Abstract Scene text detection has become a popular research in computer vision and pattern recognition field in recent years because of the accurate and rich information carried by scene text. Now component-based methods have become the trend, and the detection result is largely determined by the success of filtering text-like non-text regions. The main task of this paper is to reduce the time complexity... Scene text detection has become a popular research in computer vision and pattern recognition field in recent years because of the accurate and rich information carried by scene text. Now component-based methods have become the trend, and the detection result is largely determined by the success of filtering text-like non-text regions. The main task of this paper is to reduce the time complexity without a big fall in recall. First an enhanced multi-channels MSER model is introduced. Before extracting MSER, the image is sharpened by using the Laplacian and Gaussian blur and multi-channel is utilized, then the step of the threshold used in MSER algorithm is set to the minimum in order to get add the more refined MSERs. Second, two novel scene text features local contrast and boundary key points are proposed to better distinguish text regions from non-text regions. Finally, a fast text grouping algorithm is achieved which reduces the time complexity from O (n2) to O(nlog2n). Experiments on both ICDAR 2011 and ICDAR 2013 show that the recall of the proposed method is improved by 3%. Read More Read Less

Abstract The rapid development of the social Web has resulted in diverse Web 2.0 applications for accessing various publishing information related to social activity such as conference activities, public welfare activities, and cultural events. Due to the masses of social activities information available on the Internet, a major challenge is how to access them efficiently. The traditional keyword search is imprecise and inconvenient for... The rapid development of the social Web has resulted in diverse Web 2.0 applications for accessing various publishing information related to social activity such as conference activities, public welfare activities, and cultural events. Due to the masses of social activities information available on the Internet, a major challenge is how to access them efficiently. The traditional keyword search is imprecise and inconvenient for web surfing to find interest information on the internet. This study addresses this issue by defining a Multi-layered Cloud Service Framework (MCSF) for integrating ontology and Web 2.0 technologies into cloud computing environment. The proposed MCSF is used to develop an Intelligent Social Activity Cloud Platform (ISACP) that provides customized social activity for personal needs. To demonstrate the feasibility of ISACP, the experimental results illustrate the efficiency and effectiveness of the proposed approach. Read More Read Less

Abstract Prepositional-phrase (PP) attachment ambiguity is a pervasive problem in natural language processing, and at times it poses significant challenges to a computer system to resolve this ambiguity. In literature, different approaches have been proposed to address PP-attachment ambiguity, but to the best of our knowledge, there is no published work which surveys such approaches. This survey paper compares the standard methods that attempt... Prepositional-phrase (PP) attachment ambiguity is a pervasive problem in natural language processing, and at times it poses significant challenges to a computer system to resolve this ambiguity. In literature, different approaches have been proposed to address PP-attachment ambiguity, but to the best of our knowledge, there is no published work which surveys such approaches. This survey paper compares the standard methods that attempt to resolve PP-attachment ambiguities in natural language processing. We also provide a taxonomy of various ambiguities, which may arise at different levels during the language-processing task. There are two methods employed in natural language processing concerning new approaches: the first technique called the rule-based method and the second called statistical approach. Read More Read Less