Two-tier architecture for unconstrained handwritten character recognition

In this paper, we propose an approach that combines the unsupervised and supervised learning techniques for unconstrained handwritten numeral recognition. This approach uses the Kohonen self-organizing neural network for data classification in the first stage and the learning vector quantization (LVQ) model in the second stage to improve classification accuracy. The combined architecture performs better than the Kohonen self-organizing map alone. In the proposed approach, the collection of centroids at different phases of training plays a vital role in the performance of the recognition system. Four experiments have been conducted and experimental results show that the collection of centroids in the middle of the training gives high performance in terms of speed and accuracy. The systems developed also resolve the confusion between handwritten numerals.     

Sub-50-nm physical gate length CMOS technology and beyond usingsteep halo

Sub-50-nm CMOS devices are investigated using steep halo and shallow source/drain extensions. By using a high-ramp-rate spike annealing (HRR-SA) process and high-dose halo, 45-nm CMOS devices are fabricated with drive currents of 650 and 300 μA/μm for an off current of less than 10 nA/μm at 1.2 V with T_ox^inv =2.5 nm. For an off current less than 300 nA/μm, 33-nm pMOSFETs have a high drive current of 400 uA/μm. Short-channel effect and reverse short-channel effect are suppressed simultaneously by using the HRR-SA process to activate a source/drain extension (SDE) after forming a deep source/drain (S/D). This process sequence is defined as a reverse-order S/D (R-S/D) formation. By using this formation, 24-nm nMOSFETs are achieved with a high drive current of 800 μA/μm for an off current of less than 300 μA/μm at 1.2 V. This high drive current might be a result of a steep halo structure reducing the spreading resistance of source/drain extensions     

Determining pheromone content of hairpencils from individual virgin males ofPseudaletia unipuncta (Haw.) (Lepidoptera: Noctuidae)

Levels of benzaldehyde recovered from virginPseudaletia unipuncta (Haw.) males were not influenced by (1) the time hairpencils remained in the solvent (1-72 hr), (2) anesthetization or agitation of males prior to excision of hairpencils, or (3) the time (photophase or scotophase) that hairpencils were excised. Thus the interindividual variability observed is not a methodological artifact. Most males had similar concentrations in both hairpencils, although in some cases only one hairpencil contained pheromone. In one case, a male with partially extruded hairpencils had no benzaldehyde at all. Neither pupal weight nor hairpencil length proved to be reliable indicators of pheromone content.Contribution No. 373, Département de biologie, Université Laval.     

Refractive index profiling-state of the art

Problems associated with different calibration techniques and some instrumental effects that can determine instrumental accuracy are discussed. It is shown how transmission effects can be observed and used to generate a correction curve for refracted near-field instruments. High-quality fiber slices needed for axial interferometry were used to obtain refractive index differences on suitable fibers to ±1% with the measurements traceable to national standards. Interferometry was used to calibrate a multiple-step fiber which is available in 10-m lengths with a certificate of calibration. The limitations of the technique are discussed, and preliminary results of a study on the effects of residual stress and stress relief on measurements are given. Comparisons are made between measurements made by axial interferometry and refracted near field on a range of fibers. It is shown that commercial profilers can produce reliable measurements of refractive index difference, absolute refractive index, profile shape, and numerical aperture that agree well with measurements by other techniques     

Influence of Al and Ni concentration on the Martensitic transformation in Cu-Al-Ni shape-memory alloys

The martensitic transformation temperatures and the types of martensitic phases have been determined in a wide concentration range of technological interest for Cu-Al-Ni shape-memory alloys (SMAs) A stability diagram of martensitic phases as a function of alloy concentration has been determined. It is found that when the aluminum content increases, the transformation changes from β ₃ ⇒ β′₃ to β ₃ ⇒ γ′₃, with an intermediate concentration range where both martensites coexist due to a β ₃ ⇒ γ′₃+β′₃ transformation. On the other hand, an increase of nickel content stabilizes the martensite β′₃, changing from a mixed β ₃ ⇒ γ′₃ + β′₃ to a single β ₃ ⇒ β′₃ transformation. Furthermore, linear relationships between M _s and Al and Ni concentrations have been obtained for all types of martensitic phases.