A comparison of deep learning and citizen science techniques for counting wildlife in aerial survey images. Issue 6 (5th March 2019)

• Record Type: Journal Article
• Authors: Torney, Colin J.; Lloyd‐Jones, David J.; Chevallier, Mark; Moyer, David C.; Maliti, Honori T.; Mwita, Machoke; Kohi, Edward M.; Hopcraft, Grant C.
• Editors: McCrea, Rachel
• Is Part Of: Methods in ecology and evolution. Volume 10:Issue 6(2019)
• Page Start: 779
• Page End: 787
• Publication Date: 2019-03-05
• Subjects: citizen science -- conservation -- deep learning -- monitoring -- population ecology -- surveys; Ecology -- Periodicals; Evolution -- Periodicals; 577
• DOI: 10.1111/2041-210X.13165
• Languages: English
• ISSNs: 2041-210X

Abstract

Abstract: Fast and accurate estimates of wildlife abundance are an essential component of efforts to conserve ecosystems in the face of rapid environmental change. A widely used method for estimating species abundance involves flying aerial transects, taking photographs, counting animals within the images and then inferring total population size based on a statistical estimate of species density in the region. The intermediate task of manually counting the aerial images is highly labour intensive and is often the limiting step in making a population estimate. Here, we assess the use of two novel approaches to perform this task by deploying both citizen scientists and deep learning to count aerial images of the 2015 survey of wildebeest ( Connochaetes taurinus ) in Serengeti National Park, Tanzania. Through the use of the online platform Zooniverse, we collected multiple non‐expert counts by citizen scientists and used three different aggregation methods to obtain a single count for the survey images. We also counted the images by developing a bespoke deep learning method via the use of a convolutional neural network. The results of both approaches were then compared. After filtering of the citizen science counts, both approaches provided highly accurate total estimates. The deep learning method was far faster and appears to be a more reliable and predictable approach; however, we note that citizen science volunteers played an important role when creating training data for the algorithm. Notably, our results show that accurate, species‐specific, automated counting of aerial wildlife images is now possible. Foreign Language Abstract: Namna bora na ya haraka ya kuidadi wanyamapori ni jambo la msingi kwenye jitihada za uhifadhi na ikolojia hasa wakati huu wa mabadiliko ya tabia nchi. Mojawapo ya njia ambayo hutumika kuidadi wanyamapori ni kupiga picha za angani kwa kutumia ndege ndogo, kisha kuhesabu wanyamapori walioko kwenye picha hizo, na baadaye kukokotoa idadi kutokana na ukubwa wa eneo husika. Kazi ambayo huwa inachukua muda mrefu ni kuhesabu wanyamapori kwenye picha, ambayo huchelewesha ukokotoaji wa idadi ya wanyamapori. Katika utafiti huu, tumetumia njia mbili za kuhesabu idadi ya wanyamapori kwenye picha anga zilizochukuliwa mwaka 2015 katika Hifadhi ya Taifa ya Serengeti nchini Tanzania. Njia hizo ni kuhesabu kwa kutumia watalaam wa kujitolea kwenye mtandao wa kijamii na nyingine ni kwa kutumia mfumo maalum wa kompyuta wa kung'amua maumbo ya vitu. Kwa kutumia mtandao wa kijamii wa "Zooniverse" tuliweza kufanya majumuisho ya idadi ya nyumbu waliohesabiwa na wataalamu mbalimbali wa kujitolea wa kijamii kutoka kwenye picha husika. Kwa upande mwingine, tuliweza kuhesabu nyumbu wote kwenye picha husika kwa kutumia mfumo maalum wa kompyuta. Baadae, matokeo ya njia zote mbili yalilinganishwa. Baada ya kulinganisha makosa ya wazi kwenye njia ya kuhesabu kwa kutumia watu wa kujitolea kwenye mtandao wa kijamii, matokeo ya ukadiriaji wa idadi ya nyumbu kwa njia zote yalikuwa sawia. Hata hivyo, njia ya mfumo wa kompyuta ilionesha kutoa matokeo kwa haraka na ya kuaminika. Aidha, tafiti inaonesha wataalamu wa kujitolea wa kijamii ni wa muhimu katika kuandaa takwimu za kuufunza mfumo wa kompyuta uliotumika. Kwa ujumla, matokeo ya utafiti wetu yameonesha uwezekano wa kuhesabu na kutambua nyumbu kwenye picha kwa kutumia mfumo wa kompyuta .