|Pattern Analysis and Machine Intelligence|
Our research strength in this area includes:
- Case-based reasoning systems
- Chaos systems
- Document analysis and recognition
- Fuzzy systems
- Genetic algorithms
- Intelligent hybrid systems
- Machine vision
- Neural networks
- Pattern recognition
- Telerobotics operation
- Traditional Chinese medicine informatics
Pattern recognition aims to classify or describe groups of measurements or observations. It is an important technology within the area of intelligent systems, often used in data preprocessing and decision making.
Pattern recognition is an innate ability that we all possess. As children we recognise the number of our classroom, using the inborn gift of number recognition. The ability of speech and character recognition allows us to understand what the teacher says in the class and writes on the chalkboard.
Interest in the research of pattern recognition applications has spawned in recent years. Popular areas include: data mining (identification of a 'pattern', i.e., a correlation, or an outlier in millions of multidimensional patterns), document classification (efficient search of text documents), financial forecasting, and biometrics (personal identification with physical or behavioral attributes).
Demand for automatic pattern recognition systems also surges, thanks to the popularity of larger databases and more stringent performance standard in speed, cost and accuracy. Recently, scientists have suggested (so-called) affective computing to give computers the ability to recognize and express emotions, to respond intelligently to human emotions and to develop models that explain the role of emotions in rational decision model.
The design of a pattern recognition system involves essentially the following aspects:
- Data acquisition and preprocessing;
- Data representation;
- Decision making.
The specific problem domain dictates the choice of sensor, preprocessing technique, representation scheme, and the decision making model. There are also a number of overlaps between pattern recognition techniques and such areas as adaptive systems and signal processing, artificial intelligence, O neural modeling, optimization/estimation theory, fuzzy sets, structural modeling, and formal languages.
Examples of pattern recognitions applications include: biometrics, computer vision, seismic analysis, character (letter or number) recognition, medical diagnosis.
To find out more, please contact Dr. James LIU
Telephone: 2766 7273