Wenting Cai

Integration of Hyperspectral Imagery and Sparse Sonar Data for Shallow Water Bathymetry Mapping

Accurate and rapid mapping of shallow water bathymetry is essential for the safe operation of many industries. Here, we propose a new approach to shallow water bathymetry mapping that integrates hyperspectral image and sparse sonar data. Our approach includes two main steps: dimensional reduction of Hyperion images and interpolation of sparse sonar data. First, we propose a new algorithm, i.e., a sonar-based semisupervised Laplacian eigenmap (LE) using both spatial and spectral distance, for dimensional reduction of Hyperion imagery. Second, we develop a new algorithm to interpolate sparse sonar points using a 3-D information diffusion method with homogeneous regions. These homogeneous regions are derived from the segmentation of the dimensional reduction results based on depth. We conduct the experimental comparison to confirm the applicability of the dimensional reduction and interpolation methods and their advantages over previously described methods. The proposed dimensional reduction method achieves better dimensional results than unsupervised method and semisupervised LE method (using only spectral distance). Furthermore, the bathymetry retrieved using the proposed method is more precise than that retrieved using common interpolation methods.

Multi-scale urban land cover extraction based on object oriented analysis

The paper classifies urban land cover object-oriented. First, use the Mean Shift algorithm to segment the digital aerial image of study area. By changing the algorithm initialparameters, different scales of the segmentation results are got, and an optimal scale segmentation result is selected from these results, as the data source of the classification process. Then, by analyzing the spectral features, texture features, as well as the topographical features of the study area, the Fisher criterion is taken to calculate the classification ability of each feature and sort them in descending order. In order to avoid the “Hughes phenomenon”, tests are taken to find the optimal feature space dimension. Finally, by using ensemble learning algorithm, decision tree algorithm, which is a weak learner will be upgraded to strong learner to improve the classification accuracy, and then the rule which is produced by the strong learner is used to classify the study area. The classification accuracy is 89.87%, which means that the method can effectively carry out urban land use/cover information extraction; furthermore, because these algorithms used in this article are no limitation in scale, so they are also suitable for multi-scale remote sensing image classification.

Automatic waterline pick-up based on improved embedded confidence

The boundary of land and water in a remote sensing image is defined as water front. Picking up the water front plays an important role in comprehensive management of waterline, strengthening of the ability in resisting nature disaster, and the research of waterlines environmental change. Using RS methods to pick up water front has advantages of wide-range, high-precision, and ability of dynamic monitoring. Based on TM/ETM+ RS multi-spectrum images, combined with the characteristic of water front, the dissertation carries through water front pick-up approach study, and forms the water front picked-up approach based on the improved embedded confidence. This approach makes full use of spectrum eigenvalue of the image, and proposes the elementary water front pick-up approach; then uses the approach of embedded confidence, finally, realizes the automatic pick-up of water front. The main study content and conclusion are listed as follows: coastline characters and the basic mentality of water front pick-up are briefly narrated, and also the main methods of the water front pick-up at home and abroad in recent years are introduced. After analyzing kinds of methods, the dissertation uses the combination of spectrum-photometric method and embedded confidence to pick up the water front. On the basis of analyzing the gray value in each band of TM image, spectrum-photometric method will be used first to pick up water body, removing part of noise in the precondition of picking up full of the water, and then, using the approach of embedded confidence to pick up waterline, during the process, the grad magnitude information is the main basis. The result indicates that: (1)the approach based on improved embedded confidence is feasible in practice; (2)this approach can detect the water front by rule and line, also can restrain the noise effectively; (3)because the study does not have the subjective infection by the researchers, so the water front pick-up result is more exact and impersonality.

Simulation and modeling of elevation variations of radial sand banks in Jiangsu province based on spatio-temporal correlation analysis

Jiangsu province, which has vast population and limited land, is facing with land resources shortage. However, the offshore radial sand banks provide huge potential land reserve resources. But, epeirogenic movement and the variation of elevation of the radial sand banks are quite frequent in coastal area of Jiangsu province; Hence, mastering law of the variation of elevation of the radial sand banks is the precondition of making use of the radial sand banks scientifically. This paper has used remote sensing method to build a series of DEMs of radial sand banks which has constant temporal intervals, and then simulated the variations of elevation of sand banks by using these DEMs. For non-spatial data, the variation law of phenomena can be revealed by time series models. However, for spatial data such as the sand bank elevation, due to the existence of spatial dependence, goes against the hypothesis that the data is independent of each other in classic statistical analysis. That means, the elevation of a point on sand bank not only dependents on its own historical values, but also affected by its neighbors. Therefore, with comprehensive consideration of time and space, we build a forecasting model based on spatio-temporal relation analysis to simulate the elevation variations of radial sands banks. Then verify the result by using the elevation in the next time interval. Results show that this method which this paper proposed can satisfy the requirement of mesoscale applications.