Large scale dynamic estimation of ocean surface temperature

Fieguth, Paul; Khellah, Fakhry; Murray, M. J.; Allen, Myles

doi:10.1109/igarss.1999.772108

Cited by 2 publications

(4 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Substituting back into (10) we find that, for a fixed dimension , the computational effort per-pixel goes as (12) That is, the complexity per pixel is strictly a function of correlation length, as is intuitive, and is not a function of , in sharp contrast to the original multiscale model (3). Although the effort (10) appears to ignore the fact that each tree has pixels on the finest scale which must be estimated, suggesting an effort more fairly represented as (13) almost all of the processing at the finest scales is the same for all models, as discussed below, so (10) is indeed a realistic reflection of the algorithm proposed in this paper. …”

Section: A Motivationmentioning

confidence: 93%

See 1 more Smart Citation

Multiply-rooted multiscale models for large-scale estimation

Fieguth

2001

IEEE Trans. on Image Process.

Self Cite

View full text Add to dashboard Cite

Divide-and-conquer or multiscale techniques have become popular for solving large statistical estimation problems. The methods rely on defining a state which conditionally decorrelates the large problem into multiple subproblems, each more straightforward than the original. However this step cannot be carried out for asymptotically large problems since the dimension of the state grows without bound, leading to problems of computational complexity and numerical stability. In this paper, we propose a new approach to hierarchical estimation in which the conditional decorrelation of arbitrarily large regions is avoided, and the problem is instead addressed piece-by-piece. The approach possesses promising attributes: it is not a local method-the estimate at every point is based on all measurements; it is numerically stable for problems of arbitrary size; and the approach retains the benefits of the multiscale framework on which it is based: a broad class of statistical models, a stochastic realization theory, an algorithm to calculate statistical likelihoods, and the ability to fuse local and nonlocal measurements.

show abstract

Section: A Motivationmentioning

confidence: 93%

“…18, via a preprocessing stage. In this collaboration [13], interest is focused on the shortterm dynamics of the ocean temperature, so the data is meanremoved (i.e., the static or systematic component is subtracted).…”

Section: B Ocean Temperaturementioning

confidence: 99%

Multiply-rooted multiscale models for large-scale estimation

Fieguth

2001

IEEE Trans. on Image Process.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Fig.3(a) and Fig.3(b) are 2 estimates of dense fields. We use the algorithm in [2] for these estimates. Fig.3(c) is the corresponding motion estimation results by Horn and Schunck algorithm.…”

Section: Experiments Resultsmentioning

confidence: 99%

“…has been studied extensively and has already been employed in a Kalman filtering scheme ( [2]). However because of some difficulties, especially data sparcity, motion estimation for SST is still an open problem.…”

Section: Introductionmentioning

confidence: 99%

Motion estimation of sparse, remotely-sensed fields

Khellah²,

Fieguth³

et al.

Canadian Conference on Electrical and Computer Engineering 2001. Conference Proceedings (Cat. No.01TH8555)

Self Cite

View full text Add to dashboard Cite

Sea surface temperature (SST) can be estimated from remotely-sensed images. Because of the sparsity of the available observation it is ideal to do estimation using dynamic methods (such as Kalman filtering). To model dynamics of SST accurately we need to know motion of sea current.The traditional video motion estimation problem is straightforward, in some ways, because there are so few constraints. That is, the motion vectors are pretty much arbitrary, and successive image frames are densely pixellated, have the same number of pixels, with similar noise statistics. However there are many motion estimation problems, particularly in the area of remote sensing, which do not share these properties. In this paper we investigate the problem of determining the motion field of the sea surface, based on infrared measurements of surface temperature. This problem is challenging in that only a subset of the whole domain is measured at each point in time; specifically, only a few stripes are imaged. In addition, because of clouds, the measured subset varies from time to time; in fact, some days absolutely nothing is imaged. The quality (level of noise) can also vary from pixel to pixel. Our research will be based on the following assumptions and observations: the motion field should be smooth and ideally divergence-free, i.e. the motion field is close to time-stationary. Based on these assumptions we choose to use optical flow method for this motion problem. We handle difticulty of data sparcity by pre-estimation to get a dense field. Pre-estimation can be refined by integrating this motion estimation result. Preliminary experiment result will be shown in the end. I . INTRODUCTIONTo track the changes of sea surface temperature (SST) we need to dynamically estimate temperature field from remotely-sensed images. For this purpose accurate model of sea dynamics is required.There are two main factors to concern in modelling SST: temperature diffusion and current motion. Diffusion model 500 t 450 I I 50 100 150 WO 250 300 350 400 450 500 Figure 1 -Sparse Field of SST has been studied extensively and has already been employed in a Kalman filtering scheme ([2]). However because of some difficulties, especially data sparcity, motion estimation for SST is still an open problem. Here data sparcity means that usually only a subset of the whole domain is measured at each point time in time; specifically, only a few stripes are imaged. In addition, because of clouds, the measured subset varies from time to time; in fact, some dayes absolutely nothing is imaged. The quality ( level of noise) can also vary from time to time.In this paper we discuss a motion estimation method for SST field. Based on the assumption that this motion field is smooth and close to be time-stationary we propose using optical flow algorithm with smoothness constraint for motion estimation. To handle sparcity problem we first use diffusion-based Kalman filter to get a dense estimate of the field. Then we apply optical flow method to this dense field. DYNAMIC ESTIMATI...

show abstract

Large scale dynamic estimation of ocean surface temperature

Cited by 2 publications

References 5 publications

Multiply-rooted multiscale models for large-scale estimation

Multiply-rooted multiscale models for large-scale estimation

Motion estimation of sparse, remotely-sensed fields

Contact Info

Product

Resources

About