Combination of Morphological Operations with Structure based Partitioning and grouping for Text String detection from Natural Scenes

Text information in natural scene images serves as important clues for many image-based applications such as scene perceptive, content-based image retrieval, assistive direction-finding and automatic geocoding. Now days different approaches like countours based, Image binarization and enhancement based, Gradient based and colour reduction based techniques can be used for the text detection from natural scenes. In this paper the combination of morphological operations with structure based partitioning and grouping is used for efficient string detection from any natural scene images. A new framework based on image partition and connected components grouping is used to locate text regions embedded in those images. These described techniques are based on sibling’s method i.e. adjacent character grouping method and Text line grouping method. Text line grouping method can locate text strings situated at arbitrary orientations. In adjacent character grouping detected string is represented by a rectangle while in text line grouping detected string is represented by fitted line. Described techniques can find text strings by using structure-based partition and grouping methods.