An automated model-image fitting algorithm is proposed in this paper for generating façade texture image from pictures taken by smartphones or tablet PCs. The façade texture generation requires tremendous labour work and thus, has been the bottleneck of 3D photo-realistic city modelling. With advanced developments of the micro electro mechanical system (MEMS), camera, global positioning system (GPS), and gyroscope (G-sensors) can all be integrated into a smartphone or a table PC. These sensors bring the possibility of direct-georeferencing for the pictures taken by smartphones or tablet PCs. Since the accuracy of these sensors cannot compared to the surveying instruments, the image position and orientation derived from these sensors are not capable of photogrammetric measurements. This paper adopted the least-squares model-image fitting (LSMIF) algorithm to iteratively improve the image's exterior orientation. The image position from GPS and the image orientation from gyroscope are treated as the initial values. By fitting the projection of the wireframe model to the extracted edge pixels on image, the image exterior orientation elements are solved when the optimal fitting achieved. With the exact exterior orientation elements, the wireframe model of the building can be correctly projected on the image and, therefore, the façade texture image can be extracted from the picture.
|頁（從 - 到）||351-354|
|期刊||International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives|
|出版狀態||已發佈 - 2012|
|事件||22nd Congress of the International Society for Photogrammetry and Remote Sensing, ISPRS 2012 - Melbourne, 澳大利亚|
持續時間: 2012 8月 25 → 2012 9月 1
ASJC Scopus subject areas