Abstract
Hyperspectral remote sensing technology can acquire nearly continuous spectrum information and rich sea ice image information, thus providing an important means of sea ice detection. However, the correlation and redundancy among hyperspectral bands reduce the accuracy of traditional sea ice detection methods. Based on the spectral characteristics of sea ice, this study presents an improved similarity measurement method based on linear prediction (ISMLP) to detect sea ice. First, the first original band with a large amount of information is determined based on mutual information theory. Subsequently, a second original band with the least similarity is chosen by the spectral correlation measuring method. Finally, subsequent bands are selected through the linear prediction method, and a support vector machine classifier model is applied to classify sea ice. In experiments performed on images of Baffin Bay and Bohai Bay, comparative analyses were conducted to compare the proposed method and traditional sea ice detection methods. Our proposed ISMLP method achieved the highest classification accuracies (91.18% and 94.22%) in both experiments. From these results the ISMLP method exhibits better performance overall than other methods and can be effectively applied to hyperspectral sea ice detection.
Highlights
Sea ice is the cause of most marine disasters in polar and high latitude regions
To choose the optimal band combination that features both a large amount of information and a low-degree of inter-band correlation, this paper proposes an improved similarity measurement method based on linear prediction (LP) (ISMLP) for hyperspectral sea ice detection
Can see thatspectral the different of curves sea ice Method can discriminated by their reflectance
Summary
Sea ice is the cause of most marine disasters in polar and high latitude regions. As a component of the global marine and atmospheric system, sea ice with high albedo impacts marine industries, and matter and momentum exchanges between the atmosphere and the ocean. Studying changes in sea ice is highly important. Because of the harsh natural environment in areas covered with sea ice, conventional observation methods such as the in situ sampling method and the visual estimate method [3] cannot acquire detailed information on sea ice changes in a timely and effective manner. Remote sensing technology, which can analyze data for large areas of sea ice rapidly and extensively, is widely used in sea ice detection. Compared with the multispectral technologies, hyperspectral remote sensing technology can acquire nearly continuous spectrum information and rich sea ice image information, providing an important resource for sea ice detection. Used sea ice detection methods include the threshold method, unsupervised classification and supervised classification [6,7] such as in the K-means and support vector machine (SVM) methods
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have