Fusing data from LiDAR and camera is conceptually attractive because of their complementary properties. For instance, camera images are of higher resolution and have colors, while LiDAR data provide more accurate range measurements and have a wider field of view. However, the sensor fusion problem remains challenging since it is difficult to find reliable correlations between data of very different characteristics (geometry versus texture, sparse versus dense). This letter proposes an offline LiDAR-camera fusion method to build dense, accurate 3-D models. Specifically, our method jointly solves a bundle adjustment problem and a cloud registration problem to compute camera poses and the sensor extrinsic calibration. In experiments, we show that our method can achieve an average accuracy of 2.7 mm and resolution of 70 points/cm ² by comparing to the ground truth data from a survey scanner. Furthermore, the extrinsic calibration result is discussed and shown to outperform the state-of-the-art method.

Accurate registration of 2D imagery with point clouds is a key technology for image-LiDAR point cloud fusion, camera to laser scanner calibration and camera localization. Despite continuous improvements, automatic registration of 2D and 3D data without using additional textured information still faces great challenges. In this paper, we propose a new 2D-3D registration method to estimate 2D-3D line feature correspondences and the camera pose in untextured point clouds of structured environments. Specifically, we first use geometric constraints between vanishing points and 3D parallel lines to compute all feasible camera rotations. Then, we utilize a hypothesis testing strategy to estimate the 2D-3D line correspondences and the translation vector. By checking the consistency with computed correspondences, the best rotation matrix can be found. Finally, the camera pose is further refined using non-linear optimization with all the 2D-3D line correspondences. The experimental results demonstrate the effectiveness of the proposed method on the synthetic and real dataset (outdoors and indoors) with repeated structures and rapid depth changes.

Detecting and finding people are complex tasks when visibility is reduced. This happens, for example, if a fire occurs. In these situations, heat sources and large amounts of smoke are generated. Under these circumstances, locating survivors using thermal or conventional cameras is not possible and it is necessary to use alternative techniques. The challenge of this work was to analyze if it is feasible the integration of an acoustic camera, developed at the University of Valladolid, on an unmanned aerial vehicle (UAV) to locate, by sound, people who are calling for help, in enclosed environments with reduced visibility. The acoustic array, based on MEMS (micro-electro-mechanical system) microphones, locates acoustic sources in space, and the UAV navigates autonomously by closed enclosures. This paper presents the first experimental results locating the angles of arrival of multiple sound sources, including the cries for help of a person, in an enclosed environment. The results are promising, as the system proves able to discriminate the noise generated by the propellers of the UAV, at the same time it identifies the angles of arrival of the direct sound signal and its first echoes reflected on the reflective surfaces.

Stereo reconstruction models trained on small images do not generalize well to high-resolution data. Training a model on high-resolution image size faces difficulties of data availability and is often infeasible due to limited computing resources. In this work, we present the Occlusion-aware Recurrent binocular Stereo matching (ORStereo), which deals with these issues by only training on available low disparity range stereo images. ORStereo generalizes to unseen high-resolution images with large disparity ranges by formulating the task as residual updates and refinements of an initial prediction. ORStereo is trained on images with disparity ranges limited to 256 pixels, yet it can operate 4K-resolution input with over 1000 disparities using limited GPU memory. We test the model’s capability on both synthetic and real-world high-resolution images. Experimental results demonstrate that ORStereo achieves comparable performance on 4K-resolution images compared to state-of-the-art methods trained on large disparity ranges. Compared to the baseline methods that are only trained on low-resolution images, our method has 60% or less error on 4K-resolution images.

This work presents dense stereo reconstruction using high-resolution images for infrastructure inspections. The state-of-the-art stereo reconstruction methods, both learning and non-learning ones, consume too much computational resource on high-resolution data. Recent learning-based methods achieve top ranks on most benchmarks. However, they suffer from the generalization issue due to lack of task-specific training data. We propose to use a less resource demanding non-learning method, guided by a learning-based model, to handle high-resolution images and achieve accurate stereo reconstruction. The deep-learning model produces an initial disparity prediction with uncertainty for each pixel of the down-sampled stereo image pair. The uncertainty serves as a self-measurement of its generalization ability and the per-pixel searching range around the initially predicted disparity. The downstream process performs a modified version of the Semi-Global Block Matching method with the up-sampled per-pixel searching range. The proposed deep-learning assisted method is evaluated on the Middlebury dataset and high-resolution stereo images collected by our customized binocular stereo camera. The combination of learning and non-learning methods achieves better performance on 12 out of 15 cases of the Middlebury dataset. In our infrastructure inspection experiments, the average 3D reconstruction error is less than 0.004m.