We study geometrical properties between 2-D image plane and 3-D error space under affine reconstruction. The purpose of our system is to contribute to more accurate 3-D reconstruction by analyzing geometrically 2-D to 3-D relationship. In situation for no missing feature points and no noise in the 2-D observation matrix, the accurate solution is known to be provided by Singular Value Decomposition. However, several feature points of the matrix have not been observed because of occlusions and low image resolution, and so on. In this case, there is no simple solution. To obtain accurate 3-D reconstruction by recovering missing feature points, we propose the analytic approach which can handle the error orientation and distance of missing feature points by the geometrical properties.
The hardware complexity of digital filters is mainly dominated by the coefficient multipliers. Implementing fixed-point coefficient multiplication as a network of adders, subtractors, and shifters, yields lower power consumption. In such filters the number of adders (and subtractors) determines the implementation cost. The reason is that shifts are implemented as hard-wired inter-block connections and are considered “free”. In transposed implementation of an FIR filter, each input is multiplied by several coefficients. Considering all the coefficients as a multiplier block and omitting the redundancies by sharing the common fundamentals among different coefficients, yields great reduction in the number of arithmetic operations. This paper presents a graph based algorithm to reduce the computational complexity of multiple constant multiplications. Simulation results show that using the proposed method results good improvement in adder cost of multiplier blocks.
We fabricated an etched grating filter on a Ti-diffused waveguide in LiNbO3 using inductively coupled plasma etching with C4F8/Ar as an etching gas, which has an etching rate of 85.1nm/min. The etched grating filter had a reflectivity of 35% and a bandwidth of 0.02nm. Maximum reflectivity was obtained when the electric field of an incident beam was perpendicular to the grating.
A special load flow method is required to solve power balance equations of Radial Distribution Systems (RDS) owing to high R/X ratio of their lines. This paper reports an application of Artificial Neural Networks (ANNs) to determine bus voltages of a radial distribution system for any given load without executing the load flow algorithm. A multi-layer feed forward ANN with error back propagation learning algorithm is proposed for calculation of bus voltages and its angles. Extensive testing of the proposed ANN based approach indicates its viability for radial system load flow assessment. Test results are presented for a sample 33-bus system.
This work proposes a pulse waveform to reduce UWB interference at narrowband (NB) systems. The pulse waveform is represented as a generalized form of Gaussian pulse family which can be used in ultra wideband (UWB) impulse radio systems. It is obtained by combining two higher order derivatives of Gaussian pulses. The pulse waveform reduces UWB interference power at NB systems by introducing spectral notches at the center frequency of NB systems.