DocumentCode
249272
Title
Recovering planar projections of printed clustered-dot halftones
Author
Pollard, Stephen ; Ulichney, Robert ; Gaubatz, Matthew
fYear
2014
fDate
27-30 Oct. 2014
Firstpage
4226
Lastpage
4230
Abstract
This paper presents a novel method to recover planar projections (homographies) from perspectively distorted images of clustered-dot halftones. The method uses the variation of the local affine deformation of the halftone (estimated using a Fourier method) to recover the additional projective components. Evidence provided from simulated distortion and mobile capture shows that the approach provides an effective first step in the problem of using mobile device to recover stenographic information embedded in unknown halftone images, as pixel-accurate alignment is needed to recover the data.
Keywords
Fourier transforms; image processing; steganography; Fourier transform; halftone images; homographies; image distortion; local affine deformation; mobile device; pixel-accurate alignment; planar projection recovery; printed clustered-dot halftones; projective components; stenographic information recovery; Accuracy; Approximation methods; Cameras; Discrete Fourier transforms; Mobile handsets; Fourier Transform; Halftone; Homography; Planar Projection; Stegatone; Vanishing Points;
fLanguage
English
Publisher
ieee
Conference_Titel
Image Processing (ICIP), 2014 IEEE International Conference on
Conference_Location
Paris
Type
conf
DOI
10.1109/ICIP.2014.7025858
Filename
7025858
Link To Document