Özet:

Text extraction is a process of converting the image text into plain text. Extraction of text plays a very important role in finding image texts, editing and archiving documents. However, extracting texts from complex degraded images is a tedious work. The main challenges in text detection and recognition from complex degraded images are orientation of text, font size, diversity of background, low quality, difference in color of texts and interference of noise. A significant problem in optical character recognition (OCR) process is to extract text from complex degraded images. Therefore, it is a challenging job to design an algorithm which can produce a good accuracy irrespective of the font sizes, background, color, quality and orientation of image or document texts. In this paper, we discuss about various text extraction techniques for complex degraded images and also discuss the performance analysis as well as compare the advantages and disadvantages of each techniques

Anahtar Kelimeler: