BibTeX:

@article{IJIRSTV4I2057,
     title={Markov Random Field Region Based Text Detection and Segmentation by Stroke Width Transformation},
     author={Renuka and Dr. Sujata Terdal},
     journal={International Journal for Innovative Research in Science & Technology},
     volume={4},
     number={2},
     pages={195--200},
     year={},
     url={http://www.ijirst.org/articles/IJIRSTV4I2057.pdf},
     publisher={IJIRST (International Journal for Innovative Research in Science & Technology)},
}

Text detection in handwritten image has gained widespread interests. Detection of the texts from handwritten images is a challenging problem due to the multiple fonts, different sizes, various orientations and alignment, reflections, shadows, the complexity of image background. Text detection and segmentation from handwritten images are useful in many applications. We present a method called Markov Random Method for image operator that seeks to find the value of each image pixel, and demonstrate their use on the task of text detection in natural, which makes it fast and robust enough to eliminate the need for multi scale computation or scanning windows. A notable work, which is Markov Random Field method (MRF), has been attracting much interest due to its simplicity and efficiency. However, the Stroke Width Transform (SWT), and OCR has difficulty in situations like blur, low contrast, and illumination change, since it is highly relies on the outcome from the edge detector. Here region based approach MRF (Markov Random Field) with stroke width transform (SWT) method is proposed for automatic detection and extraction of text from handwritten images and explains the methodology to extract and recognize text. The applications of region based image segmentation by MRF for text detection from image has given the scope to us to include the important technologies like Text Information Extraction, Stroke Width Transformation etc. which will helps to improve the efficiency of work.

Bounding box, discrete wavelet transform, Markov random field, Text localization and Stroke width transform