Learn More
In this study, we investigated the changes in topological architectures of brain functional networks in attention-deficit/hyperactivity disorder (ADHD). Functional magnetic resonance images (fMRI) were obtained from 19 children with ADHD and 20 healthy controls during resting state. Brain functional networks were constructed by thresholding the correlation(More)
We explore an approach to speaker identification called speaker clustering in the GMM-based speaker recognition system in order to reduce the computational complexity. The ISODATA algorithm adapted for our purpose works well when we cluster speakers whose acoustic characteristics are similar to a distance measure. The time spent on HSI (hierarchical speaker(More)
This paper presents a fast motion compensated de-interlacing algorithm based on true sub-pixel accurate motion estimation scheme. This algorithm aims at two existing motion compensated de-interlacing problems: one is the computational complexity of motion compensated de-interlacing method and the other is robust to erroneous motion vectors. A new motion(More)
This paper presents an efficient AVS-P2 encoding scheme which is based on motion compensation de-interlacing information. This scheme uses the similarity between motion compensation of encoding and motion compensation of de-interlacing to reduce the overall computational complexity of encoding system. A system analysis is performed to study this similarity.(More)
We put forward a new algorithm for speaker identification. The difficulties for speaker recognition were first analyzed. Because most methods for speaker identification are based on parameter estimation, we put forward a nonparameter method for speaker identification. The method is based on Fisher differentiation vector. The influences of different factors(More)
  • 1