Region-Based CNN for Anomaly Detection in PV Power Plants Using Aerial Imagery.

Today, solar energy is taking an increasing share of the total energy mix. Unfortunately, many operational photovoltaic plants suffer from a plenitude of defects resulting in non-negligible power loss. The latter highly impacts the overall performance of the PV site; therefore, operators need to regularly inspect their solar parks for anomalies in order to prevent severe performance drops. As this operation is naturally labor-intensive and costly, we present in this paper a novel system for improved PV diagnostics using drone-based imagery. Our solution consists of three main steps. The first step locates the solar panels within the image. The second step detects the anomalies within the solar panels. The final step identifies the root cause of the anomaly. In this paper, we mainly focus on the second step comprising the detection of anomalies within solar panels, which is done using a region-based convolutional neural network (CNN). Experiments on six different PV sites with different specifications and a variety of defects demonstrate that our anomaly detector achieves a true positive rate or recall of more than 90% for a false positive rate of around 2% to 3% tested on a dataset containing nearly 9000 solar panels. Compared to the best state-of-the-art methods, the experiments revealed that we achieve a slightly higher true positive rate for a substantially lower false positive rate, while tested on a more realistic dataset.

Solar Panel Detection within Complex Backgrounds Using Thermal Images Acquired by UAVs.

The installation of solar plants everywhere in the world increases year by year. Automated diagnostic methods are needed to inspect the solar plants and to identify anomalies within these photovoltaic panels. The inspection is usually carried out by unmanned aerial vehicles (UAVs) using thermal imaging sensors. The first step in the whole process is to detect the solar panels in those images. However, standard image processing techniques fail in case of low-contrast images or images with complex backgrounds. Moreover, the shades of power lines or structures similar to solar panels impede the automated detection process. In this research, two self-developed methods are compared for the detection of panels in this context, one based on classical techniques and another one based on deep learning, both with a common post-processing step. The first method is based on edge detection and classification, in contrast to the second method is based on training a region based convolutional neural networks to identify a panel. The first method corrects for the low contrast of the thermal image using several preprocessing techniques. Subsequently, edge detection, segmentation and segment classification are applied. The latter is done using a support vector machine trained with an optimized texture descriptor vector. The second method is based on deep learning trained with images that have been subjected to three different pre-processing operations. The postprocessing use the detected panels to infer the location of panels that were not detected. This step selects contours from detected panels based on the panel area and the angle of rotation. Then new panels are determined by the extrapolation of these contours. The panels in 100 random images taken from eleven UAV flights over three solar plants are labeled and used to evaluate the detection methods. The metrics for the new method based on classical techniques reaches a precision of 0.997, a recall of 0.970 and a F1 score of 0.983. The metrics for the method of deep learning reaches a precision of 0.996, a recall of 0.981 and a F1 score of 0.989. The two panel detection methods are highly effective in the presence of complex backgrounds.

Have I Seen This Place Before? A Fast and Robust Loop Detection and Correction Method for 3D Lidar SLAM.

In this paper, we present a complete loop detection and correction system developed for data originating from lidar scanners. Regarding detection, we propose a combination of a global point cloud matcher with a novel registration algorithm to determine loop candidates in a highly effective way. The registration method can deal with point clouds that are largely deviating in orientation while improving the efficiency over existing techniques. In addition, we accelerated the computation of the global point cloud matcher by a factor of 2⁻4, exploiting the GPU to its maximum. Experiments demonstrated that our combined approach more reliably detects loops in lidar data compared to other point cloud matchers as it leads to better precision⁻recall trade-offs: for nearly 100% recall, we gain up to 7% in precision. Finally, we present a novel loop correction algorithm that leads to an improvement by a factor of 2 on the average and median pose error, while at the same time only requires a handful of seconds to complete.

3D Scene Reconstruction Using Omnidirectional Vision and LiDAR: A Hybrid Approach.

In this paper, we propose a novel approach to obtain accurate 3D reconstructions of large-scale environments by means of a mobile acquisition platform. The system incorporates a Velodyne LiDAR scanner, as well as a Point Grey Ladybug panoramic camera system. It was designed with genericity in mind, and hence, it does not make any assumption about the scene or about the sensor set-up. The main novelty of this work is that the proposed LiDAR mapping approach deals explicitly with the inhomogeneous density of point clouds produced by LiDAR scanners. To this end, we keep track of a global 3D map of the environment, which is continuously improved and refined by means of a surface reconstruction technique. Moreover, we perform surface analysis on consecutive generated point clouds in order to assure a perfect alignment with the global 3D map. In order to cope with drift, the system incorporates loop closure by determining the pose error and propagating it back in the pose graph. Our algorithm was exhaustively tested on data captured at a conference building, a university campus and an industrial site of a chemical company. Experiments demonstrate that it is capable of generating highly accurate 3D maps in very challenging environments. We can state that the average distance of corresponding point pairs between the ground truth and estimated point cloud approximates one centimeter for an area covering approximately 4000 m 2 . To prove the genericity of the system, it was tested on the well-known Kitti vision benchmark. The results show that our approach competes with state of the art methods without making any additional assumptions.