Publications-Proceedings

Article View/Open

Publication Export

Google ScholarTM

NCCU Library

Citation Infomation

Related Publications in TAIR

題名 Feature descriptor based on local intensity order relations of pixel group
作者 Liao, Wen-Hung;Wu, Chia-Chen;Lin, Ming-Ching
廖文宏
貢獻者 資訊科學系
關鍵詞 Image recognition; Pattern recognition; Feature descriptors; Histogram of oriented gradients; Intensity difference; Local descriptors; Order patterns; Order relation; Recognition engines; Storage requirements; Pixels
日期 2017-04
上傳時間 3-Aug-2017 14:12:02 (UTC+8)
摘要 Robust image features are essential in building effective image recognition engines. These features can be constructed according to various principles, such the distribution of local gradients (Histogram of Oriented Gradients, HOG), the relationship between two pixels (Local Binary Descriptors, LBD), or local intensity order statistics (Local Intensity Order Patterns, LIOP). Because the feature dimension grows quickly as one considers the ordering relations of a group of N (N>2) pixels, few researchers have exploited local order statistics among a pixel set to define an image feature. In this paper, we propose a novel approach to construct a feature descriptor using local intensity order relations (LIOR) in a pixel group. In contrast to LIOP where the feature dimension increases drastically with the number of elements in a set, the size of LIOR is manageable. Moreover, LIOR ensures the stability of ordering by encoding the intensity differences as weights. Two different strategies for assigning the weights have been devised and tested. Experimental results indicate that the proposed methods yield better or comparable performance for different types of image degradation when compared to the original LIOP. Additionally, the storage requirement is significantly lower when the number of pixels in a group increases. © 2016 IEEE.
關聯 Proceedings - International Conference on Pattern Recognition, , 1977-1981
23rd International Conference on Pattern Recognition, ICPR 2016; Cancun CenterCancun; Mexico; 4 December 2016 到 8 December 2016; 類別編號CFP16182-ART; 代碼 127420
資料類型 conference
DOI http://dx.doi.org/10.1109/ICPR.2016.7899926
dc.contributor 資訊科學系zh_Tw
dc.creator (作者) Liao, Wen-Hung;Wu, Chia-Chen;Lin, Ming-Chingen_US
dc.creator (作者) 廖文宏zh_TW
dc.date (日期) 2017-04en_US
dc.date.accessioned 3-Aug-2017 14:12:02 (UTC+8)-
dc.date.available 3-Aug-2017 14:12:02 (UTC+8)-
dc.date.issued (上傳時間) 3-Aug-2017 14:12:02 (UTC+8)-
dc.identifier.uri (URI) http://nccur.lib.nccu.edu.tw/handle/140.119/111615-
dc.description.abstract (摘要) Robust image features are essential in building effective image recognition engines. These features can be constructed according to various principles, such the distribution of local gradients (Histogram of Oriented Gradients, HOG), the relationship between two pixels (Local Binary Descriptors, LBD), or local intensity order statistics (Local Intensity Order Patterns, LIOP). Because the feature dimension grows quickly as one considers the ordering relations of a group of N (N>2) pixels, few researchers have exploited local order statistics among a pixel set to define an image feature. In this paper, we propose a novel approach to construct a feature descriptor using local intensity order relations (LIOR) in a pixel group. In contrast to LIOP where the feature dimension increases drastically with the number of elements in a set, the size of LIOR is manageable. Moreover, LIOR ensures the stability of ordering by encoding the intensity differences as weights. Two different strategies for assigning the weights have been devised and tested. Experimental results indicate that the proposed methods yield better or comparable performance for different types of image degradation when compared to the original LIOP. Additionally, the storage requirement is significantly lower when the number of pixels in a group increases. © 2016 IEEE.en_US
dc.format.extent 209 bytes-
dc.format.mimetype text/html-
dc.relation (關聯) Proceedings - International Conference on Pattern Recognition, , 1977-1981en_US
dc.relation (關聯) 23rd International Conference on Pattern Recognition, ICPR 2016; Cancun CenterCancun; Mexico; 4 December 2016 到 8 December 2016; 類別編號CFP16182-ART; 代碼 127420en_US
dc.subject (關鍵詞) Image recognition; Pattern recognition; Feature descriptors; Histogram of oriented gradients; Intensity difference; Local descriptors; Order patterns; Order relation; Recognition engines; Storage requirements; Pixelsen_US
dc.title (題名) Feature descriptor based on local intensity order relations of pixel groupen_US
dc.type (資料類型) conference
dc.identifier.doi (DOI) 10.1109/ICPR.2016.7899926
dc.doi.uri (DOI) http://dx.doi.org/10.1109/ICPR.2016.7899926