Bundle Block Adjustment of Cbers 2b Hrc Images Using Control Lines

China-Brazil Earth Resources Satellite (CBERS) images present strategic importance for Brazil, providing support in several applications such as deforestation management and fire control, mainly in Amazon region. CBERS 2B carries three imagery sensors: High Resolution Camera (HRC), CCD Camera and Wide Field Imager (WFI), which provide images with GSD of 2.5 meters (m), 20 m and 260 m, respectively. One problem with these images is the accuracy of their georreferencing, being necessary to correct them with ground control information. Although points are generally used as control elements in spatio-triangulation, some advantages motivate the use of linear features in Photogrammetry, including the plenty of this feature in man-made environments and the improvement of the robustness and geometric strength in bundle block adjustment. The aim of this work is to experimentally assess a block adjustment method based on linear features with CBERS 2B HRC images. The model being used states the coplanarity condition between the projection ray in the image and the projection plane in the object space. This model was implemented in the TMS (Triangulation with multi sensors) software that uses multifeatures control (points and lines). Experiments using a block composed by four CBERS 2B HRC images from two adjacent orbits were carried out and the results showed that the model based on straight lines works successfully in spatio-triangulation with CBERS 2B HRC images providing better results when compared to those obtained by triangulation with collinearity model, based on points.