The Green Toad Example: A Comparison of Pattern Recognition Software

Abstract

Abstract Background Individual identification of animals is important for assessing the size and status of populations. Photo-based approaches, where animals are recognized by naturally occurring and visually identifiable features, such as color patterns, are cost-effective methods for this purpose. We compared five available programs for their power to semi-automatically identify dorsal patterns of the European green toad (Bufotes viridis). Method We created a data set of 200 pictures of known identity, two pictures for each individual, and analyzed the percentage of correctly identified animals for each software. Furthermore, we employed a generalized linear mixed model to identify important factors contributing to correct identifications. We used these results to estimate the population size of our hypothetical population. Conclusions The freely available HotSpotter application was the software which performed by far the best for our green toad example, identifying close to 100% of the photos correctly. The animals’ sex highly significantly influenced detection probability, presumably because of sex-specific differences in the pattern contrast. Population estimates were close to the expected 100 for HotSpotter, but for the other applications population size was highly overestimated. Given the clarity of our results we strongly recommend the HotSpotter software, which is a highly efficient tool for individual pattern recognition.