Tomas Naprstek

Dr. Tomas Naprstek

National Research Council Canada
orcid

Research Focus & Metrics

Academic Career Length

13Years (2014-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
15 January 2019 in Drones
Verified
Journal Article
Open access
PDF file
JAJ. Pablo Arroyo-Mora
J. Pablo Arroyo-Mora
Margaret KalacskaMargaret Kalacska
Margaret Kalacska
DIDeep Inamdar
Deep Inamdar
RSRaymond Soffer
Raymond Soffer
OLOliver Lucanus
Oliver Lucanus
George LeblancGeorge Leblanc
George Leblanc
Hyperspectral remote sensing provides a wealth of data essential for vegetation studies encompassing a wide range of applications (e.g., species diversity, ecosystem monitoring, etc.). The development and implementation of UAV-based hyperspectral systems have gained popularity over the last few years with novel efforts to demonstrate their operability. Here we describe the design, implementation, testing, and early results of the UAV-μCASI system, which showcases a relatively new hyperspectral sensor suitable for ecological studies. The μCASI (288 spectral bands) was integrated with a custom IMU-GNSS data recorder built in-house and mounted on a commercially available hexacopter platform with a gimbal to maximize system stability and minimize image distortion. We deployed the UAV-μCASI at three sites with different ecological characteristics across Canada: The Mer Bleue peatland, an abandoned agricultural field on Ile Grosbois, and the Cowichan Garry Oak Preserve meadow. We examined the attitude data from the flight controller to better understand airframe motion and the effectiveness of the integrated Differential Real Time Kinematic (RTK) GNSS. We describe important aspects of mission planning and show the effectiveness of a bundling adjustment to reduce boresight errors as well as the integration of a digital surface model for image geocorrection to account for parallax effects at the Mer Bleue test site. Finally, we assessed the quality of the radiometrically and atmospherically corrected imagery from the UAV-μCASI and found a close agreement (<2%) between the image derived reflectance and in-situ measurements. Overall, we found that a flight speed of 2.7 m/s, careful mission planning, and the integration of the bundling adjustment were important system characteristics for optimizing the image quality at an ultra-high spatial resolution (3–5 cm). Furthermore, environmental considerations such as wind speed (<5 m/s) and solar illumination also play a critical role in determining image quality. With the growing popularity of “turnkey” UAV-hyperspectral systems on the market, we demonstrate the basic requirements and technical challenges for these systems to be fully operational.

1 May 2019 in Geophysics
Verified
Journal Article
Open access
TNTomas Naprstek
Tomas Naprstek
RSRichard S. Smith
Richard S. Smith

22 September 2021 in Geophysics
Verified
Journal Article
Closed access
TNTomas Naprstek
Tomas Naprstek
RSRichard S. Smith
Richard S. Smith
Parameter estimation in aeromagnetics is an important tool for geologic interpretation. Due to aeromagnetic data being highly prevalent around the world, it can often be used to assist in understanding the geology of an area as a whole or for locating potential areas of further investigation for mineral exploration. Methods that automatically provide information such as the location and depth to the source of anomalies are useful to the interpretation, particularly in areas where a large number of anomalies exist. Unfortunately, many current methods rely on high-order derivatives and are therefore susceptible to noise in the data. Convolutional neural networks (CNNs) are a subset of machine-learning methods that are well-suited to image processing tasks, and they have been shown to be effective at interpreting other geophysical data, such as seismic sections. Following several similar successful approaches, we have developed a CNN methodology for estimating the location and depth of lineament-type anomalies in aeromagnetic maps. To train the CNN model, we used a synthetic aeromagnetic data modeler to vary the relevant physical parameters, and we developed a representative data set of approximately 1.4 million images. These were then used for training classification CNNs, with each class representing a small range of depth values. We first applied the model to a series of difficult synthetic data sets with varying amounts of noise, comparing the results against the tilt-depth method. We then applied the CNN model to a data set from northeastern Ontario, Canada, that contained a dike with known depth that was correctly estimated. This method is shown to be robust to noise, and it can easily be applied to new data sets using the trained model, which has been made publicly available.

27 April 2021 in Sensors
Verified
Journal Article
Open access
PDF file
GIGabriela Ifimov
Gabriela Ifimov
TNTomas Naprstek
Tomas Naprstek
JJJoshua Johnston
Joshua Johnston
JAJ. Pablo Arroyo-Mora
J. Pablo Arroyo-Mora
George LeblancGeorge Leblanc
George Leblanc
MLMadeline Lee
Madeline Lee
The increase in annual wildfires in many areas of the world has triggered international efforts to deploy sensors on airborne and space platforms to map these events and understand their behaviour. During the summer of 2017, an airborne flight campaign acquired mid-wave infrared imagery over active wildfires in Northern Ontario, Canada. However, it suffered multiple position-based equipment issues, thus requiring a non-standard geocorrection methodology. This study presents the approach, which utilizes a two-step semi-automatic geocorrection process that outputs image mosaics from airborne infrared video input. The first step extracts individual video frames that are combined into orthoimages using an automatic image registration method. The second step involves the georeferencing of the imagery using pseudo-ground control points to a fixed coordinate systems. The output geocorrected datasets in units of radiance can then be used to derive fire products such as fire radiative power density (FRPD). Prior to the georeferencing process, the Root Mean Square Error (RMSE) associated with the imagery was greater than 200 m. After the georeferencing process was applied, an RMSE below 30 m was reported, and the computed FRPD estimations are within expected values across the literature. As such, this alternative geocorrection methodology successfully salvages an otherwise unusable dataset and can be adapted by other researchers that do not have access to accurate positional information for airborne infrared flight campaigns over wildfires.

18 February 2016 in Geophysics
Verified
Journal Article
Open access
TNTomas Naprstek
Tomas Naprstek
RSRichard S. Smith
Richard S. Smith
The radio imaging method (RIM) is an electromagnetic cross-borehole method with applications in mineral exploration, as well as in the coal industry, where it is used across mine drives. Attenuation of the signal from conductive regions may indicate areas of mineralization, and these conductive effects in general dominate the response. In an effort to better understand the effect of a material’s dielectric permittivity on the response of the RIM, we have developed a simple program to model an electric dipole in a homogeneous whole space. When increasing the dielectric permittivity, the amplitude peak broadened and increased, whereas the phase peak sharpened and shifted negatively. To showcase the effect of dielectric permittivity on RIM data, data recorded from two transmitter positions in a moderately homogeneous zone in the Sudbury Basin were curve fit, and we concluded that despite the stronger effect that conductivity has on the signal, RIM is still sensitive to dielectric permittivity, and appropriate values must be used when developing conductivity tomograms. In addition, we found that for the given situation and frequencies used, an increase in either the conductivity or dielectric permittivity could be accounted for by a decrease of approximately the same factor in the other variable. However, the low-conductivity, high-permittivity case seemed to fit the shape of the amplitude and phase curves better. For the sulfide impregnated crystalline rocks at our field site, relative dielectric constants of 26.4 and 31 at 1250 and 625 kHz, respectively, were inferred.

27 July 2021 in MethodsX
Verified
Journal Article
Open access
TNTomas Naprstek
Tomas Naprstek
JAJ. Pablo Arroyo-Mora
J. Pablo Arroyo-Mora
JJJoshua Johnston
Joshua Johnston
GLGeorge Leblanc
George Leblanc
Airborne remotely sensed data (e.g. hyperspectral imagery, thermal videography, full frame RGB photography) often requires post-processing to be combined into a series of images or a mosaic for analysis. This is generally accomplished through the use of position and attitude hardware (i.e. Global Navigation Satellite System - GNSS / Inertial Measurement Unit - IMU) in combination with specialized software. Occasionally, hardware failure in the GNSS/IMU instrumentation occurs, however the data are still recoverable through a correction process, which allows image registration to mosaic the data. Here we present a simple and flexible MATLAB® code package that has been developed to combine video-based remotely sensed data. It first applies an iterative image registration process to align all frames, using pre-existing GPS information if supplied by the user, and then grids the frame data together to develop a final, single mosaic dataset that can be used for analysis. An example of this method using airborne infrared video data of a wildfire is shown as a demonstration. • MATLAB functions are easily adaptable to specific user needs and datasets. • The method outputs the combined data and positional information in three separate MATLAB variables that can be readily used for analysis in MATLAB or exported for use in other software.