Vision is a useful information for manipulation tasks since it provides a large spectrum of details about the environment. The visual feedback information can be used to guarantee the manipulation accuracy and improve robustness to uncertainty. In this paper, a vision-based adaptive control method is proposed for micromanipulation using optical tweezers. A general model of robotic manipulator is introduced in the optical tweezers system so that the movement of the laser beam is controlled by vision-based closed-loop robotic manipulation techniques. A new adaptive Jacobian controller is proposed for the regulation problem in presence of camera calibration errors. The stability of the closed-loop system is analyzed with consideration of the dynamics of both the particle and the robotic manipulator. Simulations are presented to illustrate the performance of the proposed method.