Time filter

Source Type

Feng Q.,Chinese Academy of Forestry | Feng Q.,State Forestry Administration Key Laboratory | Chen E.-X.,Chinese Academy of Forestry | Chen E.-X.,State Forestry Administration Key Laboratory | And 11 more authors.
International Geoscience and Remote Sensing Symposium (IGARSS)

In order to improve the land cover classification accuracy for SAR image, Support Vector Machine (SVM), which has wide applicability is used on the land cover classification of POLSAR image in this paper. The study site is located in Tahe County, Heilongjiang Province, China, and two scenes of quad-polarization Radarsat-2 SAR images were acquired. the land cover classification of single-temporal POLSAR image by SVM, and multi-temporal POLSAR image by SVM and maximum likelihood classification (MLC) is studied separately. Then all the classification results are evaluated. Some conclusions can be got according to the analysis of all results and accuracy: Firstly, it is difficult to distinguish the different types of vegetation for the similar scattering among them in July. However, water, whose scattering characteristic is simplex, can be distinguished from others easily. Scondly, in October, the scattering characteristics among forest, shrub, grass, crop are different, therefore it is easy to distinguish vegetation because of their one from others in this period. But for water, with reduced in winter, the river width narrows, compared with it in summer, water classification accuracy is lower in this period. Thirdly, joint July and October SAR data for classification, can offset espective their own disadvantages. and improve overall accuracy. And the last one, With the characteristics that different probability density distribution, small sample, non-linear and so on, SVM shows the wide applicability. © 2012 IEEE. Source

Guo Y.,Chinese Academy of Forestry | Guo Y.,State Forestry Administration Key Laboratory | Li Z.,Chinese Academy of Forestry | Li Z.,State Forestry Administration Key Laboratory | And 13 more authors.
International Geoscience and Remote Sensing Symposium (IGARSS)

The main objective of this study was to investigate the potential of using Support Vector Machines (SVM) and Random forest (RF) to estimate forest above ground biomass (FAGB) by using multi-source remote sensing data. To do so, we introduced a basic flow of SVM to estimate FAGB from multisource remote sensing data. RF method was adept at identifying relevant features having main effects in multisource remote sensing data. Results show that: (i) In the stage of feature selection, the Random Forest model provide better results compared to the typical F-scores method. (ii) The optimal SVM model, based on the selection of features clearly demonstrate that the estimation accuracy increased by feature selection algorithm. (iii) Compared to the optimal KNN, BPNN and RBFNN model, the optimal SVM algorithm provided more accurate and robust result on the considered case. © 2012 IEEE. Source

Zhao L.,Chinese Academy of Forestry | Zhao L.,State Forestry Administration Key Laboratory | Chen E.,Chinese Academy of Forestry | Chen E.,State Forestry Administration Key Laboratory | And 8 more authors.
Wuhan Daxue Xuebao (Xinxi Kexue Ban)/Geomatics and Information Science of Wuhan University

A new segmentation approach of polarimetric synthetic aperture radar (PolSAR) data is proposed based on mean shift and spectral graph partitioning. First, Mean-shift algorithm is used to generate the over-segmentation of PolSAR image. In order to extract edge information, we apply a set of detectors based on the Wishart distribution with a hypothesis testing method that has fully considered the polarization information in PolSAR images. Then, a similarity matrix is constructed based on the over-segmentation results and image edge information. The graph partitioning process is performed using the normalized cut criterion. With this method, we improve the segmentation efficiency of spectral graph partitioning based on the over-segmentation results generated by Mean-shift. The quality of segmentation results is also improved as a result of the global optimization of spectral graph partitioning algorithm. We applied this method on Radarsat-2 full polarization images and evaluated the segmentation results. The experiement showed that this scheme can realize PolSAR segmentation effectively, speed up the original algorithm, and also demonstrates a better result than eCognition's Multi-resolution segmentation approach. Source

Discover hidden collaborations