Geology topics

Mark D. Koneff

Publications and source records attributed to Mark D. Koneff.

4 recordsLinked to original sources

USGS2024

Deep learning workflow to support in-flight processing of digital aerial imagery for wildlife population surveys

Deep learning shows promise for automating detection and classification of wildlife from digital aerial imagery to support cost-efficient remote sensing solutions for wildlife population monitoring. To support in-flight orthorectification and machine learning processing to detect and classify wildlife from imagery in near real-time, we evaluated deep learning methods that address hardware limitations and the need for processing efficiencies to support the envisioned in-flight workflow. We developed an annotated dataset for a suite of marine birds from high-resolution digital aerial imagery collected over open water environments to train the models. The proposed 3-stage workflow for automated, in-flight data processing includes: 1) image filtering based on the probability of any bird occurrence, 2) bird instance detection, and 3) bird instance classification. For image filtering, we compared the performance of a binary classifier with Mask Region-based Convolutional Neural Network (Mask R-CNN) as a means of sub-setting large volumes of imagery based on the probability of at least one bird occurrence in an image. On both the validation and test datasets, the binary classifier achieved higher performance than Mask R-CNN for predicting bird occurrence at the image-level. We recommend the binary classifier over Mask R-CNN for workflow first-stage filtering. For bird instance detection, we leveraged Mask R-CNN as our detection framework and proposed an iterative refinement method to bootstrap our predicted detections from loose ground-truth annotations. We also discuss future work to address the taxonomic classification phase of the envisioned workflow.

Massachusetts, Wisconsin↗

USGS2023

Challenges and solutions for automated avian recognition in aerial imagery

Remote aerial sensing provides a non-invasive, large geographical-scale technology for avian monitoring, but the manual processing of images limits its development and applications. Artificial Intelligence (AI) methods can be used to mitigate this manual image processing requirement. The implementation of AI methods, however, has several challenges: (1) imbalanced (i.e., long-tailed) data distribution, (2) annotation uncertainty in categorization, and (3) dataset discrepancies across different study sites. Here we use aerial imagery data of waterbirds around Cape Cod and Lake Michigan in the United States to examine how these challenges limit avian recognition performance. We review existing solutions and demonstrate as use cases how methods like Label Distribution Aware Marginal Loss with Deferred Re-Weighting, hierarchical classification, and FixMatch address the three challenges. We also present a new approach to tackle the annotation uncertainty challenge using a Soft-fine Pseudo-Label methodology. Finally, we aim with this paper to increase awareness in the ecological remote sensing community of these challenges and bridge the gap between ecological applications and state-of-the-art computer science, thereby opening new doors to future research.

Remote Sensing in Ecology and Conservation↗

USGS2022

Automating sandhill crane counts from nocturnal thermal aerial imagery using deep learning

Population monitoring is essential to management and conservation efforts for migratory birds, but traditional low-altitude aerial surveys with human observers are plagued by individual observer bias and risk to flight crews. Aerial surveys that use remote sensing can reduce bias and risk, but manual counting of wildlife in imagery is laborious and may be cost-prohibitive. Therefore, automated methods for counting are critical to cost-efficient application of remote sensing for wildlife surveys covering large areas. We conducted nocturnal surveys of sandhill cranes ( Antigone canadensis ) during spring migration in the Central Platte River Valley of Nebraska, USA, using midwave thermal infrared sensors. We developed a framework for automated counting of sandhill cranes from thermal imagery using deep learning, assessed and compared the performance of two automated counting models, and quantified the effect of spatial resolution on counting accuracy. Aerial thermal imagery data were collected in March 2018 and 2021; 40 images were analyzed. We applied two deep learning models: an object detection approach, Faster R-CNN and a recently developed pixel-density estimation approach, ASPDNet. Model performance was determined using data independent of the training imagery. The effect of spatial resolution was quantified with a beta regression on relative error. Our results showed model accuracy of 9% mean percent error for ASPDNet and 18% for Faster R-CNN. Most error was related to the undercounting of sandhill cranes. ASPDNet had < 50% of the error of Faster R-CNN as measured by mean percent error, root-mean-squared error and mean absolute error. Spatial resolution affected accuracy of both models, with error rate increasing with coarser resolution, particularly with Faster R-CNN. Deep learning models, particularly pixel-density estimators, can accurately automate counting of migratory birds in a dense, aggregate setting such as nocturnal roosting sites.

Nebraska↗

USGS2017

Evaluation of harvest and information needs for North American sea ducks

Wildlife managers routinely seek to establish sustainable limits of sport harvest or other regulated forms of take while confronted with considerable uncertainty. A growing body of ecological research focuses on methods to describe and account for uncertainty in management decision-making and to prioritize research and monitoring investments to reduce the most influential uncertainties. We used simulation methods incorporating measures of demographic uncertainty to evaluate risk of overharvest and prioritize information needs for North American sea ducks (Tribe Mergini ). Sea ducks are popular game birds in North America, yet they are poorly monitored and their population dynamics are poorly understood relative to other North American waterfowl. There have been few attempts to assess the sustainability of harvest of North American sea ducks, and no formal harvest strategy exists in the U.S. or Canada to guide management. The popularity of sea duck hunting, extended hunting opportunity for some populations (i.e., special seasons and/or bag limits), and population declines have led to concern about potential overharvest. We used Monte Carlo simulation to contrast estimates of allowable harvest and observed harvest and assess risk of overharvest for 7 populations of North American sea ducks: the American subspecies of common eider ( Somateria mollissima dresseri ), eastern and western populations of black scoter ( Melanitta americana ) and surf scoter ( M . perspicillata ), and continental populations of white-winged scoter ( M . fusca ) and long-tailed duck ( Clangula hyemalis ). We combined information from empirical studies and the opinions of experts through formal elicitation to create probability distributions reflecting uncertainty in the individual demographic parameters used in this assessment. Estimates of maximum growth ( r max ), and therefore of allowable harvest, were highly uncertain for all populations. Long-tailed duck and American common eider appeared to be at high risk of overharvest (i.e., observed harvest < allowable harvest in 5–7% and 19–26% of simulations, respectively depending on the functional form of density dependence), whereas the other populations appeared to be at moderate risk to low risk (observed harvest < allowable harvest in 22–68% of simulations, again conditional on the form of density dependence). We also evaluated the sensitivity of the difference between allowable and observed harvest estimates to uncertainty in individual demographic parameters to prioritize information needs. We found that uncertainty in overall fecundity had more influence on comparisons of allowable and observed harvest than adult survival or observed harvest for all species except long-tailed duck. Although adult survival was characterized by less uncertainty than individual components of fecundity, it was identified as a high priority information need given the sensitivity of growth rate and allowable harvest to this parameter. Uncertainty about population size was influential in the comparison of observed and allowable harvest for 5 of the 6 populations where it factored into the assessment. While this assessment highlights a high degree of uncertainty in allowable harvest, it provides a framework for integration of improved data from future research and monitoring. It could also serve as the basis for harvest strategy development as management objectives and regulatory alternatives are specified by the management community.

PLoS ONE↗