label the image and image data according to the recognition result Finally, the image and video data of the stop obtained in step S2, the image and video data of each device obtained in step S3, and the image and video data after identification and labeling are kept in the data storage module.
  • the recognizable level described in step S3 is a level recognizable by the naked eye or a machine algorithm.
  • the image and video data captured in the present invention are light field information data.
  • the present invention has no specific limitation on the position of the gimbal relative to the inspection robot, and can be installed on the top of the inspection robot or in front of the inspection robot, but it is not limited thereto.
  • the present invention has no specific limitation on the position of the light field shooting device relative to the gimbal, and may be on the top of the gimbal or on the upper side wall of the gimbal, but is not limited thereto.
  • connection relationship and structure of the gimbal and the light field shooting equipment all adopt the prior art.
  • the present invention has the following beneficial effects:
  • the traditional image and image acquisition technology can only shoot one device at a time, while the image acquisition and processing system and method of electric equipment inspection map based on light field photography technology provided by the present invention adopt light field photography technology, which can simultaneously shoot and collect multiple devices
  • the image and image data of power equipment greatly shorten the inspection time of power equipment robots and improve the efficiency of inspection; at the same time, because the position accuracy requirements of light field shooting technology are lower than traditional shooting technology, it can significantly reduce the power production environment such as substations.
  • the technical difficulty of rebuilding construction when the inspection robot is able to run reduces the cost of rebuilding.
  • FIG. 1 is a schematic structural diagram of an image acquisition and processing system of a patrol chart for electric equipment based on light field shooting technology in Embodiment 1;
  • FIG. 2 is a structural schematic diagram of an image acquisition and processing system of a patrol chart for electric equipment based on light field shooting technology in Embodiment 2;
  • FIG. 3 is a flow chart of the image acquisition and processing method of the electric equipment inspection chart based on the light field shooting technology of the present invention
  • FIG. 4 is a schematic diagram of the image acquisition and processing system of the electric equipment inspection chart based on the light field shooting technology of the present invention
  • FIG. 5 is a schematic diagram comparing the principles of the traditional shooting technology and the light field shooting technology of the present invention.
  • 100 electric inspection robot platform; 101, inspection robot; 102, motion system; 103, PTZ; 104, first communication module; 200, light field shooting equipment; 300, light field data processing system; 301, Second communication module; 302, light field image processing module; 303, image recognition module; 304, data storage module; 305, display module.
  • plural means two or more.
  • the terms “inner”, “upper”, “lower”, etc. indicate the orientation or state relationship is based on the orientation or state relationship shown in the drawings, only to facilitate the description of the present invention and simplify the description, not to indicate or imply The device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be construed as limiting the present invention.
  • connection should be understood in a broad sense, for example, it can be a fixed connection or a Disassembly connection, or integral connection; it can be mechanical connection or electrical connection; it can be directly connected or indirectly connected through an intermediary.
  • connection should be understood in a broad sense, for example, it can be a fixed connection or a Disassembly connection, or integral connection; it can be mechanical connection or electrical connection; it can be directly connected or indirectly connected through an intermediary.
  • the image acquisition and processing system of the electric equipment inspection chart based on the light field shooting technology includes the electric power inspection robot platform 100, the light field shooting equipment 200, and the light field data processing system 300;
  • the electric power inspection robot platform 100 includes an inspection robot 101, a motion system 102, a gimbal 103, and a first communication module 104;
  • the inspection robot 101 is connected to the motion system 102, the gimbal 103, the first communication module 104, and the light field shooting device 200, respectively;
  • the gimbal 103 is also connected to the light field shooting device 200;
  • the motion system 102 is used to control the walking route of the inspection robot 101;
  • the gimbal 103 is used to carry the light field shooting device 200, and is used to adjust the height and angle of the light field shooting device 200;
  • the inspection robot 101 is used to control the gimbal 103 to adjust the height and angle of the light field shooting device 200, and to control the work of the light field shooting device 200, and also to pass the images and image data shot by the light field shooting device 200 through the first
  • the communication module 104 transmits to the light field data processing system 300;
  • the first communication module 104 is used to communicate with the light field data processing system 300 and transmit the captured image and image data to the light field data processing system 300;
  • the walking route of the inspection robot 101 pre-stored in the motion system 102;
  • the inspection robot 101 has pre-stored the height and angle of the light field shooting device 200 at each stop on the walking route of the inspection robot 101;
  • the light field data processing system 300 includes a second communication module 301, a light field map image processing module 302, a map image recognition module 303, and a data storage module 304;
  • the first communication module 104, the second communication module 301, the light field image processing module 302, the image recognition module 303 and the data storage module 304 are sequentially connected in sequence;
  • the second communication module 301 is used for data communication with the electric power inspection robot platform 100 through the first communication module 104, so as to obtain the captured image and image data;
  • the light field image processing module 302 is used to process the captured image and video data transmitted from the first communication module 104 to obtain the image and video data of each device;
  • the image recognition module 303 is used to match the image and image data of each device from the light field image processing module 302 with the device parameters pre-stored in the image recognition module 303, so as to identify the image and image data of each device The name of the device to which it belongs and mark it on the image and image data;
  • the data storage module 304 is also connected to the second communication module 301 and the light field image processing module 302, respectively, for storing the captured image and image data transmitted from the second communication module 301, and storing the light field image processing module 302 to obtain
  • the image and video data of each device also stores the recognition and annotation results of the image recognition module 303.
  • An image acquisition and processing method for electric equipment patrol map based on light field shooting technology includes the following steps:
  • Step S1 According to the pre-stored walking route, move the inspection robot to the dock of the equipment to be inspected via the motion system;
  • Step S2 According to the height and angle of the light field shooting equipment at each stop on the walking path of the pre-stored inspection robot, the inspection robot controls the gimbal to adjust the height and angle of the light field shooting equipment, and then the light field shooting equipment shoots, Obtain the image and image data of the stop;
  • Step S3 The image and image data of the stop obtained in step S2 are transmitted to the light field image image processing module through the first communication module and the second communication module, and the light field image image processing module processes the image and image of the stop
  • the data obtains the complete outline boundary of each device and its position in the image and image data of the docking point, and records the correspondence between each device and position; then, with each device position as the focus, adjust the optical parameters, re-imaging, Improve the sharpness of the image and image of the device corresponding to the focus to a level that can be recognized, and intercept the image and image data of each device from the image and image data of the docking point;
  • Step S4 Recognize the image and image data of each device obtained in step S3 through the image and image recognition module, identify the device name to which the image and image data of each device belongs, and label the image and image data according to the recognition result Finally, the image and video data of the stop obtained in step S2, the image and video data of each device obtained in step S3, and the image and video data after identification and labeling are kept in the data storage module.
  • the image acquisition and processing system of the electric equipment inspection map based on the light field shooting technology includes the electric power inspection robot platform 100, the light field shooting equipment 200, and the light field data processing system 300;
  • the electric power inspection robot platform 100 includes an inspection robot 101, a motion system 102, a gimbal 103, and a first communication module 104;
  • the inspection robot 101 is connected to the motion system 102, the gimbal 103, the first communication module 104, and the light field shooting device 200, respectively;
  • the gimbal 103 is also connected to the light field shooting device 200;
  • the motion system 102 is used to control the walking route of the inspection robot 101;
  • the gimbal 103 is used to carry the light field shooting device 200, and is used to adjust the height and angle of the light field shooting device 200;
  • the inspection robot 101 is used to control the gimbal 103 to adjust the height and angle of the light field shooting device 200, and to control the work of the light field shooting device 200, and also to pass the images and image data shot by the light field shooting device 200 through the first
  • the communication module 104 transmits to the light field data processing system 300;
  • the first communication module 104 is used to communicate with the light field data processing system 300 and transmit the captured image and image data to the light field data processing system 300;
  • the walking route of the inspection robot 101 pre-stored in the motion system 102;
  • the inspection robot 101 has pre-stored the height and angle of the light field shooting device 200 at each stop on the walking route of the inspection robot 101;
  • the light field data processing system 300 includes a second communication module 301, a light field map image processing module 302, a map image recognition module 303, and a data storage module 304;
  • the first communication module 104, the second communication module 301, the light field image processing module 302, the image recognition module 303 and the data storage module 304 are sequentially connected in sequence;
  • the second communication module 301 is used for data communication with the electric power inspection robot platform 100 through the first communication module 104, so as to obtain the captured image and image data;
  • the light field image processing module 302 is used to process the captured image and video data transmitted from the first communication module 104 to obtain the image and video data of each device;
  • the image recognition module 303 is used to match the image and image data of each device from the light field image processing module 302 with the device parameters pre-stored in the image recognition module 303, so as to identify the image and image data of each device The name of the device to which it belongs and mark it on the image and image data;
  • the data storage module 304 is also connected to the second communication module 301 and the light field image processing module 302, respectively, for storing the captured image and image data transmitted from the second communication module 301, and storing the light field image processing module 302 to obtain
  • the image and video data of each device also stores the recognition and annotation results of the image recognition module 303.
  • a display module 305 which is connected to the data storage module 304, is used to display the captured images and image data stored in the storage module 304, and is also used to display the images of each device stored in the storage module 304, Image data and recognition and annotation results.
  • the first communication module 104 and the second communication module 301 are connected by wire or wirelessly, including but not limited to Ethernet and WiFi.
  • the light field shooting device 200 includes a main lens and a micro lens array, and the light passing through the main lens passes through each micro lens of the micro lens array again, thereby obtaining the direction and position information of the light, and making the imaging result more adjustable in the later period To achieve the effect of taking pictures first and then focusing.
  • An image acquisition and processing method for electric equipment patrol map based on light field shooting technology includes the following steps:
  • Step S1 According to the pre-stored walking route, move the inspection robot to the dock of the equipment to be inspected via the motion system;
  • Step S2 According to the height and angle of the light field shooting equipment at each stop on the walking path of the pre-stored inspection robot, the inspection robot controls the gimbal to adjust the height and angle of the light field shooting equipment, and then the light field shooting equipment shoots, Obtain the image and image data of the stop;
  • Step S3 The image and image data of the stop obtained in step S2 are transmitted to the light field image image processing module through the first communication module and the second communication module, and the light field image image processing module processes the image and image of the stop
  • the data obtains the complete outline boundary of each device and its position in the image and image data of the docking point, and records the correspondence between each device and position; then, with each device position as the focus, adjust the optical parameters, re-imaging, Improve the sharpness of the image and image of the device corresponding to the focus to a level that can be recognized, and intercept the image and image data of each device from the image and image data of the docking point;
  • Step S4 Recognize the image and image data of each device obtained in step S3 through the image and image recognition module, identify the device name to which the image and image data of each device belongs, and label the image and image data according to the recognition result Finally, the image and video data of the stop obtained in step S2, the image and video data of each device obtained in step S3, and the image and video data after identification and labeling are kept in the data storage module.
  • the recognizable level described in step S3 is a level recognizable by the naked eye or a machine algorithm.
  • the embodiment of the present invention uses light field shooting technology, which can obtain data of multiple devices in only one light field shooting (traditional image and image acquisition technology can only shoot one device at a time), and light field data after shooting
  • the processing system takes each device position as the focus, adjusts the optical parameters, re-images, and intercepts the data of different devices, so as to first capture and obtain all light field information, and then focus to intercept the specific device data, reduce the number of shots, and greatly shorten the power equipment robot patrol.
  • the inspection time improves the inspection efficiency; at the same time, because the light field shooting technology has lower position accuracy requirements than traditional shooting technology, it can significantly reduce the technical difficulty of retrofitting construction for substations and other power production environments for the operation of the inspection robot, and reduce the transformation. cost.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
  • Image Analysis (AREA)
  • Image Processing (AREA)
  • Manipulator (AREA)

Abstract

一种基于光场拍摄技术的电力设备巡检图影像采集处理系统,属于电力设备巡检机器人技术领域。该系统包括电力巡检机器人平台(100)、光场拍摄设备(200)和光场数据处理系统(300),电力巡检机器人平台(100)包括巡检机器人(101)、运动系统(102)、云台(103)和第一通信模块(104);光场数据处理系统(300)包括第二通信模块(301)、光场图影像处理模块(302)、图影像识别模块(303)和数据存储模块(304),系统采用光场拍摄技术,可以一次性同时拍摄采集多个电力设备的图像和影像数据,大大缩短电力设备机器人巡检的时间,提高巡检效率;同时显著降低变电站等电力生产环境为巡检机器人能够运行而进行改造施工的技术难度,降低改造成本。

Description

基于光场拍摄技术的电力设备巡检图影像采集处理系统 技术领域
本发明属于电力设备巡检机器人技术领域,具体涉及一种基于光场拍摄技术的电力设备巡检图影像采集处理系统。
背景技术
电力设备巡检机器人,已经在电力行业中各个生产环境(如变电站、电缆沟等)得到广泛应用,主要用于代替人力来采集电力设备的图像、影像、声音、温度等数据。其中采集图像和影像数据,现有技术主要使用高清摄像头进行拍摄,巡检机器人需要精确运动到采集数据设备的附近,调整摄像头位置、高度和角度,并以该设备为摄像头镜头焦点,调整摄像头光学参数,以保证拍摄的设备图像和影像的清晰。每采集一个设备的数据,就需要移动、调焦、拍摄一次,由于每个设备到摄像头的距离、角度、焦点不同,无法一次拍摄多个设备的图像和影像,数据采集时间长,效率低。
现有的高清摄像头以主镜头捕捉光线,再聚焦在镜头后的胶片或感光器上,所有光线的总和形成相片上的小点,显示影像,这种方式只能清晰呈现焦点的影像。因此如何克服现有技术的不足是目前电力设备巡检机器人技术领域亟需解决的问题。
发明内容
本发明提供一种基于光场拍摄技术的电力设备巡检图影像采集处理系统,以解决电力设备巡检机器人一次只能拍摄采集一个设备图像和影像数据的问题。
为实现上述目的,本发明采用的技术方案如下:
基于光场拍摄技术的电力设备巡检图影像采集处理系统,包括电力巡检机器人平台、光场拍摄设备和光场数据处理系统;
所述的电力巡检机器人平台包括巡检机器人、运动系统、云台和第一通信模块;
巡检机器人分别与运动系统、云台、第一通信模块、光场拍摄设备相连;
云台还与光场拍摄设备相连;
运动系统用于控制巡检机器人的行走路线;
云台用于承载光场拍摄设备,并用于调整光场拍摄设备的高度和角度;
巡检机器人用于控制云台调整光场拍摄设备的高度和角度,并用于控制光场拍摄设备的工作,还用于将光场拍摄设备拍摄的图像、影像数据通过第一通信模块传输至光场数据处理系统;
第一通信模块用于与光场数据处理系统进行通信,向光场数据处理系统传输拍摄的图像、影像数据;
运动系统内预存巡检机器人的行走路线;
巡检机器人内预存有巡检机器人行走路线上各停靠点的光场拍摄设备的高度和角度;
光场数据处理系统包括第二通信模块、光场图影像处理模块、图影像识别模块和数据存储模块;
第一通信模块、第二通信模块、光场图影像处理模块、图影像识别模块和数据存储模块依次顺序相连;
第二通信模块用于通过第一通信模块与电力巡检机器人平台进行数据通信,从而获取拍摄的图像、影像数据;
光场图影像处理模块用于将第一通信模块传来的拍摄的图像、影像数据进行处理,得到每个设备的图像、影像数据;
图影像识别模块用于对光场图影像处理模块传来的每个设备的图像、影像数据与图影像识别模块内预存的设备参数进行匹配,从而识别每个设备的图像、影像数据所属的设备名称并在图像、影像数据上进行标注;
数据存储模块还分别与第二通信模块、光场图影像处理模块相连,用于存储第二通信模块传来的拍摄的图像、影像数据,并存储光场图影像处理模块得到的每个设备的图像、影像数据,还存储图影像识别模块的识别及标注结果。
进一步,优选的是,还包括显示模块,所述的显示模块与数据存储模块相连,用于显示存储模块中存储的拍摄的图像、影像数据,还用于显示存储模块中存储每个设备的图像、影像数据及识别、标注结果。
进一步,优选的是,第一通信模块和第二通信模块通过有线或无线连接。
进一步,优选的是,光场拍摄设备包括主镜头和微型透镜阵列,将穿过主镜头的光线再次穿过微型透镜阵列的每个微型透镜,从而收获到光线的方向与位置信息。
本发明同时提供基于光场拍摄技术的电力设备巡检图影像采集处理方法,采用上述采集处理系统,包括如下步骤:
步骤S1:根据预存的行走路线,通过运动系统将巡检机器人移动到待检查设备停靠点停靠;
步骤S2:根据预存的巡检机器人行走路线上各停靠点的光场拍摄设备的高度和角度,巡检机器人控制云台来调整光场拍摄设备的高度和角度,之后光场拍摄设备进行拍摄,得到该停靠点的图像、影像数据;
步骤S3:将步骤步骤S2得到该停靠点的图像、影像数据通过第一通信模块、第二通信模块传输至光场图影像处理模块中,光场图影像处理模块处理该停靠点的图像、影像数据得到各个设备的完整轮廓边界及其在该停靠点的图像、影像数据中的位置,并记录每个设备和位置的对应关系;之后以每个设备位置为焦点,调整光学参数,重新成像,将焦点对应设备的图像、影像清晰度提升至能够识别的等级,并从该停靠点的图像、影像数据分别截取每个设备的图像、影像数据;
步骤S4:通过图像和影像识别模块对步骤S3得到的每个设备的图像、影像数据进行识别,识别每个设备的图像和影像数据所属的设备名称,并按照识别结果对图像、影像数据进行标注,最后将步骤S2得到该停靠点的图像、影像数据、步骤S3得到的每个设备的图像、影像数据,以及识别和标注后的图像、影像数据保持至数据存储模块中。
进一步,优选的是,步骤S3所述的能够识别的等级为肉眼或机器算法能够识别的等级。
本发明中拍摄的图像、影像数据为光场信息数据。
本发明对于云台相对于巡检机器人的位置没有具体限制,可以设置在巡检机器人的顶部,也可以设置在巡检机器人的前方,但不限于此。
本发明对于光场拍摄设备相对于云台的位置没有具体限制,可以在云台的顶部,也可以在云台的上部一侧壁上,但不限于此。
本发明中云台与光场拍摄设备的连接关系及构造均采用现有技术。
本发明与现有技术相比,其有益效果为:
传统图像和影像采集技术一次只能拍摄一个设备,而本发明提供的基于光场拍摄技术的电力设备巡检图影像采集处理系统和方法,采用光场拍摄技术,可以一次性同时拍摄采集多个电力设备的图像和影像数据,大大缩短电力设备机器人巡检的时间,提高巡检效率;同时,由于光场拍摄技术对位置精度要求低于传统拍摄技术,能够显著降低变电站等电力生产环境为巡检机器人能够运 行而进行改造施工的技术难度,降低改造成本。
附图说明
为了更清楚地说明本申请的技术方案,下面将对实施例中所需要使用的附图作简单地介绍。
图1为实施例1基于光场拍摄技术的电力设备巡检图影像采集处理系统的结构示意图;
图2为实施例2基于光场拍摄技术的电力设备巡检图影像采集处理系统的结构示意图;
图3为本发明基于光场拍摄技术的电力设备巡检图影像采集处理方法的流程图;
图4为本发明基于光场拍摄技术的电力设备巡检图影像采集处理系统使用时的示意图;
图5为传统拍摄技术与本发明光场拍摄技术原理对比示意图;
其中,100、电力巡检机器人平台;101、巡检机器人;102、运动系统;103、云台;104、第一通信模块;200、光场拍摄设备;300、光场数据处理系统;301、第二通信模块;302、光场图影像处理模块;303、图影像识别模块;304、数据存储模块;305、显示模块。
具体实施方式
下面结合实施例对本发明作进一步的详细描述。
本领域技术人员将会理解,下列实施例仅用于说明本发明,而不应视为限定本发明的范围。实施例中未注明具体技术或条件者,按照本领域内的文献所描述的技术或条件或者按照产品说明书进行。所用材料或设备未注明生产厂商者,均为可以通过购买获得的常规产品。
本技术领域技术人员可以理解,除非特意声明,这里使用的单数形式“一”、“一个”、“所述”和“该”也可包括复数形式。应该进一步理解的是,本发明的说明书中使用的措辞“包括”是指存在所述特征、整数、步骤、操作、元件和/或组件,但是并不排除存在或添加一个或多个其他特征、整数、步骤、操作、元件、组件和/或它们的组。应该理解,当我们称元件被“连接”或“耦接”到另一元件时,它可以直接连接或耦接到其他元件,或者也可以存在中间元件。此外,这里使用的“连接”或“耦接”可以包括无线连接或耦接。
在本发明的描述中,除非另有说明,“多个”的含义是两个或两个以上。术语 “内”、“上”、“下”等指示的方位或状态关系为基于附图所示的方位或状态关系,仅是为了便于描述本发明和简化描述,而不是指示或暗示所指的装置或元件必须具有特定的方位、以特定的方位构造和操作,因此不能理解为对本发明的限制。
在本发明的描述中,需要说明的是,除非另有明确的规定和限定,术语“安装”、“连接”、“设有”应做广义理解,例如,可以是固定连接,也可以是可拆卸连接,或一体地连接;可以是机械连接,也可以是电连接;可以是直接相连,也可以通过中间媒介间接相连。对于本领域的普通技术人员而言,根据具体情况理解上述术语在本发明中的具体含义。
本技术领域技术人员可以理解,除非另外定义,这里使用的所有术语(包括技术术语和科学术语)具有与本发明所属领域中的普通技术人员的一般理解相同的意义。还应该理解的是,诸如通用字典中定义的那些术语应该被理解为具有与现有技术的上下文中的意义一致的意义,并且除非像这里一样定义,不会用理想化或过于正式的含义来解释。
实施例1
如图1所示,基于光场拍摄技术的电力设备巡检图影像采集处理系统,包括电力巡检机器人平台100、光场拍摄设备200和光场数据处理系统300;
所述的电力巡检机器人平台100包括巡检机器人101、运动系统102、云台103和第一通信模块104;
巡检机器人101分别与运动系统102、云台103、第一通信模块104、光场拍摄设备200相连;
云台103还与光场拍摄设备200相连;
运动系统102用于控制巡检机器人101的行走路线;
云台103用于承载光场拍摄设备200,并用于调整光场拍摄设备200的高度和角度;
巡检机器人101用于控制云台103调整光场拍摄设备200的高度和角度,并用于控制光场拍摄设备200的工作,还用于将光场拍摄设备200拍摄的图像、影像数据通过第一通信模块104传输至光场数据处理系统300;
第一通信模块104用于与光场数据处理系统300进行通信,向光场数据处理系统300传输拍摄的图像、影像数据;
运动系统102内预存巡检机器人101的行走路线;
巡检机器人101内预存有巡检机器人101行走路线上各停靠点的光场拍摄 设备200的高度和角度;
光场数据处理系统300包括第二通信模块301、光场图影像处理模块302、图影像识别模块303和数据存储模块304;
第一通信模块104、第二通信模块301、光场图影像处理模块302、图影像识别模块303和数据存储模块304依次顺序相连;
第二通信模块301用于通过第一通信模块104与电力巡检机器人平台100进行数据通信,从而获取拍摄的图像、影像数据;
光场图影像处理模块302用于将第一通信模块104传来的拍摄的图像、影像数据进行处理,得到每个设备的图像、影像数据;
图影像识别模块303用于对光场图影像处理模块302传来的每个设备的图像、影像数据与图影像识别模块303内预存的设备参数进行匹配,从而识别每个设备的图像、影像数据所属的设备名称并在图像、影像数据上进行标注;
数据存储模块304还分别与第二通信模块301、光场图影像处理模块302相连,用于存储第二通信模块301传来的拍摄的图像、影像数据,并存储光场图影像处理模块302得到的每个设备的图像、影像数据,还存储图影像识别模块303的识别及标注结果。
一种基于光场拍摄技术的电力设备巡检图影像采集处理方法,采用本实施例采集处理系统,包括如下步骤:
步骤S1:根据预存的行走路线,通过运动系统将巡检机器人移动到待检查设备停靠点停靠;
步骤S2:根据预存的巡检机器人行走路线上各停靠点的光场拍摄设备的高度和角度,巡检机器人控制云台来调整光场拍摄设备的高度和角度,之后光场拍摄设备进行拍摄,得到该停靠点的图像、影像数据;
步骤S3:将步骤步骤S2得到该停靠点的图像、影像数据通过第一通信模块、第二通信模块传输至光场图影像处理模块中,光场图影像处理模块处理该停靠点的图像、影像数据得到各个设备的完整轮廓边界及其在该停靠点的图像、影像数据中的位置,并记录每个设备和位置的对应关系;之后以每个设备位置为焦点,调整光学参数,重新成像,将焦点对应设备的图像、影像清晰度提升至能够识别的等级,并从该停靠点的图像、影像数据分别截取每个设备的图像、影像数据;
步骤S4:通过图像和影像识别模块对步骤S3得到的每个设备的图像、影像 数据进行识别,识别每个设备的图像和影像数据所属的设备名称,并按照识别结果对图像、影像数据进行标注,最后将步骤S2得到该停靠点的图像、影像数据、步骤S3得到的每个设备的图像、影像数据,以及识别和标注后的图像、影像数据保持至数据存储模块中。
实施例2
如图2所示,基于光场拍摄技术的电力设备巡检图影像采集处理系统,包括电力巡检机器人平台100、光场拍摄设备200和光场数据处理系统300;
所述的电力巡检机器人平台100包括巡检机器人101、运动系统102、云台103和第一通信模块104;
巡检机器人101分别与运动系统102、云台103、第一通信模块104、光场拍摄设备200相连;
云台103还与光场拍摄设备200相连;
运动系统102用于控制巡检机器人101的行走路线;
云台103用于承载光场拍摄设备200,并用于调整光场拍摄设备200的高度和角度;
巡检机器人101用于控制云台103调整光场拍摄设备200的高度和角度,并用于控制光场拍摄设备200的工作,还用于将光场拍摄设备200拍摄的图像、影像数据通过第一通信模块104传输至光场数据处理系统300;
第一通信模块104用于与光场数据处理系统300进行通信,向光场数据处理系统300传输拍摄的图像、影像数据;
运动系统102内预存巡检机器人101的行走路线;
巡检机器人101内预存有巡检机器人101行走路线上各停靠点的光场拍摄设备200的高度和角度;
光场数据处理系统300包括第二通信模块301、光场图影像处理模块302、图影像识别模块303和数据存储模块304;
第一通信模块104、第二通信模块301、光场图影像处理模块302、图影像识别模块303和数据存储模块304依次顺序相连;
第二通信模块301用于通过第一通信模块104与电力巡检机器人平台100进行数据通信,从而获取拍摄的图像、影像数据;
光场图影像处理模块302用于将第一通信模块104传来的拍摄的图像、影像数据进行处理,得到每个设备的图像、影像数据;
图影像识别模块303用于对光场图影像处理模块302传来的每个设备的图像、影像数据与图影像识别模块303内预存的设备参数进行匹配,从而识别每个设备的图像、影像数据所属的设备名称并在图像、影像数据上进行标注;
数据存储模块304还分别与第二通信模块301、光场图影像处理模块302相连,用于存储第二通信模块301传来的拍摄的图像、影像数据,并存储光场图影像处理模块302得到的每个设备的图像、影像数据,还存储图影像识别模块303的识别及标注结果。
还包括显示模块305,所述的显示模块305与数据存储模块304相连,用于显示存储模块304中存储的拍摄的图像、影像数据,还用于显示存储模块304中存储每个设备的图像、影像数据及识别、标注结果。
第一通信模块104和第二通信模块301通过有线或无线连接,包括但不限于以太网、WiFi。
光场拍摄设备200包括主镜头和微型透镜阵列,将穿过主镜头的光线再次穿过微型透镜阵列的每个微型透镜,从而收获到光线的方向与位置信息,使成像结果在后期更加可调节,达到先拍照,后聚焦的效果。
一种基于光场拍摄技术的电力设备巡检图影像采集处理方法,采用本实施例采集处理系统,包括如下步骤:
步骤S1:根据预存的行走路线,通过运动系统将巡检机器人移动到待检查设备停靠点停靠;
步骤S2:根据预存的巡检机器人行走路线上各停靠点的光场拍摄设备的高度和角度,巡检机器人控制云台来调整光场拍摄设备的高度和角度,之后光场拍摄设备进行拍摄,得到该停靠点的图像、影像数据;
步骤S3:将步骤步骤S2得到该停靠点的图像、影像数据通过第一通信模块、第二通信模块传输至光场图影像处理模块中,光场图影像处理模块处理该停靠点的图像、影像数据得到各个设备的完整轮廓边界及其在该停靠点的图像、影像数据中的位置,并记录每个设备和位置的对应关系;之后以每个设备位置为焦点,调整光学参数,重新成像,将焦点对应设备的图像、影像清晰度提升至能够识别的等级,并从该停靠点的图像、影像数据分别截取每个设备的图像、影像数据;
步骤S4:通过图像和影像识别模块对步骤S3得到的每个设备的图像、影像数据进行识别,识别每个设备的图像和影像数据所属的设备名称,并按照识别 结果对图像、影像数据进行标注,最后将步骤S2得到该停靠点的图像、影像数据、步骤S3得到的每个设备的图像、影像数据,以及识别和标注后的图像、影像数据保持至数据存储模块中。
其中,步骤S3所述的能够识别的等级为肉眼或机器算法能够识别的等级。
由上述实施例可见,本发明实施例采用光场拍摄技术,只需要一次光场拍摄就能获取多个设备数据(传统图像、影像采集技术一次只能拍摄一个设备),拍摄完成后光场数据处理系统以每个设备位置为焦点,调整光学参数,重新成像,截取不同设备的数据,从而实现先拍摄获取所有光场信息,后对焦截取特定设备数据,减少拍摄次数,大大缩短电力设备机器人巡检的时间,提高巡检效率;同时,由于光场拍摄技术对位置精度要求低于传统拍摄技术,能够显著降低变电站等电力生产环境为巡检机器人能够运行而进行改造施工的技术难度,降低改造成本。
以上显示和描述了本发明的基本原理、主要特征和本发明的优点。本行业的技术人员应该了解,本发明不受上述实施例的限制,上述实施例和说明书中描述的只是说明本发明的原理,在不脱离本发明精神和范围的前提下,本发明还会有各种变化和改进,这些变化和改进都落入要求保护的本发明范围内。本发明要求保护范围由所附的权利要求书及其等效物界定。

Claims (6)

  1. 基于光场拍摄技术的电力设备巡检图影像采集处理系统,其特征在于,包括电力巡检机器人平台(100)、光场拍摄设备(200)和光场数据处理系统(300);
    所述的电力巡检机器人平台(100)包括巡检机器人(101)、运动系统(102)、云台(103)和第一通信模块(104);
    巡检机器人(101)分别与运动系统(102)、云台(103)、第一通信模块(104)、光场拍摄设备(200)相连;
    云台(103)还与光场拍摄设备(200)相连;
    运动系统(102)用于控制巡检机器人(101)的行走路线;
    云台(103)用于承载光场拍摄设备(200),并用于调整光场拍摄设备(200)的高度和角度;
    巡检机器人(101)用于控制云台(103)调整光场拍摄设备(200)的高度和角度,并用于控制光场拍摄设备(200)的工作,还用于将光场拍摄设备(200)拍摄的图像、影像数据通过第一通信模块(104)传输至光场数据处理系统(300);
    第一通信模块(104)用于与光场数据处理系统(300)进行通信,向光场数据处理系统(300)传输拍摄的图像、影像数据;
    运动系统(102)内预存巡检机器人(101)的行走路线;
    巡检机器人(101)内预存有巡检机器人(101)行走路线上各停靠点的光场拍摄设备(200)的高度和角度;
    光场数据处理系统(300)包括第二通信模块(301)、光场图影像处理模块(302)、图影像识别模块(303)和数据存储模块(304);
    第一通信模块(104)、第二通信模块(301)、光场图影像处理模块(302)、图影像识别模块(303)和数据存储模块(304)依次顺序相连;
    第二通信模块(301)用于通过第一通信模块(104)与电力巡检机器人平台(100)进行数据通信,从而获取拍摄的图像、影像数据;
    光场图影像处理模块(302)用于将第一通信模块(104)传来的拍摄的图像、影像数据进行处理,得到每个设备的图像、影像数据;
    图影像识别模块(303)用于对光场图影像处理模块(302)传来的每个设 备的图像、影像数据与图影像识别模块(303)内预存的设备参数进行匹配,从而识别每个设备的图像、影像数据所属的设备名称并在图像、影像数据上进行标注;
    数据存储模块(304)还分别与第二通信模块(301)、光场图影像处理模块(302)相连,用于存储第二通信模块(301)传来的拍摄的图像、影像数据,并存储光场图影像处理模块(302)得到的每个设备的图像、影像数据,还存储图影像识别模块(303)的识别及标注结果。
  2. 根据权利要求1所述的基于光场拍摄技术的电力设备巡检图影像采集处理系统,其特征在于,还包括显示模块(305),所述的显示模块(305)与数据存储模块(304)相连,用于显示存储模块(304)中存储的拍摄的图像、影像数据,还用于显示存储模块(304)中存储每个设备的图像、影像数据及识别、标注结果。
  3. 根据权利要求1所述的基于光场拍摄技术的电力设备巡检图影像采集处理系统,其特征在于,第一通信模块(104)和第二通信模块(301)通过有线或无线连接。
  4. 根据权利要求1所述的基于光场拍摄技术的电力设备巡检图影像采集处理系统,其特征在于,光场拍摄设备(200)包括主镜头和微型透镜阵列,将穿过主镜头的光线再次穿过微型透镜阵列的每个微型透镜,从而收获到光线的方向与位置信息。
  5. 基于光场拍摄技术的电力设备巡检图影像采集处理方法,采用权利要求1~4任意一项所述的基于光场拍摄技术的电力设备巡检图影像采集处理系统,其特征在于,包括如下步骤:
    步骤S1:根据预存的行走路线,通过运动系统将巡检机器人移动到待检查设备停靠点停靠;
    步骤S2:根据预存的巡检机器人行走路线上各停靠点的光场拍摄设备的高度和角度,巡检机器人控制云台来调整光场拍摄设备的高度和角度,之后光场拍摄设备进行拍摄,得到该停靠点的图像、影像数据;
    步骤S3:将步骤步骤S2得到该停靠点的图像、影像数据通过第一通信模块、第二通信模块传输至光场图影像处理模块中,光场图影像处理模块处理该停靠点的图像、影像数据得到各个设备的完整轮廓边界及其在该停靠点的图像、影 像数据中的位置,并记录每个设备和位置的对应关系;之后以每个设备位置为焦点,调整光学参数,重新成像,将焦点对应设备的图像、影像清晰度提升至能够识别的等级,并从该停靠点的图像、影像数据分别截取每个设备的图像、影像数据;
    步骤S4:通过图像和影像识别模块对步骤S3得到的每个设备的图像、影像数据进行识别,识别每个设备的图像和影像数据所属的设备名称,并按照识别结果对图像、影像数据进行标注,最后将步骤S2得到该停靠点的图像、影像数据、步骤S3得到的每个设备的图像、影像数据,以及识别和标注后的图像、影像数据保持至数据存储模块中。
  6. 根据权利要求5所述的基于光场拍摄技术的电力设备巡检图影像采集处理方法,其特征在于,步骤S3所述的能够识别的等级为肉眼或机器算法能够识别的等级。
PCT/CN2019/077607 2018-10-31 2019-03-11 基于光场拍摄技术的电力设备巡检图影像采集处理系统 WO2020087831A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811291050.4A CN109343573A (zh) 2018-10-31 2018-10-31 基于光场拍摄技术的电力设备巡检图影像采集处理系统
CN201811291050.4 2018-10-31

Publications (1)

Publication Number Publication Date
WO2020087831A1 true WO2020087831A1 (zh) 2020-05-07

Family

ID=65313020

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/077607 WO2020087831A1 (zh) 2018-10-31 2019-03-11 基于光场拍摄技术的电力设备巡检图影像采集处理系统

Country Status (2)

Country Link
CN (1) CN109343573A (zh)
WO (1) WO2020087831A1 (zh)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109343573A (zh) * 2018-10-31 2019-02-15 云南兆讯科技有限责任公司 基于光场拍摄技术的电力设备巡检图影像采集处理系统
CN110113592A (zh) * 2019-05-05 2019-08-09 云南兆讯科技有限责任公司 基于相机阵列技术的电力设备巡检图影像采集处理系统
CN110274160B (zh) * 2019-06-13 2021-05-11 咏峰(大连)科技有限公司 一种基于红外可见光融合图像的管道巡检系统
CN110443123A (zh) * 2019-06-26 2019-11-12 云南兆讯科技有限责任公司 一种电力设备巡检机器人图影像采集处理系统及方法
WO2021168707A1 (zh) * 2020-02-26 2021-09-02 深圳市大疆创新科技有限公司 对焦方法、装置及设备
CN114302031A (zh) * 2021-11-15 2022-04-08 华能国际电力股份有限公司上海石洞口第二电厂 基于5g通讯的相机阵列传输方法及系统

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102082466A (zh) * 2010-10-15 2011-06-01 重庆市电力公司超高压局 变电站设备智能巡检机器人系统
KR20140000379A (ko) * 2012-06-22 2014-01-03 전자부품연구원 카메라와 로봇 기반의 풍력 발전용 블레이드 검사 장치 및 방법
CN104391506A (zh) * 2014-09-19 2015-03-04 福州大学 变电站导轨式巡检机器人
CN108053471A (zh) * 2017-11-30 2018-05-18 南京绿新能源研究院有限公司 基于搭载光场相机的无人机光伏电站地形三维测绘建模系统
CN108494097A (zh) * 2018-05-09 2018-09-04 国网浙江省电力有限公司温州供电公司 用于变电站的机器人巡检系统
CN109343573A (zh) * 2018-10-31 2019-02-15 云南兆讯科技有限责任公司 基于光场拍摄技术的电力设备巡检图影像采集处理系统

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103417181B (zh) * 2013-08-01 2015-12-09 北京航空航天大学 一种光场视频摄像机内窥方法
CN106225931A (zh) * 2016-08-02 2016-12-14 北京殷图网联科技股份有限公司 基于红外点测温云台的变电站设备远程巡检方法及装置
CN106950469A (zh) * 2017-03-07 2017-07-14 北京瑞盈智拓科技发展有限公司 红外紫外一体化检测装置以及紫外成像装置

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102082466A (zh) * 2010-10-15 2011-06-01 重庆市电力公司超高压局 变电站设备智能巡检机器人系统
KR20140000379A (ko) * 2012-06-22 2014-01-03 전자부품연구원 카메라와 로봇 기반의 풍력 발전용 블레이드 검사 장치 및 방법
CN104391506A (zh) * 2014-09-19 2015-03-04 福州大学 变电站导轨式巡检机器人
CN108053471A (zh) * 2017-11-30 2018-05-18 南京绿新能源研究院有限公司 基于搭载光场相机的无人机光伏电站地形三维测绘建模系统
CN108494097A (zh) * 2018-05-09 2018-09-04 国网浙江省电力有限公司温州供电公司 用于变电站的机器人巡检系统
CN109343573A (zh) * 2018-10-31 2019-02-15 云南兆讯科技有限责任公司 基于光场拍摄技术的电力设备巡检图影像采集处理系统

Also Published As

Publication number Publication date
CN109343573A (zh) 2019-02-15

Similar Documents

Publication Publication Date Title
WO2020087831A1 (zh) 基于光场拍摄技术的电力设备巡检图影像采集处理系统
CN107659774B (zh) 一种基于多尺度相机阵列的视频成像系统及视频处理方法
CN112164015B (zh) 单目视觉自主巡检图像采集方法、装置及电力巡检无人机
TWI521255B (zh) 自動對焦方法、其自動對焦裝置和其影像擷取裝置
CN107471218B (zh) 一种基于多目视觉的双臂机器人手眼协调方法
WO2018103408A1 (zh) 一种无人机勘查交通事故现场的航拍图像采集方法及系统
CN106537214A (zh) 高速可变焦域透镜组件及相关方法
US11184539B2 (en) Intelligent dual-lens photographing device and photographing method therefor
JP2017108387A (ja) パノラマ魚眼カメラの画像較正、スティッチ、および深さ再構成方法、ならびにそのシステム
JP2017134830A (ja) 顔認識システム及び顔認識方法
CN107343174A (zh) 远距离移动目标防伪人脸虹膜抓拍装置及方法
CN112949478A (zh) 基于云台相机的目标检测方法
CN104079916A (zh) 一种全景三维视觉传感器及使用方法
CN206181190U (zh) 一种自动跟随的云台相机系统
CN101651784A (zh) 云台摄像机全景的视频跟踪系统
CN104394316B (zh) 摄像机云台姿态测试系统
WO2020207185A1 (zh) 基于三维光场技术的光学无人机监测系统
CN109525752A (zh) 拍摄设备及拍摄方法
KR101452342B1 (ko) 감시카메라 유닛 및 그 구동방법
CN106991701A (zh) 一种用于全景拍摄设备参数的标定系统及方法
TWI573104B (zh) 室內監控系統及其方法
KR20030064437A (ko) 파노라마 촬영용 감시 장치 및 그 방법
CN206195919U (zh) 一种室内虹膜自动抓拍装置
CN207027526U (zh) 一种基于双目视觉的具有视频分析功能的移动安防机器人
TW201410015A (zh) 多攝影機的整合處理系統及其方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19880592

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19880592

Country of ref document: EP

Kind code of ref document: A1