Scene sketch generation using mixture of gradient kernels and adaptive thresholding

Sidike Paheding; Almabrok Essa; Vijayan Asari

doi:10.1117/12.2226032

Translator Disclaimer

20 April 2016 Scene sketch generation using mixture of gradient kernels and adaptive thresholding

Sidike Paheding, Almabrok Essa, Vijayan Asari

Author Affiliations +

Proceedings Volume 9845, Optical Pattern Recognition XXVII; 98450N (2016) https://doi.org/10.1117/12.2226032
Event: SPIE Defense + Security, 2016, Baltimore, Maryland, United States

Abstract

This paper presents a simple but eﬀective algorithm for scene sketch generation from input images. The proposed algorithm combines the edge magnitudes of directional Prewitt diﬀerential gradient kernels with Kirsch kernels at each pixel position, and then encodes them into an eight bit binary code which encompasses local edge and texture information. In this binary encoding step, relative variance is employed to determine the object shape in each local region. Using relative variance enables object sketch extraction totally adaptive to any shape structure. On the other hand, the proposed technique does not require any parameter to adjust output and it is robust to edge density and noise. Two standard databases are used to show the eﬀectiveness of the proposed framework.

Citation Download Citation

Sidike Paheding, Almabrok Essa, and Vijayan Asari "Scene sketch generation using mixture of gradient kernels and adaptive thresholding", Proc. SPIE 9845, Optical Pattern Recognition XXVII, 98450N (20 April 2016); https://doi.org/10.1117/12.2226032

ACCESS THE FULL ARTICLE