Coastal wetland vegetation classification with a Landsat Thematic Mapper image

Abstract

Coastal wetland vegetation classification with remotely sensed data has attracted increased attention but remains a challenge. This paper explored a hybrid approach on a Landsat Thematic Mapper (TM) image for classifying coastal wetland vegetation classes. Linear spectral mixture analysis was used to unmix the TM image into four fraction images, which were used for classifying major land covers with a thresholding technique. The spectral signatures of each land cover were extracted separately and then classified into clusters with the unsupervised classification method. Expert rules were finally used to modify the classified image. This research indicates that the hybrid approach employing sub-pixel information, an analyst's knowledge and characteristics of coastal wetland vegetation distribution shows promise in successfully distinguishing coastal vegetation classes, which are difficult to separate with a maximum likelihood classifier (MLC). The hybrid method provides significantly better classification results than MLC.

Use of bi-Seasonal Landsat-8 Imagery for Mapping Marshland Plant Community Combinations at the Regional Scale

“…For example, Carreño et al, 2008 mapped salt marshes, salt steppes and reeds in a 5 km 2 area of wetland; similarly, Li et al, 2010;Zhang et al, 2011 discriminated only three habitat classes (Suaeda salsa, Spartina anglica and Spartina alterniflora) on a 100 km 2 salt marsh site. Recent work by Valentini et al (2015) also aimed to map heterogeneous vegetation, approached by the dominant plant species, but the Table 4 Under-(A) and over-detection rates (B) expressed in percent per class for the vegetation classifications taken from mono-temporal and biseasonal Landsat-8 images, using the Maximum likelihood classifier (Géhu, 2011).…”

Flood Mapping Based on Multiple Endmember Spectral Mixture Analysis and Random Forest Classifier—The Case of Yuyao, China

“…Above all, optical remote sensing can also play an important role in flood mapping given cloud free conditions. Nevertheless, mixed pixels are common in medium spatial resolution data such as TM and ETM+, and such pixels have been recognized as a problem for remote sensing applications [13][14][15][16][17][18][19][20]. In terms of flood monitoring using medium resolution data, flooded and non-flooded features may co-exist within one pixel, resulting in the challenge to extract these small and fragmented flooded patches especially under complex urban landscapes.…”

“…To deal with the mixed pixel problem, several approaches such as linear spectral mixture analysis (LSMA) [13], fuzzy-set possibilities [14], and Bayesian possibilities [15] have been developed to partition the proportions of each pixel between classes. Among these methods, LSMA appears to be the most promising and has been widely used to extract sub-pixel information with physical meaning [13,[16][17][18][19]. LSMA assumes that the reflectance of each pixel can be modeled as a linear combination of a few spectrally pure land cover components, known as endmembers [13].…”

Classification of land-cover types in muddy tidal flat wetlands using remote sensing data

“…22,23 Previously, several studies have employed Landsat TM images for mapping wetland vegetation types. [24][25][26] They attempted to combine TM imagery and ancillary environmental data using classification trees. 27,28 However, complete utilization of image and environmental characteristics to quickly and accurately extract tidal flat wetland vegetation information is still an urgent and important challenge.…”