Single Frame Resection of Compact Digital Cameras for UAV Imagery

Abstract

Recently, UAVs (Unmanned Aerial Vehicles) gain a wider acceptance from many disciplines. One major application is for monitoring and mapping. Flying beyond eye sight autonomously and collecting data over large areas are their obvious advantages. To support a large scale urban city mapping, we have developed a UAV system which can carry a compact digital camera as well as a navigational grade of a Global Positioning System (GPS) board mounted on the vehicle. Unfortunately, such a navigational system fails to provide sufficient accuracy required to process images become a large scale map. Ubiquitous digital compact cameras, despite their low cost benefits, are widely known to suffer instabilities in their internal lenses and electronics imaging system. Hence these cameras are less suitable for mapping related purposes. However, this paper presents a photogrammetric technique to precisely determine intrinsic and extrinsic camera parameters of photographed images provided that sufficient numbers of surveyed control points are available. A rigorous Mathematical model is derived to compute each image position with respect to the imaging coordinate system as well as a location of the principal point of an image sensor and the focal length of the camera. An iterative Gaussian-Newton least squares adjustment method is utilized to compute those parameters. Finally, surveyed data are processed and elaborated to justify the mathematical models.

A Relative Rotation between Two Overlapping UAV's Images

“…The projective center coordinates of each image in the Cartesian coordinate system are then utilized to calculate the baseline vector components between two images as in (2). If the GCPs are surveyed using geodetic type of GPS, the obtained geographical coordinates can be verified using space resection methods [27][28][29]. The resection method needs at least three GCPs appeared on both images.…”

Uji Lendut Deformasi Jembatan Dengan Teknik Pemotretan Konvergen Menggunakan Teknologi Drone

“…Output tersebut diperoleh dari hasil pengukuran koordinat foto (xy) dengan metode centroid pada masing-masing retro untuk setiap foto (Tjahjadi, 2010). Titik pusat tengah retro pada foto yang diamati merupakan lokasi jatuhnya berkas sinar pada bidang sensor yang dipantulkan dari objek dan memenuhi kondisi kolinieritas, di mana pusat perspektif (O), titik P pada bidang sensor, dan objek P yang dicitrakan semuanya berada pada satu garis lurus (Tjahjadi, 2017), kondisi kolinieritas ini dimodelkan secara matematis di dalam (Luhmann, 2011;Wolf, 2000;.…”

Assessing a 35mm Fixed-Lens Sony Alpha-5100 Intrinsic Parameters Prior to, During, and Post UAV Flight Mission

“…The self-calibration method is performed by means of the bundle adjustment [6] which provides a simultaneous determination of the interior and exterior orientation parameters, as well as the object point coordinates performed with or without a provision of known coordinates of control points [11], [13], [15], [21]-- [23]. However, while object space control points of known 3D coordinates are available, calibration parameters can be recovered using space resection methods [24]-- [26], relative orientation methods [27], [28], or bundle adjustment with fixed 3D coordinates. Before elaborating the result, a brief discussion about the self-calibration method is following.…”