Extracting local reliable text regions to segment complex handwritten textlines

سال انتشار: 1392
نوع سند: مقاله کنفرانسی
زبان: انگلیسی
مشاهده: 1,193

فایل این مقاله در 5 صفحه با فرمت PDF قابل دریافت می باشد

استخراج به نرم افزارهای پژوهشی:

لینک ثابت به این مقاله:

شناسه ملی سند علمی:

ICMVIP08_129

تاریخ نمایه سازی: 9 بهمن 1392

چکیده مقاله:

Textline segmentation is an important preprocessbefore trying to recognize words. Handwritten texts includecomplex lines such as connected/overlapped, multi skewed, andcurved textlines. In the proposed approach, to overcome theseproblems, local reliable text regions are locally extracted for eachblock of a handwritten text. Text image is first filtered by a set ofdirectional 2D filters and filtered images are divided to a numberof overlapping blocks. The filtered block with the highestcontrast is selected to be used for text region detection.Experiments show that our proposed method accuratelysegments complex handwritten textlines

کلیدواژه ها:

Textline segmentation ، Local text region Directional filtering

نویسندگان

Majid Ziaratban

Engineering Department Golestan University, Gorgan, Iran

Fatemeh Bagheri

Engineering Department Golestan University, Gorgan, Iran