Sensors Fusion and Multidimensional Point Cloud Analysis for Electrical Power System Inspection

Abstract

Thermal inspection is a powerful tool that enables the diagnosis of several components at its early stages. One critical aspect that influences thermal inspection outputs is the infrared reflection from external sources. This situation may change the readings, demanding that an expert correctly define the camera position, which is a time consuming and expensive operation. To mitigate this problem, this work proposes an autonomous system capable of identifying infrared reflections by filtering and fusing data obtained from both stereo and thermal cameras. The process starts by acquiring readings from multiples Observation Points (OPs) where, at each OP, the system processes the 3D point cloud and thermal image by fusing them together. The result is a dense point cloud where each point has its spatial position and temperature. Considering that each point’s information is acquired from multiple poses, it is possible to generate a temperature profile of each spatial point and filter undesirable readings caused by interference and other phenomena. To deploy and test this approach, a Directional Robotic System (DRS) is mounted over a traditional human-operated service vehicle. In that way, the DRS autonomously tracks and inspects any desirable equipment as the service vehicle passes them by. To demonstrate the results, this work presents the algorithm workflow, a proof of concept, and a real application result, showing improved performance in real-life conditions.

Review: Research on product surface quality inspection technology based on 3D point cloud

“…84 Lo´pez-Ferna´ndez et al 84 used SFM to reconstruct the point cloud model of RGB photovoltaic (PV) panels, firstly normalized the sum of 8-bit RGB channel numbers of 2D images, and then achieved the thermal image mapping of 3D point clouds through the interactive recognition of homologous entities to obtain the 5D model (X, Y, Z, T, RGB-I), where (X, Y, Z) and RGB-I are used for the denoising and segmentation of PV panel point clouds, respectively, while the identification of panel defect regions is achieved by calculating the absolute difference between the temperature value of the current point and the median value of the total temperature concentration trend. Vidal et al 85 projected the temperature value from the thermal image to the point cloud data generated by binocular stereo vision through the calibration matrix of the camera, forming a spatial temperature distribution map with a dense lattice ( see Figure 6 ), which can clearly observe the quality mutation area.…”

“…In processing 3D point cloud data, some researchers have also followed the idea of image detection. In the detection of weld defects, in order to enhance the presentation effect of weld defects, Shao et al 97 used six depth images projected from different angles of the solder joint point cloud model as the input of the model, and used standard CNN to detect and classify solder joint defects, which made up for the 85 shortcomings of insufficient global feature capture ability of point cloud based on depth image. Eguchi et al 98 transformed the 3D point cloud data of the pavement into color information images, and used the AlexNet model to detect the spots on the asphalt surface.…”

An Edge-Fog Architecture for Distributed 3D Reconstruction and Remote Monitoring of a Power Plant Site in the Context of 5G

“…The 3D reconstruction is an essential and challenging analytical solution for large structures in this environment [11]. For a complete scene comprehension, several views from different perspectives are needed [12]. Other challenges also arise from the gathering and processing of information from various sensors.…”

“…The current literature in 3D reconstruction shows that most of the research focuses on optimizing the reconstruction quality in a centralized manner [11,12], new reconstruction approaches [13][14][15], and enhancing algorithms performance [16]. Only a few studies have focused on developing a scalable distributed system regarding remote 3D supervision [17][18][19].…”

Towards an Optimal Footprint Based Area Coverage Strategy for a False-Ceiling Inspection Robot

“…A vast volume of research works is reported on the development of strategies for automated inspection tasks for complex environments [ 18 , 19 , 20 , 21 ]. Robot-aided inspection is one of the key factors among automated inspection, which has been widely researched recently.…”