Because of the close range photogrammetry has wide measuring range, high precision and high efficiency, the precision measurement of large size tasks take more and more important role Among them, the self-calibration measurement model based on adjustment optimization is the important reason to ensure the method to achieve high-precision measurement. However, with commercial grade SLR camera more and more applied to three-dimensional measurement, the measurement accuracy and the professional camera compared to a certain gap A large number of analyses have found that, in addition to the camera itself, the self -calibration model relies too much on the internal parameters of the camera, especially the distortion parameter, which is the important reason leading to the decrease of the measurement accuracy. In order to reduce the influence of the parameterized model on the measurement results, we propose a photogrammetric method that does not rely on the intrinsic parameters of the camera. Firstly, a non-parameterized calibration method for large field of view camera is designed by combining the perpendicular method and Zeiss calibration method. Then, the non-parameterized measurement model based on the angle information can be established after the matching of the same point and the initial value of the difference between different images. Finally, combined with adjustment optimization algorithm, the three-dimensional coordinate of the measured point in space is calculated accurately. Compared with the traditional photogrammetry results, it is proved that this method can effectively improve the photogrammetric accuracy of the large field SLR camera.