Xi, ZHANG
A Supervised Learning Approach towards Ranking and Sorting Problems
[Abstract]
Recently, classification problems have been well developed. The popular methods for classification, Neural Network, SVM and Boosting have also been applied to related fields such as ranking and sorting. Although ranking and sorting problems have attracted the attention of researchers in the machine learning community, most of the solutions simply treat the ranking/sorting problems as multi-class classification problems. In this paper, we will propose a specified approach for the ranking / sorting problems based on 2-class classification SVM method.[Keywords] Ranking, Sorting, Supervised Learning, Kernel Method, Support Vector Machines, SVM
[Summary]
[Reference]
Prof. Shuyi ZHANG.A New Method for Spatial Feature Extraction and Classification of Remote Sensing Image
[Abstract]
The extraction and classification problem of spatial features from high r esolution satellite sensor image, especially from the image covering urban areas, is a very significant but challenging task. However, it is very difficult to be implemented and the main obstacle comes from high-dimensional and complicated properties of spatial features. In this paper, we propose to use a two-dimension wave-let transform as well as a classification method---support vector machine (SVM) to address this issue. Also, a boosting method is involved in order to improve the accuracy of classification. We will show in our experiment that SVM with Boosting leads to a more admissible result by choosing several SVM kernels, including the linear kernel and the Gaussian kernel.[Keywords] spatial feature exaction; classification;
[Summary]
[Reference]
*IGARSS : IEEE International Geosciences and Remote Sensing SymposiumDenver, Colorado 2006 Seoul, Korea 2005
Feature Extraction and Classification with Wavelet Transform and Support Vector Machines
[Abstract]
Abstract This paper introduces texture classification method by using wavelet transform and support vector machines. First, a wavelet-based texture feature set is obtained by the overcomplete wavelet decomposition of local areas in remote sensing images, then the texture classification is carried out by the SVM technique. The experiment results show that the method above discussed is admissible.[Keywords] Feature extraction; texture classification; wavelet; support vector machines; neural networks
[Summary]
[Reference]
*IGARSS : IEEE International Geosciences and Remote Sensing SymposiumDenver, Colorado 2006 Seoul, Korea 2005
A New Method for Multi-Source Remote Sensing Image Fusion
[Abstract]
Multi-source remote sensing image fusion is one of the advanced image processing technologies. The main purpose of image fusion is to create a new image regrouping the complementary information of the original images. One of the most relevant applications is fusion of a high spatial resolution image with a low spatial resolution image. In this paper, we study the Intensity-Hue-Saturation Transformation(HIS), Principal Components Analysis, High Pass Filter(HPF), Wavelet transformation(WT), and show their deficiencies, and then propose a novel image fusion method based on CMYK. Experiment results show that the fusion method based on CMYK is remarkably better than the classical image fusion methods.[Keywords]Remote sensing ,image fusion, ihs, wavelet cmyk
[Summary]
[Reference]
*IGARSS : IEEE International Geosciences and Remote Sensing SymposiumDenver, Colorado 2006 Seoul, Korea 2005
The Construct and Implementation of Complex Document/View System Based on .NET Framework
[Abstract]
Among the design of applications' user interface, there is usually a demand that more than two views displayed on the screen at the same time. Sometimes models of a combined system have many common methods that only need one existence, as well as separated menu, data and other resources. In the two cases, the Document/View system could provide with more effective functions and friendlier user interface. This paper include a method of the construct of the Document/View applications, mainly refer to the complex Document/View system based on .NET Framework.[Keywords] .NET Framework; MFC; Document/View; MDI; Splitter Window