粒子成像测速图像的模糊聚类识别

聚类是流体运动的基本规律 ,聚类由几个相邻的粒子开始。随着流体的运动 ,粒子簇在第二次曝光时改变其形状及位置 ,通过模糊聚类可获得粒子运动的规律。应用聚类分析方法可改善非标识PIV图像的处理效率。将粒子运动轨迹追踪算法应用于PIV图像分析 ,取得了较好的效果

FUZZY CLUSTERING RECOGNITION IN PIV IMAGE ANALYSIS

The principle of fuzzy logic is applied to analyze the motion of the particles. The clustering procedure begins from several adjacent particles, and the clustering principles are based on the fluid mechanics. As the particles move, the topological structure of the particle changes in the next exposure, and tracing of the consequent clustering with the flow of particle in a whole field will be obtained. Fuzzy clustering method can be used to improve the processing efficiency of PIV measurements. A particle image tracking algorithm based on this method is developed for the PIV image analysis.