An Efficient k-Means Clustering Algorithm: Analysis and Implementation
IEEE Transactions on Pattern Analysis and Machine Intelligence
K-means clustering via principal component analysis
ICML '04 Proceedings of the twenty-first international conference on Machine learning
Hi-index | 0.00 |
K-means clustering algorithm is applied to process the colorful frames of UAV reconnaissance videos. After conversing the RGB color space into L*a*b* color space for reducing the dimensions, the appointed distinct segment can be partitioned from the colorful image with K-means clustering algorithm. This method can avoid the questions of poor precision to image segments with the algorithms of background subtraction, optical flows and feature matching about the complicated locomotive and shaking backgrounds. The experimental result shows that the adoption of the color-based segmentation algorithm can really produce high image segmentation precision, in UAV's civil actual application. At last, a GUI is designed to carry out the image segmentation and video processing expediently.