1) 三重大学大学院 工学研究科 情報工学専攻
光学文字認識, テロップ認識, アラビア語認識, コーミングノイズ, 編集距離, 流れテロップ接続
The authors have conducted studies on Arabic telop recognition to develop a system for video retrieval by keyword to index and edit Arabic broadcast programs received daily and stored in a big database. This paper describes a dedicated OCR for recognizing low resolution telop in video images. A telop recognition system consisting of text line extraction, word segmentation and segmentation-recognition of words is developed and the performance was experimentally evaluated using datasets of frame images extracted from AlJazeera broadcasting programs. Character recognition of moving telop is difficult due to combing noise caused by the interlacing of scan-lines. A technique to detect and eliminate the combing noise to correctly recognize the moving telop is proposed. This paper also proposes a technique based on insertion operation with minimum edit distance between successive two telops to connect them. The method to connect the moving telops is necessary for automatic language translation. The proposed method using edit distance for bi-gram sequence of telops (Method-B) is shown to be robust to recognition error of characters and successfully connect the telops.
編集・発行 : 一般社団法人 電気学会 制作・登載者 : 三美印刷株式会社