DocumentCode
2551690
Title
A method of container codes detection and segmentation by using improved projection algorithm
Author
Tai, Shen-Chuan ; Chang, Tang-Yu ; Wang, Shih-Yu
Author_Institution
Dept. of Electr. Eng., Nat. Cheng Kung Univ., Tainan, Taiwan
fYear
2011
fDate
21-25 June 2011
Firstpage
587
Lastpage
592
Abstract
In recent year, surveillance system is widely used in our daily life. The surveillance system is usually applied to monitor specify target for some purpose. Intelligence surveillance system can helps us to get interested information from the monitor screen. In addition to applied to security monitor in our living environment and people or traffic flow statistics, it also widely used in management. Therefore, how to use existing surveillance system to improve the accuracy rate with reducing the human resources is an important issue. Most of the current surveillance systems focus on license plate detection, and we try to extend the detect object to the containers. The existing license plate detection algorithm would face some difficulties with using on container codes detection. This paper presents a character region extraction method for containers. We also improved projection method to reduce the effect caused by bars or marks on container and make it can be used to detect irregular rows of characters.
Keywords
containers; feature extraction; image segmentation; monitoring; object detection; road traffic; surveillance; character region extraction method; container code detection; container code segmentation; human resources; intelligence surveillance system; license plate detection; object detection; screen; security monitoring; target monitoring; traffic flow statistics; Containers; Image color analysis; Image edge detection; Image segmentation; Licenses; Pixel; Vehicles;
fLanguage
English
Publisher
ieee
Conference_Titel
Intelligent Control and Automation (WCICA), 2011 9th World Congress on
Conference_Location
Taipei
Print_ISBN
978-1-61284-698-9
Type
conf
DOI
10.1109/WCICA.2011.5970581
Filename
5970581
Link To Document