Dilmurat Tursun

Research on on-line Uyghur character recognition technology based on center distance feature

In this paper, the center distance feature (CDF) is presented as an efficient approach for on-line Uyghur handwritten character recognition. Based on early research for on-line Uyghur handwritten character recognition, a further research is conducted with center distance feature, abbreviated as CDF. This paper introduces the extraction of center distance feature and its three different methods such as CDF-2, CDF-4 and CDF-8 which have improved the average recognition accuracy respectively to 78.17%, 90.47% and 94.50% for the 32 isolated forms of Uyghur characters. 12800 samples from 400 different writers are participated into experiments. The system is trained using 70 percent of total samples and tested on the remained 30 percent.

A Feature Selection and Extraction Method for Uyghur Handwriting-Based Writer identification

This paper proposes a method for texture feature extraction by integrating Gabor filters and independent component analysis (ICA) for Uyghur handwriting based writer identification. That is, the texture image is firstly filtered by a given bank of Gabor filters, and then higher dimensional feature vectors are constructed from the filtered texture images. Next, the dimensionality of these vectors is reduced by means of principal component analysis (PCA). Finally, the independent components in the resulting vectors with dimensionality reduced are analyzed and extracted by us. Experiments were performed using KNN-5 classifier to Uyghur handwriting samples from 55 different people and promising results of 92.5% correct identification rate were achieved.

A Particle Filter and Fuzzy Clustering Based Algorithm for Tracking Dim Moving Multiple Point Targets in IR Image Sequence

The tracking algorithm described in this paper adopts the fuzzy clustering method to fuse the single frame detection measurements from different tracking windows, then the fused measurements will be inputted to the Particle filters, via the recursive prediction and updating, getting the moving state of each target at each discrete time. The theoretical and experimental results show that the algorithm presented in this paper can keep dynamic stability and robustness while providing good real-time tracking performance.

MCDF based on-line handwritten character recognition for total Uyghur character forms

This paper proposed the Modified Center Distance Feature (MCDF) and its different forms for Uyghur handwritten character recognition. By combination with some low dimensional features, MCDF gifted remarkable recognition accuracy of 87.6% for total Uyghur character forms. This result is higher than previous record by more than 11 points. Samples from 400 volunteers are used in experiments.

Gaussian Particle Filter Based Algorithm for Tracking of a Dim Moving Point Target in IR Image Sequences

In order to solve the problems of deviations with large error occurred at the sharp corners of motion trajectories and poor stability of tracking algorithm presented in previous paper, and further make improvements to the tracking precision, in this paper the rather appropriate state model is established first of all, and then the effective observations are collected by using spatial-temporal detection and fusion in which the brightness information of the target included in the decision criteria for the first time. Under the different circumstances of Gaussian noise and non-Gaussian noise, the experimental results show that take the Gaussian particle filter as the tracking algorithm which has no use of re-sampling could have effectively improved the precision of algorithm for tracking dim moving point target in IR image sequences, and has a good real-time performances and good stability.

Research on Prosodic Features and Their Prediction Issues in Uyghur Text-to-Speech System

As one of the core technologies of minority language information processing, in recent years, the Uyghur speech synthesis technology has made great progress, but in TTS (Text To Speech) systems, prosodic phrases are not predicted with high accuracy which slows down the improvement of naturalness of synthesized speech. In this paper, Uyghur prosodic features was studied and the Context features which affects the Uyghur Prosodic phrases was analyzed on the basis of Uyghur prosodic features and large Uyghur speech Corpus. The context parameters that have an important impact on Uyghur prosodic features were collected and question sets well fits the training of Uyghur prosodic prediction was designed, and the training data was prepared. Consequently, the Uyghur prosodic training was processed and how to further apply it in the Uyghur speech synthesis system to attain higher naturalness was discussed.

Experimental Phonetics B ased Analysis of Semivowels in Uyghur Language

According to the research needs in speech synthesis and speech recognition applications, this paper analysis the acoustic features of semivowel in Uyghur language from the view of experimental phonetics that some acoustic parameters such as formant, pitch, durations were measured, and finally verified the fact by means of experimental phonetics analysis method that the semivowels have the typical phonetic features of vowel’s in Uyghur language.