TY - GEN
T1 - Riesz Transform Associated with Linear Canonical Transform
T2 - 6th International Conference on Image, Video and Signal Processing, IVSP 2024
AU - Chen, Jian Yi
AU - Li, Bing Zhao
N1 - Publisher Copyright:
© 2024 ACM.
PY - 2024/3/14
Y1 - 2024/3/14
N2 - This article presents a novel approach called the generalized Riesz transform (GRT) that addresses the issue of negative frequency components in linear canonical transform (LCT). The GRT is an isotropic extension of the two-dimensional Hilbert transform and finds applications in computer vision, medical imaging, and pattern recognition, with specific emphasis on edge detection, texture analysis, and feature extraction. The article defines the GRT within the framework of linear canonical domains using three-dimensional harmonic fields and properties of the LCT, and provides a spatial domain filtering formula for GRT. The effectiveness of GRT in capturing texture features is demonstrated through the concept of generalized monogenic signal, and an envelope detection algorithm based on GRT and phase consistency is proposed, showcasing superior performance compared to conventional methods.
AB - This article presents a novel approach called the generalized Riesz transform (GRT) that addresses the issue of negative frequency components in linear canonical transform (LCT). The GRT is an isotropic extension of the two-dimensional Hilbert transform and finds applications in computer vision, medical imaging, and pattern recognition, with specific emphasis on edge detection, texture analysis, and feature extraction. The article defines the GRT within the framework of linear canonical domains using three-dimensional harmonic fields and properties of the LCT, and provides a spatial domain filtering formula for GRT. The effectiveness of GRT in capturing texture features is demonstrated through the concept of generalized monogenic signal, and an envelope detection algorithm based on GRT and phase consistency is proposed, showcasing superior performance compared to conventional methods.
KW - envelope detection
KW - generalized monogenic signal
KW - Generalized Riesz transform
KW - linear canonical transform
UR - http://www.scopus.com/inward/record.url?scp=85196218910&partnerID=8YFLogxK
U2 - 10.1145/3655755.3655773
DO - 10.1145/3655755.3655773
M3 - Conference contribution
AN - SCOPUS:85196218910
T3 - ACM International Conference Proceeding Series
SP - 135
EP - 141
BT - IVSP 2024 - 2024 6th International Conference on Image, Video and Signal Processing
PB - Association for Computing Machinery
Y2 - 14 March 2024 through 16 March 2024
ER -