The paper explorers the application of ANFIS (Adaptive Network-based Fuzzy Inference System) in the extraction of
remote sensing information of wetland, which integrates the advantages of fuzzy inference and neural network. Firstly,
the paper discusses characteristics, principles and process of the method. Secondly, Taking the wetland of Yellow River
Delta in KenLi County as the experimental area, it analysis the extraction of remote sensing information of wetland
based on ANFIS. Finally, it compares the extraction result of the method with the traditional classification method, in
which show ANFIS's better performance.
In the process of remote-sensed image fusion with wavelet packet transform, wavelet basis with different properties can
exhibit different fusion performance. It is significant to find the best wavelet packet basis and apply it in the process.
However, for image fusion, best basis searching algorithm must works within two wavelet packet trees, in the case that
the present algorithm only works within one tree. The paper firstly proposes a new searching algorithm working in two
trees, then realizes a new image fusion method using wavelet packet transform with the best basis that is developed from
the new algorithm. Experiment testifies: under the fusion rule based on texture, the method develops more advantage of
wavelet packet transform, and gains a better fusion performance compared with other image fusion method using
wavelet packet transform (including wavelet transform).
Tobacco is one of important crops in our country, and brings the significant irreplaceable effect into playing in
countrywide economic growth. So the monitoring and scientific management of tobacco fields show especially important
to us. To monitor growing crops in a large scale is a complicated problem and a satisfied method to know what the way a
crop is growing has been sought by the scientists in the field. At present, the study of tobacco remote sensing monitoring
is less both at home and abroad. In this paper, we try to obtain tobacco field and area by remote sensing with Yunan
Province Honghe State Tobacco County as example. We adopt rejecting interfering tobacco field information
classification method of supervision while monitoring and get an ideal result. Simultaneity, we also offered the
suggestion of further improving classification precision.
The quantitative evaluation of desertification extent with remotely sensed imagery has been a hot spot of remote sensing
application research. The evaluation process should consider the principles of dominance, integration and so on.
Traditional evaluation methods to desertification extent are usually carried out at the scale of discrete pixels, which fails
to taken into account of the influence of adjacent pixels and results in noises on the evaluation result images, inducing
the unilateralism result. If we try to use filters to reduce the noises, then the evaluation results will be wrong contrasting
with its real result. Based on former researches and the geographic science principle, this paper discusses the method of
assessing desertification extent at the scale of geographic unit, in which the geographic unit is determined by vegetation
coverage index and spatial information. The test results show that this method provides more accurate assessment of the
ground situation avoiding the limitations of traditional methods.