The Improved SVM Multi Objects’s Identification For the Uncalibrated Visual Servoing

For the assembly of multi micro objects in micromanipulation, the first task is to identify multi microparts. We present an improved support vector machine algorithm, which employs invariant moments based edgeextraction to obtain feature attribute and then presents a heuristic attribute reduction algorithm based on roughset’s discernibility matrix to obtain attribute reduction, with using support vector machine to identify andclassify the targets. The visual servoing is the second task. For avoiding the complicated calibration of intrinsicparameter of camera, We apply an improved broydenʹs method to estimate the image jacobian matrix online, which employs chebyshev polynomial to construct a cost function to approximate the optimization value, obtaining a fast convergence for online estimation. Last, a two DOF visual controller based fuzzy adaptive PDcontrol law for micro-manipulation is presented. The experiments of micro-assembly of micro parts in microscopesconfirm that the proposed methods are effective and feasible