作者: Andreas Girgensohn , Qiong Liu , Patrick Chiu
DOI:
关键词: Discrete cosine transform 、 Face detection 、 Compressibility 、 Variation (game tree) 、 Computer science 、 Artificial intelligence 、 Computer vision 、 Image (mathematics)
摘要: An algorithm for finding regions of interest (ROI) in images and photos based on an information driven approach which sub-blocks image are analyzed content or compressibility the discrete cosine transform. The low grouped into ROIs using a morphological technique. Unlike other algorithms that geared highly specific types ROI (e.g. face detection), method present invention is generally applicable to arbitrary photos. A center-weighted variation can produce better results certain photo applications. be used with several applications, including Stained-Glass collages Pan-and-Scan presentations.