作者: Tariq Alusaifeer , Sheela Ramanna , Christopher J. Henry , James Peters
DOI: 10.1007/978-3-642-41299-8_24
关键词:
摘要: This paper presents a parallel version of the Maximal Clique Enumeration MCE approach for discovering tolerance classes. Finding such classes is computationally complex problem, especially in case large data sets or content-based retrieval applicationsCBIR. The GPU implementation an extension earlier work by authors on finding efficient methods computing images. experimental results demonstrate that GPU-based algorithm faster than serial and can perform computations with higher values e.