Title :
Address block location on complex mail pieces
Author :
Yu, Bin ; Jain, Anil K. ; Mohiuddin, Moidin
Author_Institution :
Dept. of Comput. Sci., Michigan State Univ., East Lansing, MI, USA
Abstract :
Unlike simple letter envelopes which have a high degree of global spatial structure among a limited number of entities, many mail pieces such as magazines usually have an address block printed on a label which can be pasted in an arbitrary position and orientation among text, graphics and images on a magazine cover, which is often shrink wrapped. This work concentrates on address block location for complex mail pieces with an arbitrary layout of printed entities. Using a bottom-up approach, a pyramid model is created for the address block. Based on this model, some features are extracted for assigning a confidence value to the located address blocks. The typical processing time for locating a destination address block is 0.1 seconds on a SGI Indy workstation
Keywords :
document image processing; feature extraction; image segmentation; mailing systems; optical character recognition; postal services; SGI Indy workstation; address block location; bottom-up approach; complex mail pieces; confidence value; feature extraction; global spatial structure; graphics; images; letter envelopes; magazines; mail sorting; processing time; pyramid model; text; Aging; Computer science; Feature extraction; Graphics; Layout; Plastics; Postal services; Printers; Sorting; Workstations;
Conference_Titel :
Document Analysis and Recognition, 1997., Proceedings of the Fourth International Conference on
Conference_Location :
Ulm
Print_ISBN :
0-8186-7898-4
DOI :
10.1109/ICDAR.1997.620641