부산대학교 도서관

With the widespread application of multimedia and network technologies, natural scene data such as images and videos are becoming increasingly abundant. The text information contained in images and videos is an important part of understanding multimedia content. The detection and recognition of text information in images and videos is of great research significance in fields such as video retrieval and visual assistance systems. This paper analyzed and studied text detection in multimedia data in natural scenes, then proposed an OCR text detection method based on Chinese character features. The experimental results show that this method has good performance in detecting Chinese characters in natural scenes.