Key Zone Auto Location of Plantar Pressure Image Using De-Dimensionality and Wavelet Transformation

摘要

The research on the comfort of shoes is a hot-issue in the field of footwear optimization design and intelligent design. To acquire the data-set of plantar pressure imaging, the foot-scan plantar pressure imaging experimental equipment was used to the plantar pressure imaging by collecting the actual test images of plantar pressure. The basic data-set of plantar pressure image is constructed by means of denoising and dimension reduction, which provides the basis for the segmentation of key feature areas and the extraction of key features of plantar pressure. The proposed methods of image de-dimensionality are used in preprocessing of image data-set. And some image filtering technologies are also applied in suppressing the noise of the target image, and subsequently to keep the original image feature details. The effect of some kinds of filtering was directly affect the subsequent image processing and the effectiveness of image fusion analysis by using Gaussian filtering, mean filtering and bilateral filtering which will guide the last surface generation in better ways.