¹School of Information and Technology, Northwest University, Xi’an, China
²School of Information and Computing Sciences, North University for Nationalities, Yinchuan, China
³School of Physics & Electrical Information Engineering, Ningxia University, Yinchuan, China

---

Abstract—In this paper we propose a novel approach called three feature functions based on Radon transform directly. The corresponding features are defined as the symmetry of the shape, the width of contour, the fullness of the shape. They reveal the information of visual and the inner structures of shape using characteristics of Radon projective space. Theoretical and experimental results verify that the proposed method has strong capability and robustness in handling a variety of shapes and is invariant to rotation.

Index Terms—Radon transformation, three feature functions of descriptors, Shape analysis, Image retrieval, Object recognition.

[PDF]

Cite: Ziping Ma, Baosheng Kang, Bin Xiao, Jinlin Ma, "Three Feature Functions based on Radon Transform for Image Retrieval and Object Recognition," Journal of Software vol. 8, no. 2, pp. 397-403, 2013.