In order to reduce the overall weight of the visible-light detector and ensure its imaging quality,a lightweight method based on approximate model technology was constructed. The data was sampled by Latin hypercube experiment design method,moreover,in order to improve computing efficiency and obtain high calculation accuracy,response surface approximate models based on polynomial fitting and moving least square fitting are introduced into the lightweight design of visible-light detector,and a mathematical model was built for this structure. The sequential quadratic programming method was used to solve the optimization problem,thus avoiding the disadvantages of large calculated amount in the traditional optimization method and slow convergence or even no convergence in the optimization of nonlinear system. Finally,experiments were designed to verify the stiffness of the visible-light detector after lightweighting. The experimental results show that the lightweight design of visible-light detector meets the requirements of design indexes.