Footprint Gender Predicting Method based on Fusion of Ruler Information
This study proposes a foot gender identification method that incorporates ruler informa-tion.The proposed method first uses the Sobel operator to extract the positions of the ruler and foot-prints in the image.Then,based on the position information,it extracts global and local features of the footprints and ruler,and fuses them to obtain the combined features.Finally,the ratio informa-tion of the ruler is used as a metric feature to predict the gender of the footprint.The experimental re-sults on a standard dataset of footprint examination show that the proposed method achieves an accu-racy of 93.2％,which is higher than the baseline method by 2.5％.The results also demonstrate that incorporating ruler information is a key factor in improving the prediction accuracy.
万方数据
维普
