skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Explore Research Products in the PAR
It may take a few hours for recently added research products to appear in PAR search results.


  • NSF Public Access
  • Search Results
  • Improving Road Safety during Nocturnal Hours by Characterizing Animal Poses Utilizing CNN-Based Analysis of Thermal Images
Title: Improving Road Safety during Nocturnal Hours by Characterizing Animal Poses Utilizing CNN-Based Analysis of Thermal Images
Animal–vehicle collision is a common danger on highways, especially during nighttime driving. Its likelihood is affected not only by the low visibility during nighttime hours, but also by the unpredictability of animals’ actions when a vehicle is nearby. Extensive research has shown that the lack of visibility during nighttime hours can be addressed using thermal imaging. However, to our knowledge, little research has been undertaken on predicting animal action through an animal’s specific poses while a vehicle is moving. This paper proposes a new system that couples the use of a two-dimensional convolutional neural network (2D-CNN) and thermal image input, to determine the risk imposed by an animal in a specific pose to a passing automobile during nighttime hours. The proposed system was tested using a set of thermal images presenting real-life scenarios of animals in specific poses on the roadside and was found to classify animal poses accurately in 82% of cases. Overall, it provides a valuable basis for implementing an automotive tool to minimize animal–vehicle collisions during nighttime hours.  more » « less
Award ID(s):
2051113
PAR ID:
10600725
Author(s) / Creator(s):
; ;
Publisher / Repository:
MDPI
Date Published:
Journal Name:
Sustainability
Volume:
14
Issue:
19
ISSN:
2071-1050
Page Range / eLocation ID:
12133
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; (, Intelligent Decision Technologies)
    Tsihrintzis, George A; Virvou, Maria; Hatzilygeroudis, Ioannis (Ed.)
    Animal Vehicle Collision (AVC) is relatively an evolving source of fatality resulting in the deficit of wildlife conservancy along with carnage. It’s a globally distressing and disturbing experience that causes monetary damage, injury, and human-animal mortality. Roadkill has always been atop the research domain and serendipitously provided heterogeneous solutions for collision mitigation and prevention. Despite the abundant solution availability, this research throws a new spotlight on wildlife-vehicle collision mitigation using highly efficient artificial intelligence during nighttime hours. This study focuses mainly on arousal mechanisms of the “Histogram of Oriented Gradients (HOG)” intelligent system with extracted thermography image features, which are then processed by a trained, convolutional neural network (1D-CNN). The above computer vision – deep learning-based alert system has an accuracy between 94%, and 96% on the arousal mechanisms with the empowered real-time data set utilization. 
    more » « less
  2. ; ; ; (, International Journal of Climatology)
    Abstract Adverse weather has been shown to be spatially and temporally variable across high‐latitude locations. The current study provides a unique investigation of limited surface visibility time periods at five coastal Greenland locations from 1979 to 2018 and identifies the coincident adverse weather, as well as the local and large‐scale atmospheric conditions during limited surface visibility time periods. Locations on the east coast of Greenland have the largest percentage of hours with limited visibility each month with maxima during February. Western and southern coastal locations have fewer occurrences of limited visibility. Warm‐season maxima are present in the northern locations, while warm‐season minima occur at all other locations. Fog is reported during each month at all five stations, however, a substantial increase of hours occurs at the northern and eastern stations during the typical melt season on Greenland. There is a seasonal difference in the percent of limited visibility hours linked to precipitation with a minimum in the warm season and maximum across the cold season. Limited visibility attributed to precipitation generally has the largest percentage of hours each month except for the northern and eastern locations during the melt season. The location and number of cyclonic circulations, as well as the spatial scale of troughs, across northeastern Canada and the North Atlantic Ocean greatly influence the adverse weather linked to prolonged limited surface visibility events. Limited visibility events at Thule, Greenland during both the warm and cold seasons are largely influenced by the presence and intensity of a cyclone over northeastern Canada. At Danmarkshavn, Greenland limited visibility events in both seasons generally have a cyclone positioned to the east, an area of higher pressure over Greenland and a cyclone positioned to the west of Greenland. 
    more » « less
  3. ; ; ; ; ; ; ; ; ; ; et al (, Nature Methods)
    Abstract The desire to understand how the brain generates and patterns behavior has driven rapid methodological innovation in tools to quantify natural animal behavior. While advances in deep learning and computer vision have enabled markerless pose estimation in individual animals, extending these to multiple animals presents unique challenges for studies of social behaviors or animals in their natural environments. Here we present Social LEAP Estimates Animal Poses (SLEAP), a machine learning system for multi-animal pose tracking. This system enables versatile workflows for data labeling, model training and inference on previously unseen data. SLEAP features an accessible graphical user interface, a standardized data model, a reproducible configuration system, over 30 model architectures, two approaches to part grouping and two approaches to identity tracking. We applied SLEAP to seven datasets across flies, bees, mice and gerbils to systematically evaluate each approach and architecture, and we compare it with other existing approaches. SLEAP achieves greater accuracy and speeds of more than 800 frames per second, with latencies of less than 3.5 ms at full 1,024 × 1,024 image resolution. This makes SLEAP usable for real-time applications, which we demonstrate by controlling the behavior of one animal on the basis of the tracking and detection of social interactions with another animal. 
    more » « less
  4. (, Atmosphere)
    The nighttime boundary layer was studied in an urban area surrounded by tropical forest by use of a copter-type unmanned aerial vehicle (UAV) in central Amazonia during the wet season. Fifty-seven vertical profiles of ozone concentration, potential temperature, and specific humidity were collected from surface to 500 m above ground level (a.g.l.) at high vertical and temporal resolutions by use of embedded sensors on the UAV. Abrupt changes in ozone concentration with altitude served as a proxy of nighttime boundary layer (NBL) height for the case of a normal, undisturbed, stratified nighttime atmosphere, corresponding to 40% of the cases. The median height of the boundary layer was 300 m. A turbulent mixing NBL constituted 28% of the profiles, while the median height of the boundary layer was 290 m. The remaining 32% of profiles corresponded to complex atmospheres without clear boundary layer heights. The occurrence of the three different cases correlated well with relative cloud cover. The results show that the standard nighttime model widely implemented in chemical transport models holds just 40% of the time, suggesting new challenges in modeling of regional nighttime chemistry. The boundary layer heights were also somewhat higher than observed previously over forested and pasture areas in Amazonia, indicating the important effect of the urban heat island. 
    more » « less
  5. (, mSystems)
    Recent research on host-microbe interactions has focused on intimate sym- bioses. Yet transient interactions, such as the stimulation of animal metamorphosis by bac- teria, can have significant impacts on each partner. During these short-lived interactions, swimming animal larvae identify a desirable location on the seafloor and undergo meta- morphosis into a juvenile based on the presence of specific bottom-dwelling bacteria. While this phenomenon is critical for seeding new animals to establish or maintain benthic ecosystems, there is an ocean of fundamental questions that remain unanswered. Here, I propose an updated model of how bacteria stimulate animal metamorphosis based on evi- dence that bacteria inject a stimulatory protein that prompts tubeworm metamorphosis. I consider what we hope to learn about stimulatory bacterial products, how animals recog- nize these products, and the consequences for both partners. Finally, I provide examples of how studying an enigmatic host-microbe interaction can serve as an engine for scientific discovery. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email