In order to realize less time consuming and on-line image classification for steel strip surface defects, an improved multiclass support vector machine (SVM) was proposed. The SVM used a novel algorithm and only constructed (k-1) two-class SVMs where K is the number of classes. In the testing phase, to identify the surface defects it used a new unidirectional acyclic graph which had internal (k-1) nodes and k leaves. Its testing time is less than traditional multiclass SVM method. The experiment results shows that this method is simple and less time consuming while preserving generalization ability and recognition accuracy toward steel strip surface defects.

In response to the dilemma for image identification by the existing classifier toward surface defects of steel ball, an improved support vector machine (SVM) for multiclass problems is proposed. Minimum distance method is presented to resolve the unclassifiable region of the multiclass SVMs. The 16 image features of the surface defects are selected as input vector of the SVMs. The experiment results show that more accurate identification toward surface defects of steel ball was achieved by the improved multiclass SVM and the accuracy can reach 95%.

This paper proposed a defects extraction method based on one-dimensional adaptive Gaussian filtering. To overcome the difficulties as various defects, low contrast, complex background and overall uneven brightness, this method designed a new one-dimensional adaptive Gaussian filter based on the defect size. According to the gray change of the magnetic tiles surface image in the different regions, we progressed the Gaussian filtering based on the selective one-dimensional scan line by region, and made the width of the filter can automatically adjust with the defect size, simulated the threshold value curve, and extracted the magnetic tiles surface defects. The experiment shows that this method can accurately and quickly extract the various types of defects of the magnetic tile surface.

The Liquid Surface Pressure Control is the key factor for the guarantee of Low Pressure Die Casting Quality. Regarding to the disadvantages of conventional PID Control such as pressure fluctuation, poor repeatability of the pressure curve, and so on, we propose Liquid Surface Pressure Control System (LSPCS) based on Fuzzy Adaptive PID. Design method of Fuzzy PID Controller has been discussed, and the realization methods of the hardware and software in this system are developed. This proposed system has a good performance in practice.