HYPERSPECTRAL COASTAL WETLAND CLASSIFICATION BASED ON A MULTIOBJECT CONVOLUTIONAL NEURAL NETWORK MODEL AND DECISION FUSION

Abstract

The phenomenon of spectral aliasing exists for coastal wetland object types, which leads to class mixing. This letter proposes a multiobject convolutional neural network (CNN) decision fusion classification method for hyperspectral images of coastal wetlands. This method adopts decision fusion based on fuzzy membership rules applied to single-object CNN classification to obtain higher classification accuracy. Experimental results demonstrate the effectiveness of the proposed method for the six object types, including water, tidal flat, reed, and other vegetation types. The overall accuracy of the decision fusion classification method based on fuzzy membership is 82.11%, which is 3.33% and 6.24% higher than those of single-object feature band CNN and support vector machine methods. The classification method based on multiobject CNN decision fusion inherits the characteristics of single-object feature bands of the CNN, making it a practical approach to image classification under the challenging conditions in which class mixing occurs.

Existing System

Spatial-spectral classification, Decision-level fusion of spectral reflectance and derivative information.

Proposed System

This system develops a hyperspectral coastal wetland classification method, using decision fusion strategies to execute multiobject CNN classification. First, the method uses spectral deviation to obtain the separable spectrum range of object types based on field spectral data, and then adopts the CNN model with spatial–spectral features to classify coastal wetland types. Based on the CNN classification results of the single-object feature bands, we obtain a final classification result by using a fuzzy membership degree decision fusion algorithm. We use the proposed classification method to carry out classification of the Yellow River Estuary coastal wetlands on a hyperspectral image sensed by the Compact High Resolution Imaging Spectrometer (CHRIS) on the Project for On Board Autonomy satellite.  

Architecture

BLOCK DIAGRAM