Acta Astronautica · 2021 · 14 citations · 27 references
ComputingAerospace EngineeringAstronomyComputer VisionMachine VisionSensorsMotion AnalysisMotion EstimationCentroid ComputationCentroid EstimationCentroiding AlgorithmsEnergyEnergy Change AnalysisGaussian Fitting CentroidingSystematic Error Mitigation
In the fields of aerospace, astronomy and biology, the observation and tracking of weak point targets is a common problem. High-accuracy centroid position is the basic and crucial information for subsequent measurement and analysis. A centroid determination method for moving dim point targets is proposed in this work. Based on the target intensity distribution and the motion model, it transforms the energy variation sensed by pixels into the displacement of the spot centroid. Time domain is taken into consideration to reduce the effect of noise based on Kalman filtering. Compared with the traditional centroiding method, the proposed method overcomes the systematic error caused by energy integral of detector and energy truncation of calculation window. It is also not as sensitive to noise as the Gaussian fitting method. Experimental results show that the accuracy of the method can reach within 0.008 pixels. The method can even work with an extremely small amount of information as long as the energy change of only one pixel is detected. The research may provide a good solution for centroid measurement of point targets in different application fields.
27
Imaging Intracellular Fluorescent Proteins at Nanometer Resolution
Eric Betzig, George H. Patterson, Rachid Sougrat et al. · Science · 2006 · 8.9K citations
Robust single-particle tracking in live-cell time-lapse sequences
Khuloud Jaqaman, Dinah Loerke, Marcel Mettlen et al. · Nature Methods · 2008 · 2.1K citations · Full text