For the problems of low motion estimation accuracy and affecting easily by various interferences an improved optical flow algorithm with five-confidence-point restraint based on gradient thresholds was put forward in this paper.The five-confidence-point weighted method was applied to calculate the time and space gradient values of each image pixel and to reduce the effect of noise on the single pixel, and the gradient threshold was adopted to process the basic optical flow constraint equation.Therefore, the interference data which will disturb the motion estimation in the optical flow field were eliminated and the motion estimation accuracy of motion image was improved. The experiments were carried out by the traditional algorithm and the improved one, respectively for global and local movements. The contrastive analysis shows that the improved optical flow algorithm can figure both scene global motion and local motion in the scene and has advantages of high motion estimation accuracy and strong anti-interference, better robustness and convergence as compared with that traditional one.Furthermore,it can the shorten computation time by 1/3. It can be expected that this algorithm have potential application in driving environment sensing and traffic flow analysis.