作者: Prasanna Mulgaonkar , Gregory K. Myers , John P. Marcotullio , Hrishikesh B. Aradhye
DOI:
关键词:
摘要: A method and apparatus are provided for analyzing an electronic communication containing imagery, e.g., to determine whether or not the is a spam communication. In one embodiment, inventive includes detecting more regions of imagery in received applying pre-processing techniques locate (e.g., blocks lines) text that may be distorted. The then analyzes content indicates spam. specialized extraction rectification embedded followed by optical character recognition processing applied extract their therefrom. another keyword shape-matching detect presence absence spam-indicative words from text. other attributes extracted regions, such as size, location, color complexity used build evidence against