Title: Multi-cell multi-Bernoulli tracking method based on fractal measurement model
Authors: Jihong Zhu; Mingli Lu; Fei Wang
Addresses: School of Electrical and Automatic Engineering, Changshu Institute of Technology Changshu, China ' School of Electrical and Automatic Engineering, Changshu Institute of Technology Changshu, China ' School of Electrical and Automatic Engineering, Changshu Institute of Technology Changshu, China
Abstract: For detecting and tracking directly from image observations without the need for any separate target detection, a novel multi-cell tracking method based on multi-Bernoulli filter using fractal measurement model is proposed. The Hurst coefficient is one of most important parameters in application of fractal theory. It is used to construct fractal measurement model in this paper, which is estimated by the rescaled range analysis method. The fractal measurement model can offer two advantages for multi-Bernoulli filter. The one is that the measurement model is easy to establish, the other is that the computation of likelihood function is simple. Experiment results show that our proposed method could achieve an accurate and joint estimate of the number of cells and their individual states especially in the case of the number of cell population varying and the cellular morphology changing. Furthermore, it shows equivalent tracking accuracy against other tracking methods.
Keywords: multi-cell tracking; multi-Bernoulli filter; Hurst coefficient; rescaled range analysis; fractal measurement model.
DOI: 10.1504/IJMIC.2021.121838
International Journal of Modelling, Identification and Control, 2021 Vol.37 No.3/4, pp.240 - 248
Received: 15 Nov 2020
Accepted: 11 Feb 2021
Published online: 07 Apr 2022 *