International Journal of Information Technology
Vol. 11 No. 7 2005 (Special Issue)
Guest Editorial
Special Issue on Vision Information Processing & its Application
Research in the field of intelligent computing has been remarkably growing in the information era, which aims to bring intelligence, reasoning, perception, information gathering and analysis to computer systems. Advanced Informatics is seen to include a range of techniques such as artificial intelligence, evolutionary and adaptive computing, and so on. The International Conference on Intelligent Computing (ICIC’2005) is the most comprehensive conference focused on the various aspects of advances in the intelligent computing fields.
The 15 papers in this special issue are selected from the ICIC’2005. The three parts, which are mainly divided, show research results in the field of vision information processing and its application. The first part of this special issue is related to human visual system. The second part covers language process such as speech recognition, word recognition, intelligent query system and the science tutoring system. Part 3 brings a number of papers that focus on the relation to image processing such as image segmentation, pattern recognition, image fusion and feature extraction.
Zheng Chang et al. propose an efficient motion-compensated 3D wavelet video coding system. The research results show that it can get better performance for low-motion video sequences, and it can not only improve the compression efficiency, but also reduce memory requirement and computational complexity for high-motion video sequences.
A novel rate-distortion optimization based on structural similarity in color image encoder is the focal point of the study by Zhiyi Mai et al.
Next, Bing Han et al. present a feature reduction method based on Rough-Fuzzy Sets for news video. The shot boundary detection method can decrease data dimensions and preserve the information of original video farthest.
Jian Yang et al. have been described a new non-native accents speech corpus of dictation Mandarin for Large Vocabulary Continuous Speech Recognition, which is contributed to recognize the non-native speech.
A new approach of applying high-order language models into speech recognition process is proposed by Ying Liu et al., which enhanced the system performance without apparent hazard on the system efficiency.
The study by Hyung Joon Kook develops a tutoring architecture for geometry where the domain principles are automatically converted to inference operators for use by the domain-independent, inferential portion of the tutoring system, which automatically generates the solutions to problems at tutoring time based on the inference operators.
Next, Hongyu Yang et al. utilize feature structure to describe the structure of vocabulary, phrase and sentence. Functional Unification Grammar (FUG) in Natural Language Process (NLP) was applied to intelligent query in Intrusion Detection System (IDS) audit database.
Yuguo Ding et al. cast the topic of isolated word recognition system. Two types of confidence measures are introduced: online garbage model based likelihood ratio and semi-syllable based posterior probability.
Yingchun Li et al. focus on the extraction particles such as long casts in microscopic images. Gabor-based combining iterative method has been presented, which is a robust method using automatic segmentation.
Ding Feng et al. exploit a diagnosis expert system, which can automatically analyze many current recording charts and identifying work condition of an electrical submersible pumping. The method of the recording chart feature pick-up is adopted in the pattern identification or recognition.
The study on the image fusion by Heng Ma et al. clarify the concepts and relationship between the fusion rules and fusion operators based on wavelet transform, and propose a new spatial frequency computing method corresponding to wavelet decomposition characteristic.
Another image fusion approach based on segmentation region is proposed by Rong Wang et al., which combines effectively the pixel-level and feature-level fusion to get a better fusion image.
In the next paper, Ying Li et al. introduce a novel method for automatic segmentation of SAR images using wavelet packet, which is based texture feature set and clustering algorithm.
Fuyuan Peng et al. present a fuzzy pattern recognition approach for undersea image based on fractal features, and realize reliable target classification in undersea image.
Finally, Guosheng Wang et al. investigate the design of reconfiguring a class of linear control systems via state feedback eigenstructure assignment.
The papers in this special issue give some glimpse into the intelligent computing research fields at ICIC 2005. We hope you enjoy this special issue and it is helpful for you.
Lili Yun, Guest Editor
Kumamoto University, Japan