The Research on Judgment Method of Powder Mixing Uniformity Based on PSO-SVDD

The uniformity of powder mixing during the preparation process is an important factor affecting the quality of media. In order to judge the uniformity of powder mixture during media preparation, a particle swarm optimization (PSO) based support vector data description (SVDD) method is proposed for detecting the uniformity of powder mixture. Firstly, uniform powder images' feature parameters based on RGB and HSV color space are extracted to build training data. Then, PSO is used to optimize the kernel parameters σ and penalty factor C of the SVDD algorithm. The judgment model of powder mixture uniformity is established. Finally, Uneven powder images and uniformity of powder images are constructed as test samples to test the effect of the judgment model. The results show that the proposed method performs better in judging powder's uniformity and higher classification accuracy.