Deep Neural Networks based Optical Character Recognition for Parking Signs
Name
Muhammed Adil Yatkin
Abstract
Optical Character Recognition(OCR) is a common computer recognition technology used to extract textual information from an image or document. OCR technologies often detect and extract the text contained in images and make it computer-readable. As shown in many previous research works, text written images do not always have good quality to be recognized by automatic OCR systems. In our research, we evaluated the performance of different automatic OCR systems such as Google, Amazon etc. by preparing our own parking sign OCR datasets, and we deeply investigated state-of-the-art mechanisms of recent works, and we develop an efficient, with high accuracy system to recognize parking sign images.
Graduation Thesis language
English
Graduation Thesis type
Master - Computer Science
Supervisor(s)
Gholamreza Anbarjafari, Till Quack
Defence year
2020