• SANTHI K

Abstract: Image-based mostly text data are effective tags or clues for several security applications. In natural pictures, text characters and strings typically seem in close signboards and supply significant information of close settings and objects. Text characters and strings in a natural scene will offer valuable data for several applications. Extracting text directly from natural scene pictures may be a difficult task owing to various text patterns and variant background interferences. This paper proposes a way of text extraction exploitation scene image. The planned paper styles a discriminative character descriptor by combining many progressive feature detectors and descriptors. Second, it models character structure at every character category by coming up with stroke configuration maps. the look is compatible with the appliance of scene text extraction in pictures. The paper is developed to point out the effectiveness of our planned text extraction methodology on image dataset with text data extraction from close objects

Keywords— Text Detection, Scene Images Dataset, Text Extraction, Segmentation, Color Model.

"EFFECTIVE SECURE TEXT EXTRACTION FOR IMAGE DATASET USING MULTI COLOR SEGMENTATION WITH STRUCTURE ANALYSIS MODEL ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 6, page no.810-818, June 2019, Available :http://www.jetir.org/papers/JETIR1907B40.pdf

"EFFECTIVE SECURE TEXT EXTRACTION FOR IMAGE DATASET USING MULTI COLOR SEGMENTATION WITH STRUCTURE ANALYSIS MODEL ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 6, page no. pp810-818, June 2019, Available at : http://www.jetir.org/papers/JETIR1907B40.pdf