Automated detection and enumeration of marine wildlife using unmanned aircraft systems (UAS) and thermal imagery

Abstract

Estimating animal populations is critical for wildlife management. Aerial surveys are used for generating population estimates, but can be hampered by cost, logistical complexity, and human risk. Additionally, human counts of organisms in aerial imagery can be tedious and subjective. Automated approaches show promise, but can be constrained by long setup times and difficulty discriminating animals in aggregations. We combine unmanned aircraft systems (UAS), thermal imagery and computer vision to improve traditional wildlife survey methods. During spring 2015, we flew fixed-wing UAS equipped with thermal sensors, imaging two grey seal (Halichoerus grypus) breeding colonies in eastern Canada. Human analysts counted and classified individual seals in imagery manually. Concurrently, an automated classification and detection algorithm discriminated seals based upon temperature, size, and shape of thermal signatures. Automated counts were within 95–98% of human estimates; at Saddle Island, the model estimated 894 seals compared to analyst counts of 913, and at Hay Island estimated 2188 seals compared to analysts’ 2311. The algorithm improves upon shortcomings of computer vision by effectively recognizing seals in aggregations while keeping model setup time minimal. Our study illustrates how UAS, thermal imagery, and automated detection can be combined to efficiently collect population data critical to wildlife management.

The Use of Unmanned Aerial Systems in Marine Mammal Research

“…Turning to smaller RPA, LASE UAS are a less expensive tool suitable for close-range abundance surveys of marine mammal aggregation sites [11] (Table 1). LASE UAS deployment in offshore waters is also feasible, provided that vessels can be equipped with adequate retrieving systems [35].…”

“…The two census methods gave similar results (within 95% and 98% in the two locations), with most of the discrepancies being associated with the detection and discrimination of seal pups. The model has been proved to be an effective and less expensive alternative to human census, provided that the targeted animals thermally contrast with the landscape and are not associated with similar species [11].…”

“…Turning to lighter fixed-wing designs, the eBee (SenseFly) has been used for the census of grey seals (Halichoerus grypus) in two breeding colonies of the Gulf of St. Lawrence, Nova Scotia, Canada [11]. The aircraft was equipped with a Canon S110 and thermal infrared camera (640 × 512-pixel resolution), obtaining a GSD of 3 cm/pixel and 7 cm/pixel, respectively.…”

“…Instead, LASE UAS could be used to survey narrower areas [37], such as marine mammal aggregation sites (e.g., breeding or molting pinniped colonies, cetacean seasonal aggregations; e.g., [11,48]). Without the need to rapidly cover extensive areas, the aircraft can be flown at lower altitude, potentially outside the controlled airspace (<400 feet, 121.92 m AGL), thus obtaining the necessary flight authorization would be easier.…”

“…In particular, UAS have been proposed as a tool for marine mammal surveys, as they allow researchers to reach remote areas and observe animals from an advantageous perspective, being less invasive than standard aircraft [3][4][5][6][7][8][9]. Despite the challenges of operating at sea [10], UAS have been used for a number of marine mammal research applications, from abundance surveys [8,[11][12][13][14][15][16][17] to the measurement of…”