A general and efficient design approach using a radial basis function (RBF) neural classifier to cope with small training sets of high dimension, which is a problem frequently encountered in face recognition, is presented. In order to avoid overfitting and reduce the computational burden, face features are first extracted by the principal component analysis (PCA) method. Then, the resulting features are further processed by the Fisher's linear discriminant (FLD) technique to acquire lower-dimensional discriminant patterns. A novel paradigm is proposed whereby data information is encapsulated in determining the structure and initial parameters of the RBF neural classifier before learning takes place. A hybrid learning algorithm is used to train the RBF neural networks so that the dimension of the search space is drastically reduced in the gradient paradigm. Simulation results conducted on the ORL database show that the system achieves excellent performance both in terms of error rates of classification and learning efficiency.

{ ehltoh, exchen, efyang}@ntu.edu.sg ABSTRACT A simple and practical anti-aliasing method for a color straight line drawing is presented in this paper. The method has been applied in a DSP-based display system to remove the undesired jaggies occurred in the line drawing. The experimental results show that this method can produce a good visual effect on the low resolution display screen.

The dynamic behavior of human skin temperature has potential clinical diagnosis values. The method and material for acquiring and implementing thermal data plays an important role for obtaining a confirmed thermographic diagnosis. High resolution, both temperature and spatial, is required in order to see a dramatically different behavior between the data obtained in a group of patients and the one observed in a group of normal subjects.In this paper, a specific exercise is designed; we visualize the skin temperature before, during and after the exercise. Over 2000 frames during the period were captured, and the skin temperature of the regions of interest is analyzed statistically. In order to visualize the FFT to yield the relative contributions of its underlying frequency over the series of frames, the average temperature of the region of interest is obtained in order to average out some of the spurious noise. The subareas of interest are detected automatically using region-growing technique.The results are obtained using an infrared imaging system at a frame rate 30 frames per second.

No abstract