SKEW DETECTION OF PRINTED DOCUMENT IMAGES USING CHARACTER ORIENTATION ESTIMATION
Yangli Wang and Rui Song, School of Telecommunications Engineering, Xidian University, Xi’an China
In the capturing process of document images, skew is usually unavoidable. This skew may seriously degrade the performance of subsequent processing steps such as layout analysis and segmentation. As such, skew detection and correction is a necessary step in the document image processing pipeline. In this paper we propose a new skew detection method using character orientation estimation. The first step of the approach is classifying the connected components in a binarized document image into text/non-text. This is accomplished by a convolutional neural network (CNN). The next step of the proposed approach
is estimating character orientation using the minimum area rect surrounding a character. Initial document skew is then estimated with the first ?? largest characters and their orientation. This initial skew is then refined by clustering deskewed nearest characters and minimizing sum of the area of their surrounding rects in a small search range. As text/non-text are separated first, and only text is involved in document skew estimation, the proposed method overcomes the drawbacks of projection profile based methods, which can not deal well with documents containing other elements, and are weak in handling multicolumn documents when the lines of different columns are not well aligned. The proposed approach also has the advantage over those methods estimating text line orientation, such as the Hough transform
based ones, they have difficulties in choosing representing points for characters as characters are of different height. Simulation results on test document images show that, the proposed approach has achieved a mean estimation error of as low as 0.1169°. It also has the advantage of potentially saving the geometric layout analysis step in the conventional document image processing pipeline.
Character Orientation, Convolutional Neural Network, Minimum Area Rect, Character
Cluster, Document Skew Detection
PERFORMANCE EVALUATION OF FEATURE EXTRACTION FROM BIOELECTRICAL SIGNAL USING SMARTWATCH
Timibloudi S Enamamu1,2, Abayomi M Otebolaku1 and Joy Dany2, 1Sheffield Hallam University, Sheffield, UK and 2Communications and Network Research (CSCAN) Plymouth University, United Kingdom
This work provides a process for selecting suitable features using biorthogonal wavelet decomposition of signal from a non-intrusive extraction. A smartwatch is used for extracting bioelectrical signal before decomposing the signal into sub-bands of Detail and Approximation Coefficient. A detail experiment is conducted extracting suitable statistical features from the bioelectrical signal from 30 subjects using different biorthogonal wavelet family. Ten features are extracted using Biorthogonal wavelet to decompose the signal into three levels of sub-band Detail and Approximation Coefficient and features extracted from each levels the decomposed Detail and Approximation Coefficients. Comparison analysis is done after the classification of the extracted features based on the Equal Error Rate (EER). Using Natural Network (NN) classifier, Biorthogonal Wavelet Detail Coefficient Sub-band level 3 of bior1.1 achieved the best result of EER 13.80% with the fusion of the best sub-band three levels of bior1.1 achieving a better result of 12.42% EER.
Bioelectrical signals, biorthogonal wavelet, approximation coefficients, detail coefficient, wavelet transform.