Abstract:Compased with vision,robotic tactile image are usually of low resolution,noisy and incomplete,so theclassification accuracy is often not satisfied when using conventional classifiers.In this paper,we present aclassifier based on the hopfield associative memory(HAM),and applied it successfully to the recognition oftactile images.The comparison between the associative memory classifier and three conventional classifiers isshown in the exnerimente.
[1] Luo. R R, Joh H H. Tactile array sensor for object identification using complex moments. Proc. IEEE Int Conf Robotics and Automation,Washington D C. 1987:1935-1940
[2] Mostafa Y S A. Psaltis D. Recognitions aspects of moment invariants. IEEE Trans PA,MI, 1984; 6(11):698-706
[3] Hopfield J J. Neural network an physical systems with emergent collective computational abilities. Proc. National Academy of Sciences. USA. 1982: 2554-2558
[4] Farsell J A. Michel A N. A synthesis procedure for Hopfield's continuoues-time AM. IEEE Trans CAS. 1990,37,877-884
[5] 徐永惠.触觉传感器与触觉信息处理.中科院自动化所硕士论文,北京,1991