Search

Methodical aerial population surveys monitoring critically endangered species in Canadian North Atlantic waters are instrumental in influencing government policies both in eco-nomic and conservational efforts. The primary factor hindering the success of these missions is poor visibility caused by glare. This paper builds off our foundational paper [1] and pushes the envelope toward a data-driven glare modelling system. Said data-driven system makes use of meteorological and astronomical data to assist aircraft in navigating in order to mitigate acquisition errors and optimize the quality of acquired data. It is found that reliably extracting aircraft orientation is critical to our approach, to that end, we present a GPS alignment methodology which makes use of the fusion of two GPS signals. Using the complementary strengths and weaknesses of these two signals a synthetic interpolation of fused data is used to generate more reliable flight tracks, substantially improving glare modelling. This methodology could be applied to any other applications with similar signal restrictions., 2023 IEEE/ION Position, Location and Navigation Symposium (PLANS), April 24-27, 2023, Monterey, California, United States

Multimodal image registration is a challenging task. To begin with, the variation of parallax in the images makes the process intrinsically tricky. Additionally, due to phenomenology differences in modalities, the appearance of the same feature may vary significantly between the images making the registration laborious. To help mitigate these issues, we propose a two-step approach targeted at visible and infrared imagery. First, we train a generative adversarial network to learn the domain transfer function between the visible and the infrared domain, thereby mitigating the impact of the visual dissimilarity between the images. Second, we train a deep Siamese network to compute a homography in an unsupervised setting. Both elements are combined and trained sequentially. Our method is evaluated on a publicly available dataset. Our results show that the proposed method provides a reduction of more than 30% on average from the previous state-of-the-art, and outperforms several baselines and recent deep homography methods., 2022 IEEE International Conference on Big Data (Big Data), December 17-20, 2022, Osaka, Japan

Current airborne camera systems (ACS) used for surveillance and reconnaissance are generally equipped with cameras operating in different wavelengths. The registration and fusion of the images generated by these cameras enable a multimodal detection capability. In this paper, we propose a method to perform the geometric registration of infrared and visible imagery collected with an electro-optical and infrared (EO-IR) ACS. The registration process is a two-step approach. First, an initial transformation is computed using the expected orientation of the scene with respect to the camera and the camera’s intrinsic and extrinsic parameters. Then, the image alignment is refined using a technique based on mutual information which adapts the registration to the scene’s geometry and copes with video streams that are not perfectly synchronized. The image registration technique is implemented on an edge computing device and currently runs at 8 Hz. The intent is to use the resulting system to augment the capability of ACS that do not have built-in image registration and fusion capabilities. We present results obtained by applying the registration approach on real airborne imagery. We also evaluate the image registration error using physical markers installed on a building., 2022 IEEE International Conference on Image Processing (ICIP), October 16-19, 2022, Bordeaux, France

Fusing thermal and visible images is a recurring challenge in computer vision, especially when the images of the two modalities are not well registered. This registration problem is traditionally solved by matching descriptors and depends on the richness and discriminating power of the representation. Ensuring that detected features are dense and uniformly distributed is not necessarily guaranteed. More recently, machine learning methods addressed the issue of visible to visible matching, but few address the multi-modality setting. In this paper, we propose to address the special case of thermal-visible image registration with small baseline parallax correction. Our deep homography model is evaluated on an open thermal and visible dataset with two training settings, unsupervised and supervised. Results demonstrate the feasibility of the approach, and performances comparison to state-of-the-art models is evaluated., 2022 25th International Conference on Information Fusion (FUSION), Julu 4-7, 2022, Linköping, Sweden

Current mission planning simulations are typically carefully designed around a tradeoff between tight integration with accurate geospatial data or meeting real-time demands. Our proposed system is designed to achieve both. Deriving from work on cloud geospatial stores and mission planning systems, we are developing a platform for consuming new data, building it into models accessible from a real-time simulation server, and constructing machine learning systems over this data. This combination of large datasets with continuous pipelines feeding a real-time simulation forces us to bring together innovations on both the cloud infrastructure and user interface level., 2021 IEEE/ACM 25th International Symposium on Distributed Simulation and Real Time Applications (DS-RT), Sept. 27-29, 2021, Valencia, Spain [Virtual Host City]

The alignment of aerial and satellite imagery with ground sensor data is an ongoing research challenge. In dense urban environments, part of this challenge is induced by the positioning error of Global Positioning System (GPS). Despite the potential for error, many studies use GPS in order to infer road networks because GPS data is inexpensive and can be acquired quickly. Major transit organizations are freely providing data on the real-time position of their buses as well as ground truth route trajectories. This work exploits geospatial open data to construct a database of historical GPS from bus roads. Using this database, the GPS error map along main arteries of major cities can be reconstructed. The extraction of error maps is highly relevant for the planning and the joint exploitation of airborne and ground-based imagery. In this work, we use bus routes in downtown Victoria, BC, Canada and Adelaide, Australia to demonstrate the extraction GPS error maps., IEEE/CVF International Conference on Computer Vision (ICCV) Workshops 2021, Oct. 11-17, 2021 [Held Virtually]

This paper presents a classifier that takes airborne imagery acquired during marine megafauna surveys and classifies the glare intensity into four classes representing the severity of the glare. The objective of the classifier is to automate labour intensive and subjective components of the work performed by trained Marine Mammal Observers (MMOs). The proposed automatic method is based on a cascaded random forest architecture. The method uses features extracted from the histogram of the survey’s images and the metadata associated with respective images. The use of metadata is justified by the image formation model and we observed that it tends to improve the accuracy of the classifier. The proposed method provides results similar to that of trained MMOs. This is critical to the adoption of machine learning and machine vision technologies since introducing a change of methodology may impact the comparability of historic and future survey results when evaluating glare intensity., OCEANS 2021: San Diego – Porto, September 20-23, 2021,San Diego, CA, USA, Series: Ocean: Conference proceedings

The Jet Propulsion Laboratory is developing two experimental Differential Absorption Radar (DAR) platforms for remote sensing of water vapor on both Earth and planetary bodies. These are VIPR (Vapor Inside-cloud Profiling Radar), which operates near the 183 GHz water absorption resonance, and WASSR (WAter Sounding Short-range Radar), tuned to the 557 GHz water line. Here we contrast the system architecture for the two experimental instruments and present preliminary results of VIPR's recent airborne deployment.

The computational prediction of transcription factor binding sites remains a challenging problems in bioinformatics, despite significant m ethodological d evelopments f rom t he field of machine learning. Such computational models are essential to help interpret the non-coding portion of human genomes, and to learn more about the regulatory mechanisms controlling gene expression. In parallel, massive genome sequencing efforts have produced assembled genomes for hundred of vertebrate species, but this data is underused. We present PhyloReg, a new semi-supervised learning approach that can be used for a wide variety of sequence-to-function prediction problems, and that takes advantage of hundreds of millions of years of evolution to regularize predictors and improve accuracy. We demonstrate that PhyloReg can be used to better train a previously proposed deep learning model of transcription factor binding. Simulation studies further help delineate the benefits o f t he a pproach. G ains in prediction accuracy are obtained over a broad set of transcription factors and cell types.

5G standards implementation drive significant challenges at the acoustic filter level and requires innovative solutions. A promising approach relies in Surface Acoustic Waves technology combined with a thin LiTaO3 piezoelectric crystal layer bonded on Silicon substrate – so called POI-substrate, but suffers from volume manufacturing solution. We will report how Smart Cut™ layer transfer technology enables it.

In this study we report on the current state-of-progress of the Water Sounding Short-range Radar (WASSR) project. The proposed WASSR instrument will facilitate estimation of near-surface water vapor profile on Mars. This project is supported under NASA's Maturation of Instruments for Solar System Exploration (MatISSE) program and is a frequency modulated continuous wave (FMCW) differential absorption radar (DAR) instrument. The scope of the work presented here includes a discussion of (i) the science background and motivation for profiling water vapor near the Martian surface, (ii) the differential absorption radar (DAR) technique implemented at the 557 GHz water vapor absorption line, and (iii) the submillimeter wave radar system implementation and characterization.

High quality single crystal LiTaO3 thin film transfer using Smart Cut™ technology has been demonstrated. Piezo On Insulator (POI) characteristics were optimized to the benefit of the guided SAW resonator performance. We will correlate these substrate parameters with some key figures of merit of guided SAW devices in terms of frequency control, k2, Q factor.