Abstract
Faster R-CNN has advantages in object detection task. But in face of the variability of text and interference of the external factors, it cannot achieve perfect detection results in natural scene text detection. Moreover, the text detection algorithms based on deep learning need to use large data sets to train the network, while in some special scenarios where a mass of samples cannot be obtained, the performance of these algorithms is likely to be limited. How to accurately detect text in natural scene based on small data sets is a challenging issue. To address this issue, a multi-scale text feature extraction network with feature pyramid based on Faster R-CNN is proposed, which can accurately and comprehensively express complex and changeable text features in natural scenes even in the small data cases. Experiment results show that the proposed MSTD method is very competitive with existing related architectures.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
