Search

A method for inspecting solder joints using support vector machines (SVMs) and a tiered circular illumination technique is proposed. The illumination technique provides visual information that allows the 3D shape of the solder joint surface to be determined. The extracted features are used to classify the solder joint using an SVM classifier. Experimental results show the effectiveness of the proposed method.

Kernel principal component analysis (PCA) is presented as a mechanism for extracting textural information. Using the polynomial kernel, higher order correlations of input pixels can be easily used as features for classification. As a result, supervised texture classification can be performed using a neural network.

A genetic algorithm-based segmentation method for video sequences is proposed. The computation is carried out by a number of chromosomes that evolve from distributed genetic algorithms. To eliminate redundant computations, only unstable chromosomes, corresponding to moving object parts, are evolved. Experimental results show the effectiveness of the method.

An approach to the problem of supervised texture segmentation using nonlinear support vector machines (SVMs) is presented. For each texture class a nonlinear SVM is constructed which separates that class from the other classes. The segmentation then works by applying all the SVMs to an input image and arbitrating between the SVM outputs. Experimental results show the effectiveness of the proposed method.