Adaption of a Self-Learning Algorithm for Dynamic Classification of Water Bodies to MERIS Data

Fichtelmann, Bernd; Borg, Erik; Guenther, Kurt P.

doi:10.1007/978-3-319-09144-0_26

Cited by 2 publications

(15 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…But it was already shown [6] that the generic principle can be transferred to other sensor data as MERIS-FR(S) or -RR data respectively. The basic principle of DySLEM described in [5] consists first in the creation of a static and two dynamic land-water masks where water pixels classified by the dynamic masks are labelled as "candidates" for water.…”

Section: Dynamically Self-learning Evaluation Methods (Dyslem) -An Adomentioning

confidence: 98%

“…2 shows the land-water masks of the Valdecanas Dam Lake of river Tejo, Spain, for two spatial resolutions. [6]) and of VGT (second row) data. Fraction of water for a VGT pixel in percentage (d), static masks with ≥10% (e) and ≥60% fraction of water (f).…”

Section: Generation Of Regional Static Land-water Mask (Ws1)mentioning

confidence: 98%

“…The structure of processor is generic, already presented in [5] and a modification to MERIS data is described in [6]. Apart from the changing frame size only the sensor specific classifications for BOA-VGT reflectance data have to be used.…”

Section: Structure Of the Processormentioning

confidence: 99%

“…As in [5], [6] the actual best available land-water mask, the SRTM Water Body Data (SWBD) acquired during the Shuttle Radar Topographic Mission was used for the geographical region between 60°S and 60°N. The spatial accuracy of SWBD is better than 30m [14].…”

Section: Available Static Land-water Masksmentioning

confidence: 99%

“…Ground observations showed that they are mostly swimming reed islands. As already mentioned in [5], [6], the precise global detection of water bodies is a complex task because of changing atmospheric conditions as haze, aerosol and cloud but also due to the surface roughness of water bodies. Additionally, different components as chlorophyll, yellow substance or suspended sediments influence the spectral characteristics of water bodies.…”

Section: Introductionmentioning

confidence: 98%

See 4 more Smart Citations

Adaption of a Self-Learning Algorithm for Dynamic Classification of Water Bodies to SPOT VEGETATION Data

Fichtelmann

Guenther

Borg

2015

Computational Science and Its Applications -- ICCSA 2015

Self Cite

View full text Add to dashboard Cite

Abstract. Within the ESA CCI "Fire Disturbance" project a dynamic selflearning water masking approach originally developed for AATSR data was modified for MERIS-FR(S) and MERIS-RR data and now for SPOT VEGETATION (VGT) data. The primary goal of the development was to apply for all sensors the same generic principles by combining static water masks on a global scale with a self-learning algorithm. Our approach results in the generation of a dynamic water mask which helps to distinguish dark burned area objects from other different types of dark areas (e.g. cloud or topographic shadows, coniferous forests). The use of static land-water masks includes the disadvantage that land-water masks represent only a temporal snapshot of the water bodies. Regional results demonstrate the quality of the dynamic water mask. In addition the advantages to conventional water masking algorithms are shown. Furthermore, the dynamic water masks of AATSR, MERIS and VGT for the same region are presented and discussed together with the use of more detailed static water masks.

show abstract

Section: Dynamically Self-learning Evaluation Methods (Dyslem) -An Adomentioning

confidence: 98%

Section: Generation Of Regional Static Land-water Mask (Ws1)mentioning

confidence: 98%

Section: Structure Of the Processormentioning

confidence: 99%

Section: Available Static Land-water Masksmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 98%

See 3 more Smart Citations

Adaption of a Self-Learning Algorithm for Dynamic Classification of Water Bodies to SPOT VEGETATION Data

Fichtelmann

Guenther

Borg

2015

Computational Science and Its Applications -- ICCSA 2015

Self Cite

View full text Add to dashboard Cite

show abstract

Semisupervised Hyperspectral Image Classification Using Spatial-Spectral Information and Landscape Features

Cui

Wang

et al. 2019

IEEE Access

View full text Add to dashboard Cite

In hyperspectral image classification, the foremost task is that: how can we apply limited labeled samples to achieve good classification results? Spatial-spectral classification methods, which assign a label to each pixel regarding both spatial and spectral information, are effective to improve classification performance. Moreover, semisupervised learning (SSL) focuses on the scenario that the number of labeled data is rather small while a large number of unlabeled data are available. To complement spatial-spectral classification methods and semisupervised learning for each other, we propose a novel learning landscape features semisupervised framework (LLFSF) based on M-training algorithm and weighted spatial-spectral double layer SVM classifiers module (WSS-DSVM). In this novel framework, we first propose a SLIC (simple linear iterative clustering) based non-local superpixel segmentation algorithm to initially learn landscape feature and spatial composition. Then, we apply WSS-DSVM module to obtain initial classification maps. To better characterize complex scenes of hyperspectral images, we quantizes both the landscape diversity and separability from the initial classification map, which increase availability of spatial details and structural information of objects. Finally, we put some patches with lower accuracy into Multiple-training algorithm for further classification. In order to achieve an unbiased evaluation, we have evaluated the performance of LLFSF on three different scene hyperspectral data sets and compare it with that of three state-of-the-art hyperspectral image classification methods. The experimental results confirm the efficacy of the proposed framework. INDEX TERMS Hyperspectral image classification, landscape features, spatial-spectral information, semisupervised learning.

show abstract

Adaption of a Self-Learning Algorithm for Dynamic Classification of Water Bodies to MERIS Data

Cited by 2 publications

References 9 publications

Adaption of a Self-Learning Algorithm for Dynamic Classification of Water Bodies to SPOT VEGETATION Data

Adaption of a Self-Learning Algorithm for Dynamic Classification of Water Bodies to SPOT VEGETATION Data

Semisupervised Hyperspectral Image Classification Using Spatial-Spectral Information and Landscape Features

Contact Info

Product

Resources

About