A Methodology for Empirical Performance Evaluationof Page Segmentation AlgorithmsSong

Document page segmentation is a crucial preprocessing step in Optical Character Recognition (OCR) systems. While numerous page segmentation algorithms have been proposed , there is relatively less literature on comparative evaluation | empirical or theoretical | of these algorithms. For the existing performance evaluation methods, two crucial components are usually missing: 1) automatic training of algorithms with free parameters and 2) statistical and error analysis of experimental results. In this thesis, we use the following ve-step methodology to quantitatively compare the performance of page segmentation algorithms: 1) First we create mutually exclusive training and test datasets with groundtruth, 2) we then select a meaningful and computable performance metric, 3) an optimization procedure is then used to search automatically for the optimal parameter values of the segmentation algorithms, 4) the segmentation algorithms are then evaluated on the test dataset, and nally 5) a statistical error analysis is performed to give the statistical signiicance of the experimental results. The automatic training of algorithms is posed as an optimization problem and a direct search method | the simplex method | is used to search for a set of optimal parameter values. A paired-model statistical analysis and an error analysis are conducted to provide conndence intervals for the experimental results and to interpret the functionalities of algorithms. This methodology is applied to the evaluation of ve page segmentation algorithms, of which three are representative research algorithms and the other two are well-known commercial products, on 978 images from the University of Washington III dataset. It is found that the performances of the Voronoi, Docstrum and Caere segmentation algorithms are not signiicantly diierent from each other, but they are signiicantly better than that of ScanSoft's segmentation algorithm, which in turn is signiicantly better than that of X-Y cut.