Adaptive mode decision for H.264 encoder

An adaptive mode decision algorithm is presented, with rate-distortion optimisation that reduces the complexity of the H.264 encoder without loss of image quality and compression ratio. The proposed algorithm uses the property of an all-zero coefficients block that is produced by quantisation and coefficient thresholding to effectively skip unnecessary modes. Experimental results show that the speed of the adaptive mode decision algorithm is two times faster than the full-mode decision algorithm of the JM72 reference encoder, without any coding loss.     

Extended adaptive RBF equaliser for overcoming co-channelinterference

The authors investigate the problem of nonlinear adaptive equalisation in the presence of intersymbol interference, additive white Gaussian noise and co-channel interference. An extended radial basis function (RBF) network is proposed, in which regression weights are used in the output layer and the hidden unit is defined to have the Gaussian formula with the Mahalanobis distance. It is shown by simulation that the proposed structure gives reduced computational complexity without performance degradation, compared to that of the conventional RBF equaliser     

Lossless subband coding system based on rounding transform

We propose a new rounding transform called the overlapping rounding transform (ORT). It is defined as a two-port input/two-port output FIR filtering system with a pair of rounding operations. The ORT is applied to develop lossless subband coding systems. The ORT approach has both a simpler representation and more possibilities for lossless subband implementation than the lifting scheme     

In vivo Doppler shift measurements using multimode fiber-opticcatheters

A new fiber-optic catheter for in vivo blood-flow measurements has been developed. The catheter is designed to measure blood flow in both the forward (toward the catheter tip) and reverse (away from the catheter tip) flow directions. It consists of two multimode optical fibers with core diameter of 50 μm and cladding diameter of 125 μm. One fiber transmits the laser beam into blood and the other receives the backscattered light from the erythrocytes within the probe volume. In the flow experiment, it was found that the flow within the boundary layer is indeed laminar and, hence, the relationship between the Doppler shift frequencies and the flow velocities is linear, thereby making the linear calibration possible for predicting the free stream flow velocity. Plots of the maximum shift frequency (frequency at which the Doppler spectrum disappeared into the noise spectrum) against the flow velocities are found to be more linear in both the forward and reverse flow directions than that of the dominant shift frequency (frequency with the highest amplitude). These results were reaffirmed by the numerical flow simulation along the catheter side wall     

An 8-bit-resolution, 360-μs write time nonvolatile analog memorybased on differentially balanced constant-tunneling-current scheme(DBCS)

This paper describes a fast and accurate nonvolatile analog memory (NVAM) and its programming scheme. Both constant programming rate and single-pulse programmability have been achieved, which drastically enhance programming speed and accuracy. A prototype chip containing 8×128 NVAM cells (cell size of 9×13.6 μm²) has been fabricated using 0.8-μm CMOS. Each cell is measured to store more than eight bit levels within 360 μs     

Improvement of linewidth enhancement factor in 1.55-μmmultiple-quantum-well laser diodes

The improvement of the linewidth enhancement factor in complex-coupled laser diode (CC-LD), or loss-coupled, was confirmed by measuring the spontaneous emission spectra below threshold from the sidewall of laser diodes. In addition, the serial resistance of the device was measured. The linewidth enhancement factor is improved by the presence of a light absorbing InGaAs grating for loss coupled distributed-feedback (DFB) laser diode (LD). We report the comparison of the linewidth enhancement factors of Fabry-Perot (FP) LD, conventional DFB-LD, and loss coupled DFB-LDs     

Fast DCT algorithm with fewer multiplication stages

A novel fast DCT scheme with reduced multiplication stages and fewer additions and multiplications is proposed. The proposed algorithm is structured so that most multiplications tend to be performed at the final stage, which reduces the propagation error that could occur in the fixed-point computation. Minimisation of the multiplication stages can further decrease the error     

Motion based object tracking with mobile camera

An efficient algorithm which detects and tracks a moving object is proposed. The global motion energy caused by camera movement is eliminated by finding the maximal matching position with the cross-correlation value between two consecutive frames using the spiral scanning technique. The region of a moving object is segmented exactly by combining the results of the temporal derivatives and edge map from one image, and the actual centroid of the moving object is determined as the mid-point of the cumulative distribution of the projection profile of the region