AQUISAR - Image Retrieval in Underwater Webcam Images

Common webcams yield a huge amount of images, but most of them are quite boring for lack of individual fascinating entities. In this project we analyse approaches of common image and video retrieval to develop a system for filtering interesting images shot by a webcam. The system AQUISAR (Aquarium Image Segmentation and Retrieval) is an initial framework for collecting, pre-processing and retrieving webcam images of a public aquarium setup. It combines various techniques of CBIR (Content Based Image Retrieval) and computer vision to extract interesting image contents. Since the used webcam is located in an aquarium this work is affected by the upcoming task of underwater computer vision. A recognition of webcam positions is performed as well as an image segmentation according to the content. Based on a "query by example" similarity search results in a list of images with regions resembling those in the query image. A specific object, in this setup mostly sharks, should be found in various images.