Fig. 8 | Scientific Reports

Fig. 8

From: Addressing significant challenges for animal detection in camera trap images: a novel deep learning-based approach

Fig. 8

Left: situations where the global model fails to detect the animal in the images due to challenging conditions such as darkness, proximity of the animal to the camera, or the animal being camouflaged against the background. Right: same situations where the expert models of our two-stage pipeline have successfully corrected these issues, accurately detecting the animal

Back to article page