To delete false intersection results produced by multiple trajectory cameras when they measure the multiple near ground coplanar targets with basic intersection method, a method for deleting false targets and improving the intersection accuracy with partial least squares multiple factors regression analysis is presented. Based on the distance between the multiple intersection results, the false targets can be recognized. The best estimation data of target on earth coordinates can be computed with the Partial Least Squares (PLS) multiple factors regression analysis method. Actual test results show that the method can recognize more than 20 coplanar targets which are near ground, and compared with two station intersection methods, PLS reduces measurement error more than 6″.