AN IMPROVED LIVE WIRE ALGORITHM FOR BAREFOOT HEEL IMPRESSION EXTRACTION

作者：Guo Li^*; Gao Liqun; Pian Zhaoyu

来源：International Journal of Innovative Computing Information and Control, 2009, 5(9): 2593-2602.

• Live wire
• Shortest path
• ACO algorithm
• Dijkstra

The distribution of barefoot impression reflects the behavioral and physiological feature of human body, and has extensive application in criminal investigations. Appropriate characteristics suitable for study can be retrieved from heel impression by scanning the bare footprint in human footprint recognition system. In this work, an improved live wire processing method has been proposed to extract the barefoot heel impression exactly. According to the characteristics of footprint image, a new cost function has been defined in the live wire method. In order to improve the efficiency and accuracy of the extraction, ant colony optimization (ACO) algorithm has been employed to replace the Dijkstra method to find the shortest path, and to obtain the effective edge. The least square method was used then for ellipse fitting to obtain the parameters. The experimental result shows that the improved live wire method is effective in extracting barefoot heel impression.

• 出版日期2009-9
• 单位东北大学