Forest precision classification products were the basic data for surveying of forest resource, updating forest subplot information, logging and design of forest. However, due to the diversity of stand structure, complexity of the forest growth environment, it's difficult to discriminate forest tree species using multi-spectral image. The airborne hyperspectral images can achieve the high spatial and spectral resolution imagery of forest canopy, so it will good for tree species level classification. The aim of this paper was to test the effective of combining spatial and spectral features in airborne hyper-spectral image classification. The CASI hyper spectral image data were acquired from Liangshui natural reserves area. Firstly, we use the MNF (minimum noise fraction) transform method for to reduce the hyperspectral image dimensionality and highlighting variation. And secondly, we use the grey level co-occurrence matrix (GLCM) to extract the texture features of forest tree canopy from the hyper-spectral image, and thirdly we fused the texture and the spectral features of forest canopy to classify the trees species using support vector machine (SVM) with different kernel functions. The results showed that when using the SVM classifier, MNF and texture-based features combined with linear kernel function can achieve the best overall accuracy which was 85.92%. It was also confirm that combine the spatial and spectral information can improve the accuracy of tree species classification.
With the widely application of Remote Sensing, the request for the accuracy of classification is getting higher and
higher in each application fields. The aim of this paper is to test whether spectra reflectance of various tree leaves
measured under ground-level conditions contain sufficient spectral information for discriminating tree species, and finds
a way to discriminate tree species from their spectra reflectance. This study is one of the most important prerequisites to
the future use of airborne and satellite hyper-spectral data. First, spectral reflectance of 8 tree species in Huazhong
district including herbaceous, conifers and hardwoods which between 400nm and 900nm were recorded from canopy,
using ASD hand-held Spectrometer. Next, the spectral were statistically tested using one-way ANOVA to see whether
they significantly differ at every spectral location. Finally, the spectral separability between each tree species was
quantified using the Jeffries-Matusita(J-M)distance measure. It turned out that the 8 species under study were statically
different at most spectral locations, with a significant level of 0.01. Moreover, the J-M distance indices calculated for all
species illustrated that the trees were spectrally separable.
Change Detection is one of the most popular topics in the field of Multi-temporal Remote Sensing (RS)
applications. In this paper, a novel approach was introduced for the change detection of the urban area. This
approach adopts the Dempster-Shafer(D-S) algorithm for feature fusion of the multi-temporal RS images. It, in
the first place,,constructs difference images of pixel and context respectively. These two difference images
present the features of changes in different scales. The pixel difference image is obtained by fusing the results
of the subtraction operation and the division operation, while the context difference image is obtained by the
image context. Secondly, by using the difference images, two evidences could be constructed. These
evidences are not certain, but they can give more reliable combination result if considering the average support
of the evidence to different subsets in the assignment framework. And based on these evidences, the
criterion function could be established by the D-S theory. At last, an improved D-S algorithm is applied to fuse
the two different features for detecting the change information of the RS images. An experiment, using the
SPOT and TM images of Wuhan urban area, has compared the accuracy of edge detection by using the new
fusion algorithm and the existent ones. The result shows that the method of improved D-S is solid and
efficacious, which has preferable value in remote sensing applications.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.