Classification and identification of body type in non-contacted body measurement system
Hao Liu · Tianjin Gongye Daxue xuebao · 2007
The size in length and width can be acquired directly from the front and side images in non-contacted body measurement system applying the technology of photography measurement, but the girth size can be acquired only by transforming from the 2D data. The calculating accuracy can be upgraded by using the fuzzy clustering function to classify the body types and using kernel-Fisher discriminant analysis to identify body types automatically. The result shows that calculating accuracy can be improved by using this classification.