Automatic individual identification of Saimaa ringed seals
In order to monitor animal population and to track individual animals in a non-invasive way, identification of individual animals based on certain distinctive characteristics is necessary. In this presentation, automatic image-based individual identification of the critically endangered Saimaa ringed seal (Phoca hispida saimensis) is considered. The ringed seals have a distinctive permanent pelage pattern that is unique to each individual. This can be used as a basis for the identification process. A framework is proposed that starts with the segmentation of the seal from the background and proceeds to various post-processing steps to make the pelage pattern more visible and the identification easier. For experiments, the novel dataset of the Saimaa ringed seal images has been collected and annotated. The framework is compared using two existing species independent individual identification methods and the dataset. The results show that the segmentation and proposed post-processing steps increase the identification performance.
Prof. Heikki Kalviainen,
Machine Vision and Pattern Recognition Laboratory (MVPR), School of Engineering Science, Lappeenranta University of Technology (LUT), Finland
Time and place: