Thermal Imaging Techniques to Survey and Monitor Animals in the Wild by Kirk J Havens & Edward J. Sharp

Author:Kirk J Havens & Edward J. Sharp [Havens, Kirk J. & Sharp, Edward J.]
Language: eng
Format: epub
ISBN: 9780128033852
Publisher: Elsevier Science
Published: 2015-01-15T00:00:00+00:00

Figure 10.1 (A) This photograph of a thermal image depicts a cat lost in the background clutter formed by a strongly reradiating background that was previously heated from solar radiation during the afternoon hours. (B) This photograph is of the thermal image after the cat moved to the right into a region of the background that was not subjected to the solar loading of the afternoon sun.

The possibility of using thermal imaging to determine deer abundance in New Zealand forests was undertaken because the only methods available at the time were counting fecal pellet groups or using hunter kill returns, and both methods are imprecise and inaccurate (Wilde, 2000). A FLIR Systems Prism DS MWIR thermal infrared imager with an IFOV = 1 mrad, FOV (17° × 13°), and an NETD = 0.1°C was used to census penned fallow deer in a 15 ha paddock that contained both pasture and open indigenous forest. They aligned a set of flight lines for each of three flight altitudes (550, 1000, and 1650 ft. agl) to completely cover the paddock without overlap of transects. They videotaped the thermal imagery in early summer from three separate flights: at dawn, late morning, and dusk, using a helicopter fitted with a differential GPS. They determined that while the technique was good for detecting the presence of animals and for showing trends related to time of survey and flight altitudes, it would be unlikely to provide a reliable count index except under some well-defined situations because of the high level of variability between counts. They found that the best time to count deer using thermal imaging was shortly after dawn when both the ground and vegetation were still cold and there existed a significant temperature difference between the deer and their background. They also found that the best flying altitude to maximize deer sightings for their imager with an IFOV of 1 mrad was 550 ft. agl. They conclude by saying that if animal detection and counting with thermal imaging is to be successful within forest or scrub lands, the canopy must be sufficiently open for a good portion of the deer to be detected.

Surveying or censusing with an imager of this resolution is capable of detecting more deer than was realized in this effort. If the flight lines are aligned so that the footprint on the ground (swath width) is to one side of the aircraft at an angle slightly over 45° with respect to the nadir and the imager is panned slowly back and forth in a direction parallel to the flight line (dithered), the problem posed by tree canopy can be mitigated or perhaps negated.

Focardi et al. (2001) conducted a study in the field to compare animal detectability achieved with spotlighting and that achieved with thermal imaging as a function of animal size, species, detection range, season, group size, and composition. They used a cryogenically cooled thermal imager with a thermal resolution of 0.2°C equipped with two lenses having 5× and 30× magnification. The imager was mounted on a van and the imagery was recorded to VHS tape.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.