Coherent lensless matched filter and an application to feature extraction in character recognition

• Main Author: Bage, Marc J.
• Language: English
• Published: 2010
• Online Access: http://hdl.handle.net/2429/19592

Reduction of space variance in coherent matched filtering is considered in the context of parallel character recognition. The space variance due to vignetting apertures is eliminated by synthesizing the effect of the second lens of the Vander Lugt matched filter (MT) inside the hologram. This leads to two types of lensless matched filter (LLMF) depending on the curvature of the reference beam and the relative scale of the recorded and readout signals. For a given space-bandwidth product, two layouts of recorded and readout signals are found to minimize the space variance due to the volume of the recording medium. They are well suited for line-by-line character recognition. A study of the holographic aberrations indicates for which of these layouts the space variance of the root mean square (rms) distortion is smallest. Computer-generated plots are used to investigate the space variance of the rms astigmatism of the MF and LLMF. The condition that makes the spherical aberration and coma null is also derived. Various aspects of the LLMF are examined. The LLMF acts simultaneously as a MF and a Fresnel zone plate. Therefore, when the hologram is displaced, the LLMF behavior results from the combined displacements of its filter component and lens component. In the output plane, the correlation component only is focused by the lens component. The other diffracted output components are unfocused images. They contribute noise that affects the correlation signal. In practical cases of parallel processing, the signal-to-noise ratios of the MF and LLMF at the same reference beam angle are comparable and the resolution of available recording media places no restriction on the field of view or the signals bandwidth. Improvement of the recognition performance is another concern of the thesis. It is shown experimentally that a filter matched to character features (FEMF) discriminates better between the similar looking characters O and Q than a high-pass filter matched to their high frequency components (HPMF). Yet its sensitivity to variants is not as severe as that of the EPMF when the feature extractor is a low-pass filter. Also this low-pass filtering reduces the aberrations and allows for a detector with a coarser resolution. The increased reliability of the FEMF is achieved at the expense of a reduced diffraction efficiency and of a more critical registration of the readout signal. Experimental considerations emphasize the importance of phase accuracy in the input plane. Finally, suggestions for further research are presented. === Applied Science, Faculty of === Electrical and Computer Engineering, Department of === Graduate