Tree species distribution is an important issue for sustainable forest resource management. However, the accuracy of tree species discrimination using remote-sensing data needs to be improved to support operational forestry-monitoring tasks. This study aimed to classify tree species in the Liangshui Nature Reserve of Heilongjiang Province, China using spectral and structural remote sensing information in an auto-mated Random Forest modelling approach. This study evaluates and compares the performance of two machine learning classifiers, random forests (RF), support vector machine (SVM) to classify the Chinese high-resolution remote sensing satellite GF-1 images. Texture factor was extracted from GF-1 image with grey-level co-occurrence matrix method. Normalized Difference Vegetation Index (NDVI), Ratio Vegetation Index (RVI), Enhanced Vegetation Index (EVI), Difference Vegetation Index (DVI) were calculated and coupled into the model. The result show that the Random Forest model yielded the highest classification accuracy and prediction success for the tree species with an overall classification accuracy of 81.07% and Kappa coefficient value of 0.77. The proposed random forests method was able to achieve highly satisfactory tree species discrimination results. And aerial LiDAR data should be further explored in future research activities.
Chlorophyll plays an important role in crop photosynthesis, and it is an indicator of crop growth and stress state.
Estimation of leaf chlorophyll content of maize from remote sensing data was investigated using radiative transfer model
inversion and wavelet analysis. Hyperspectral data of maize were measured in two natural fields using ASD field
spectrometer, chlorophyll content was collected with a SPAD-502 chlorophyll meter. The bands between 350-1300nm
spectra region were selected for the preprocessing, 10 spectra of each sampling point measurements of maize were
averaged for smoothing. PROSPECT was used to generate very large spectral data sets, with which spectra region was set
to 350-1300nm. The original hyperspectral of maize were applied wavelet transform with wavelet function of Haar, DB9,
sym6, coif3, bior4.4, dmey to get transform coefficients, spectral reflectance of maize were obtained after the de-noising
processing. Support vector machine was trained the training data set, in order to establish hyperspectral estimation model
of chlorophyll content. A validation data set was established based on hyperspectral data, and the leaf chlorophyll content
estimation model was applied to the validation data set to estimate leaf chlorophyll content of maize. The hyperspectral
estimation model yielded results with a coefficient of determination of 0.8712 and a mean square error (MSE) of 76.1786.
The results indicated that by decomposing leaf spectra, the wavelet analysis can be used to a fast and accurate method for
estimations of chlorophyll content.
Accurate retrieval of crop chlorophyll content is of great importance for crop growth monitoring, crop stress
situations, and the crop yield estimation. This study focused on retrieval of rice chlorophyll content from data through
radiative transfer model inversion. A field campaign was carried out in September 2009 in the farmland of ChangChun,
Jinlin province, China. A different set of 10 sites of the same species were used in 2009 for validation of methodologies.
Reflectance of rice was collected using ASD field spectrometer for the solar reflective wavelengths (350-2500 nm),
chlorophyll content of rice was measured by SPAD-502 chlorophyll meter. Each sample sites was recorded with a
Global Position System (GPS).Firstly, the PROSPECT radiative transfer model was inverted using support vector
machine in order to link rice spectrum and the corresponding chlorophyll content. Secondly, genetic algorithms were
adopted to select parameters of support vector machine, then support vector machine was trained the training data set, in
order to establish leaf chlorophyll content estimation model. Thirdly, a validation data set was established based on
hyperspectral data, and the leaf chlorophyll content estimation model was applied to the validation data set to estimate
leaf chlorophyll content of rice in the research area. Finally, the outcome of the inversion was evaluated using the
calculated R<sup>2</sup> and RMSE values with the field measurements. The results of the study highlight the significance of
support vector machine in estimating leaf chlorophyll content of rice. Future research will concentrated on the view of
the definition of satellite images and the selection of the best measurement configuration for accurate estimation of rice