Combining Slanted-Frame Classifiers for Improved HMM-Based Arabic Handwriting Recognition

The problem addressed in this study is the offline recognition of handwritten Arabic city names. The names are assumed to belong to a fixed lexicon of about 1,000 entries. A state-of-the-art classical right-left hidden Markov model (HMM)-based recognizer (reference system) using the sliding window approach is developed. The feature set includes both baseline-independent and baseline-dependent features. The analysis of the errors made by the recognizer shows that the inclination, overlap, and shifted positions of diacritical marks are major sources of errors. In this paper, we propose coping with these problems. Our approach relies on the combination of three homogeneous HMM-based classifiers. All classifiers have the same topology as the reference system and differ only in the orientation of the sliding window. We compare three combination schemes of these classifiers at the decision level. Our reported results on the benchmark IFN/ENIT database of Arabic Tunisian city names give a recognition rate higher than 90 percent accuracy and demonstrate the superiority of the neural network-based combination. Our results also show that the combination of classifiers performs better than a single classifier dealing with slant-corrected images and that the approach is robust for a wide range of orientation angles.     

IBLC effect leading to colossal dielectric constant in layered structured Eu2CuO4 ceramic

Dielectric (ε_r′) studies of phase pure T′-type Eu₂CuO₄ ceramics of two markedly different grain sizes (D), prepared by (i) conventional powder mixing and (ii) citrate complexation-Pechini process, have been carried out in the frequency range 0.1 Hz to 1 MHz, and at temperatures −100 °C to 150 °C. ε_r′ is found to be highly grain size dependent. For the sample with coarse bar-like grains (D²~17×6 μm²) ε_r′ is >10³, and for the finer grain size sample with bimodal distribution (D₁~1 μm, D₂~3 μm) ε_r′ is ~10⁵; for both the samples, high ε_r′ value is nearly frequency independent over 500 Hz≤f<100 kHz and T≥30 °C. The impedance spectroscopy (IS) study has clearly shown that both, the coarse- and the fine-grained samples consist of semiconducting grains and insulating grain boundaries that primarily lead to an internal barrier layer capacitance (IBLC) effect. And thus, manifest colossal dielectric constant (ε_r′>10³) in Eu₂CuO₄ ceramics. The smaller grain size (Pechini) sample, with over an order higher number of grains and grain boundary network, showing over an order higher ε_r′ (~10⁵) compared to the coarse grained one, further endorses the IBLC effect.