摘要 |
The invention relates to a method and a device for identification of a character string in an image. The method comprises the following steps of: extracting a character string area in the image, carrying out an over-segmentation to the character string area, and carrying out the identification of the character string included in the character string area with use of a route searching strategy and based on at least one feature selected from a language category context feature and a character placeholder context feature. |