A Multiobjective Method Leveraging Spatial–Spectral Relationship for Hyperspectral Unmixing

Liu, Erfeng; Wu, Zikai; Zhang, Hongjuan

doi:10.1109/tgrs.2022.3210198

Cited by 11 publications

(3 citation statements)

References 76 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To avoid the problem posed by hyperparameters, sparse unmixing methods based on multiobjective optimization have been developed [11], [12], [14]. Considering the sparsity level ||x|| 0 and the squared reconstruction error ||y − Φx|| 2 as the objectives, sparse unmixing may be formulated as a biobjective optimization problem…”

Section: A Sparse Unmixingmentioning

confidence: 99%

“…Several different approaches have been developed for finding the subset of endmembers that neither overfits nor underfits the pixel spectrum y. Greedy and relaxation-based unmixing algorithms select the best subset with the help of hyperparameters whose values need to be determined by a human expert. However, the optimal values of the hyperparameters are typically data dependent and difficult to find [11], [12], [13], [14]. Furthermore, the need for expert intervention reduces the speed at which a large number of pixels can be analyzed.…”

Section: Introductionmentioning

confidence: 99%

“…Angle-based elbow estimation (ABEE) [14] has been used to select the best solution from the set of Pareto-optimal solutions in many studies, e.g., [17], [18], [19], [20]. However, ABEE was criticized in [11] by stating that it sometimes selects wrong solutions or fails to unambiguously identify the best solution. Typical downsides of Bayesian unmixing algorithms include computational heaviness and general complexity [21], [22].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A New Elbow Estimation Method for Selecting the Best Solution in Sparse Unmixing

Sarjonen

Räty

2023

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

View full text Add to dashboard Cite

The goal of hyperspectral image analysis is often to determine which materials, out of a given set of possibilities, are present in each pixel. As hyperspectral data are being gathered in rapidly increasing amounts, automatic image analysis is becoming progressively more important. Automatic identification of materials from a mixed pixel is possible with 1) Bayesian unmixing algorithms and 2) multiobjective sparse unmixing algorithms when a method such as elbow estimation is used to select the best solution from the set of Pareto-optimal solutions. We develop a new elbow estimation method called termination condition adaptive elbow (TCAE) for selecting the best solution from the set of Paretooptimal solutions to a biobjective unmixing problem. Specifically, the two objectives are assumed to be the sparsity level of the fractional abundance vector and the reconstruction error. We conduct experiments with real-world unmixing applications in mind, and TCAE performs significantly better than a state-of-the-art elbow estimation method when they are both used to select the best solution from the sequence of fractional abundance vectors generated by iterative spectral mixture analysis (ISMA). Furthermore, the combination of ISMA and TCAE is able to identify endmembers from mixed pixels several times faster and with higher F1-score than the two Bayesian unmixing algorithms used as a reference. We conclude that the combination of ISMA and TCAE facilitates automatic, reliable, and rapid identification of endmembers from mixed pixels.

show abstract

Section: A Sparse Unmixingmentioning

confidence: 99%