作者: J. Marcus Rowcliffe , Juliet Field , Samuel T. Turvey , Chris Carbone
DOI: 10.1111/J.1365-2664.2008.01473.X
关键词: Camera trap 、 Image sensor 、 Computer vision 、 Ecology 、 Range (statistics) 、 Process (computing) 、 Computer science 、 Estimation 、 Artificial intelligence 、 Abundance estimation 、 Density estimation 、 Trapping
摘要: Summary 1Density estimation is of fundamental importance in wildlife management. The use camera traps to estimate animal density has so far been restricted capture–recapture analysis species with individually identifiable markings. This study developed a method that eliminates the requirement for individual recognition animals by modelling underlying process contact between and cameras. 2The model provides factor linearly scales trapping rate density, depending on two key biological variables (average group size day range) characteristics sensor (distance angle within which it detects animals). 3We tested approach an enclosed park known abundances four species, obtaining accurate estimates three out cases. Inaccuracy fourth was because biased placement cameras respect distribution this species. 4Synthesis applications. Subject unbiased measurement parameters, opens possibility reduced labour costs estimating may make possible where not previously. We provide guidelines effort required obtain reasonably precise estimates.