The detection of oil spills in water is a frequently researched area, but most of the research has been based on very large patches of crude oil on offshore areas. We present a novel framework for detecting oil spills inside a port environment, while using unmanned areal vehicles (UAV) and a thermal infrared (IR) camera. This framework is split into a training part and an operational part. In the training part, we present a process for automatically annotating RGB images and matching them with the IR images in order to create a dataset. The infrared imaging camera is crucial to be able to detect oil spills during nighttime. This dataset is then used to train on a convolutional neural network (CNN). Seven different CNN segmentation architectures and eight different feature extractors are tested in order to find the best suited combination for this task. In the operational part, we propose a method to have a real-time, onboard UAV oil spill detection using the pre-trained network and a low power interference device. A controlled experiment in the port of Antwerp showed that we are able to achieve an accuracy of 89% while only using the IR camera.
Shape from focus is an accurate, but relatively time-consuming, 3D profilometry technique (compared to e.g., laser triangulation or fringe projection). This is the case because a large amount of data that needs to be captured and processed to obtain 3D measurements. In this paper, we propose a two-step shape-from-focus measurement approach that can improve the speed with 40%. By using a faster profilometry technique to create a coarse measurement of an unknown target, this coarse measurement can be used to limit the data capture to only the required frames. This method can significantly improve the measurement and processing speed. The method was tested on a 40 mm by 40 mm custom target and resulted in an overall 46% reduction of measurement time. The accuracy of the proposed method was compared against the conventional shape from focus method by comparing both methods with a more accurate reference.
The focus of this study is to design a backlit vision instrument capable of measuring surface roughness and to discuss its metrological performance compared to traditional measurement instruments. The instrument is a non-contact high-magnification imaging system characterized by short inspection time which opens the perspective of in-line implementation. We combined the use of the modulation transfer function to evaluate the imaging conditions of an electrically tunable lens to obtain an optimally focused image. We prepared a set of turned steel samples with different roughness in the range Ra 2.4 µm to 15.1 µm. The layout of the instrument is presented, including a discussion on how optimal imaging conditions were obtained. The paper describes the comparison performed on measurements collected with the vision system designed in this work and state-of-the-art instruments. A comparison of the results of the backlit system depends on the values of surface roughness considered; while at larger values of roughness the offset increases, the results are compatible with the ones of the stylus at lower values of roughness. In fact, the error bands are superimposed by at least 58% based on the cases analyzed.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations –citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.