Object tracking using particle filter in the wavelet subspace

Rui, Ting; Zhang, Qi; Zhou, You; Xing, Jianchun

doi:10.1016/j.neucom.2012.03.036

Cited by 20 publications

(8 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, in contrast with EnKF, PF can perfectly accommodate the propagation of nonGaussian distributions through nonlinear models. This approach has been successfully applied to studies on target tracking [27], hydrological parameter estimation and uncertainty analysis [28]- [32], and land surface processing simulations [33]- [36]. Accordingly, the PF-based strategy appears to be a better choice for complicated crop growth models, i.e., the Crop Environment Resource Synthesis (CERES)-Wheat model.…”

Section: Introductionmentioning

confidence: 99%

Application of Crop Model Data Assimilation With a Particle Filter for Estimating Regional Winter Wheat Yields

Jiang

Chen

Jin

et al. 2014

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

To improve the performance of crop models for regional crop yield estimates, a particle filter (PF) was introduced to develop a data assimilation strategy using the Crop Environment Resource Synthesis (CERES)-Wheat model. Two experiments involving winter wheat yield estimations were conducted at a field plot and on a regional scale to test the feasibility of the PF-based data assimilation strategy and to analyze the effects of the PF parameters and spatiotemporal scales of assimilating observations on the performance of the crop model data assimilation. The significant improvements in the yield estimation suggest that PF-based crop model data assimilation is feasible. Winter wheat yields from the field plots were forecasted with a determination coefficient ( ) of 0.87, a root-mean-square error (RMSE) of 251 kg/ha, and a relative error (RE) of 2.95%. An acceptable yield at the county scale was estimated with a of 0.998, a RMSE of 9734 t, and a RE of 4.29%. The optimal yield estimates may be highly dependent on the reasonable spatiotemporal resolution of assimilating observations. A configuration using a particle size of 50, LAI maps with a moderate spatial resolution (e.g., 1 km), and an assimilation interval of 20 d results in a reasonable tradeoff between accuracy and effectiveness in regional applications.

show abstract

Section: Introductionmentioning

confidence: 99%

Application of Crop Model Data Assimilation With a Particle Filter for Estimating Regional Winter Wheat Yields

Jiang

Chen

Jin

et al. 2014

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

show abstract

“…Assuming the target region moving with constant velocity and scale change, Nummiaro et al [11] use a first-order dynamic model to propagate the sample set. These kinds of dynamic propagate modes are very popular now, some with slight modification (e.g., using standard second-order autoregressive process to replace the first-order model) [13,[23][24][25]. Sullivan and Rittscher [26] propose a deterministic search method to guide the random samples.…”

Section: Related Workmentioning

confidence: 99%

Visual Tracking Using an Insect Vision Embedded Particle Filter

Guo

Zhao

2015

Mathematical Problems in Engineering

View full text Add to dashboard Cite

Particle filtering (PF) based object tracking algorithms have drawn great attention from lots of scholars. The core of PF is to predict the possible location of the target via the state transition model. One commonly adopted approach is resorting to prior motion cues under the smooth motion assumption, which performs well when the target moves with a relatively stable velocity. However, it would possibly fail if the target is undergoing abrupt motion. To address this problem, inspired by insect vision, we propose a simple yet effective visual tracking framework based on PF. Utilizing the neuronal computational model of the insect vision, we estimate the motion of the target in a novel way so as to refine the position state of propagated particles using more accurate transition mode. Furthermore, we design a novel sample optimization framework where local and global search strategies are jointly used. In addition, we propose a new method to monitor long duration severe occlusion and we could recover the target. Experiments on publicly available benchmark video sequences demonstrate that the proposed tracking algorithm outperforms the state-of-the art methods in challenging scenarios, especially for tracking target which is undergoing abrupt motion or fast movement.

show abstract

“…Wavelet theory is an analytical method used in many scientific fields [10][11][12][13][14][15][16][17][18]. A. Raheja introduced a wavelet based multiresolution algorithm by extending the concept of switching resolutions in both image and data spaces [10].…”

Section: Identify Noise Variance Using Waveletmentioning

confidence: 99%