Sorting camera trap images

Camera trapping is used by conservation biologists to study snow leopards. In this research, we introduce techniques that sort camera trap images into sets with snow leopards and those without. We use Robust Principal Component Analysis, thresholding, and binary morphology to create motion templates. Cascade Object Detector finds spots in the camera trap images. The number of spots that overlap with the motion template and their density are among the features input to the Support Vector Machine classifier that sorts the images. Our results are very promising: we achieve an average classification accuracy of 93.74%.