This paper presents a genetic algorithm for bundle adjustment in aerial panoramic stitching. Compared with the conventional LM (Levenberg-Marquardt) algorithm for bundle adjustment, the proposed bundle adjustment combining the genetic algorithm optimization eliminates the possibility of sticking into the local minimum, and not requires the initial estimation of desired parameters, naturally avoiding the associated steps, that includes the normalization of matches, the computation of homography transformation, the calculations of rotation transformation and the focal length. Since the proposed bundle adjustment is composed of the directional vectors of matches, taking the advantages of genetic algorithm (GA), the Jacobian matrix and the normalization of residual error are not involved in the searching process. The experiment verifies that the proposed bundle adjustment based on the genetic algorithm can yield the global solution even in the unstable aerial imaging condition.
For high resolution satellite remote sensing cameras, the line of sight (LOS) moving during the image exposure period
will cause the modulation transfer function (MTF) degradation and image blurring. Image stabilization component is
used to improve image quality by actively removing the apparent motion induced by vibration, tracking error and attitude
instability. In this paper, the image stabilization component is considered as a kind of closed loop servo control system,
and the image stabilization effect is converted into servo control performance for research. Firstly, the image
stabilization servo loop scheme and transfer function model are constructed and the LOS jitter is considered as the output
of a stochastic system derived by white-Gaussian noise. Based on the proposed model, the demand boundary of jitter
rejection function is described, and the design criterion to be satisfied is obtained according to the requirement of image
stabilization performance. And then, a discrete Kalman estimation algorithm is introduced into image stabilization servo
loop to filter out the noise caused by pixel-shift sensor (PSS) and compensate for the delay due to the PSS measurement.
Based on the given design criterion, the control law is designed by using the output of Kalman filter. The computer
simulation is achieved to show that the proposed control strategy can significantly improve the image stabilization
Inertially stabilized platform (ISP) is indispensable for various imaging systems to segregate the base angular
movement and achieve high LOS (Line-Of-Sight) stability. The disturbance rejection ratio and command following
performance are of primary concern in designing ISP control systems. In this paper, the redundant gimbals ISP system is considered and it is shown to experience complex disturbance and parameter variation during operation. To meet advanced LOS stabilization requirement, a disturbance observer based (DOB) dual-loop controller design for ISP is proposed of which the DOB is the internal-loop. Using a nominal plant model and a low-pass filter, the disturbance signal is estimated and used as a cancellation input added to the current command of torque motor. If the DOB works well, the disturbance torque and mismatch between nominal plant and actual plant will be compensated and the internal-loop will behave as nominal model parameters. On the other hand, the external-loop will be designed for nominal model parameters to meet stabilization requirements. This paper will mainly focus on the DOB design method. Since the low-pass filter of DOB determines the sensitivity and complementary sensitivity function as will be shown in this paper, designing the filter is the most important consideration. In this paper, an optimal low-pass filter design method is proposed. The method is intuitive, simple to implement and allows on-line tuning. Simulation results show the performance enhancement of our control structure in the presence of disturbance and measurement noise.