Document Image Analysis: An Executive Briefing
Document Image Analysis: An Executive Briefing
Skew Detection of Document Images by Focused Nearest-Neighbor Clustering
ICDAR '99 Proceedings of the Fifth International Conference on Document Analysis and Recognition
Multi-Skew Detection of Indian Script Documents
ICDAR '01 Proceedings of the Sixth International Conference on Document Analysis and Recognition
Improved Nearest Neighbor Based Approach to Accurate Document Skew Estimation
ICDAR '03 Proceedings of the Seventh International Conference on Document Analysis and Recognition - Volume 1
DAS '08 Proceedings of the 2008 The Eighth IAPR International Workshop on Document Analysis Systems
SURF: speeded up robust features
ECCV'06 Proceedings of the 9th European conference on Computer Vision - Volume Part I
Hi-index | 0.00 |
This paper proposes a new approach to a method to estimate a skew angle of a rotated document image. This is realized by using Speeded-Up Robust Features (SURF), and the goal is that it enables the image to be rotated back to the correct orientation. SURF detects a number of keypoints both from the reference image on which a set of standard alphabets (e.g. letter eaf through ezf in a certain font) are written, and the image of the rotated document. Two nearest features each from the reference image and the input image are compared to decide to how many degrees the feature in the input image is rotated. Finally the skew angle of the whole input image(the global skew angle) is decided by the majority of the total votes of angles that have been calculated as mentioned above.