During counting, we remove previous track IDs to allow track IDs to be reused counting forwards or backwards just in case, the tracking algorithm cannot distinguish unique salmon that swim by. This however creates an issue for reviewers as these seperate tracks get lumped together in a single track because of them being the same track ID. This will set the track IDs to be unique when saving them in the detection text files.
During counting, we remove previous track IDs to allow track IDs to be reused counting forwards or backwards just in case, the tracking algorithm cannot distinguish unique salmon that swim by. This however creates an issue for reviewers as these seperate tracks get lumped together in a single track because of them being the same track ID. This will set the track IDs to be unique when saving them in the detection text files.