基于Surf算法的穿鞋足迹特征匹配识别OA

Against the backdrop of the rapid development of big data and artificial intelligence,the field of footprint examination urgently requires advancements in professionalization and informatization.Currently,research on the identification methods of barefoot footprints has made certain progress,but the automatic recognition of shod footprints remains a tough challenge in the field.Therefore,this study introduces the Surf algorithm to explore its potential application in the automatic recognition of shod footprints.More specifically,the study employs the Surf algorithm to perform feature matching on four types of footprint images(i.e.,the same person with the same shoe,the same person with different shoes,different people with the same type of shoes,and different people with different shoes),and conducts an in-depth analysis of the similarity between footprints by mapping the matching points through geometric transformations.The results show that footprints from the same person with the same shoe exhibit a significantly greater number of Surf feature point matches compared to other types;after geometric transformation mapping,footprint images from the same person,the same shoe and same time have a greater number of matching points,with accurate matching positions,while footprints from different people exhibit fewer matches.Additionally,even when considering footprints from the same person,those formed in closer time intervals have more matching points than those formed further apart in time.In summary,the Surf algorithm exhibits efficiency and reliability in recognizing footprints from the same person with the same shoe.