Extraction And Application Of Forest Resources Information Of The Fusion Image Based On Sub-compartment

The forest and vegetation resources are the main body of the terrestrial ecosystem.The study of forest and vegetation coverage and its change information has a great significance for the accurate understanding of the dynamic changes of forest and vegetation and its impact on the ecological environment.With the development of aerospace technology,remote sensing has become an important technical means to obtain forest and vegetation change information.ZY-1-02 C as Chinese high-resolution satellite has great significance to improve the capacity of data security of forest resources survey and monitoring in our country.The pre-processing includes ortho rectification based on 10 m multi-spectral and 2.36 m panchromatic data of ZY-1-02 C satellite images of Kaihua,Zhejiang in 2012,2013 and 2014.This paper explores different relative radiation correction methods and a comparative analysis are made on the methods of extracting information about the change of forest sub-compartment in the study area.This study can provide a basis and reference for the rapid updating of forest resources,the results show that:(1)The relative radiometric correction result based on wavelet transform is better.Wavelet transform has minimum value which is-0.5394,the order is 1,histogram matching has maximum value which is 0.6035 based on comprehensive evaluation value of relative radiometric correction image based on RMSE and average gradient by standardization method.The relative radiometric correction of wavelet transform is good,which shows that the method is suitable for the radiation correction of ZY-1-02 C satellite image in Kaihua,Zhejiang.(2)The relative radiometric correction result of wavelet transform based on haar wavelet basis is better.The wavelet transform relative radiometric correction image comprehensive evaluation results based on rbio3.1,haar wavelet basis is 0.4236 and-0.3525 based on RMSE and average gradient by standardization method,wavelet transform image value based on haar basis ranked 1,this result show that the wavelet transform based on haar basis relative radiometric correction image is better and this method can improve the accuracy of forest resources dynamic monitoring.(3)The Wavelet algorithm has better fusion effect.Wavelet,IHS,HSV,HPF,GS,Brovey and PCA fusion methods are used to fuse the multi spectral and panchromatic images of the study area.Qualitative and standardized quantitative synthesis method based on forest resources sub-compartment is used to evaluate the effect of fusion image in Kaihua,the result show that Wavelet fusion algorithm fusion image of ZY-1-02 C satellite is better,which has good spectral fidelity and enhance information richness.(4)Extraction of forest sub-compartment changing information based on statistical distribution.The detection accuracy of the relative radiometric correction image is superior to the ZY-1-02C1 product ortho-rectified image according to the comprehensive evaluation index of the correct rate,the false detection rate and the missed detection rate based on the value of visual interpretation,and the overall accuracy of forest sub-compartment change was 88.88%.The results show that relative radiometric correction image can get better results but there is still a certain degree of missed detection and false detection when extracting the changing information of forest sub-compartment.(5)Extraction of forest sub-compartment changing information by the mean of building statistics based on the NDVI difference characteristic variable and the overall accuracy of the change information extraction is 96.87%,the missed detection rate is about 3.13%,and the false detection rate is about 7.7%,the detection accuracy is better than the statistical distribution method and which is consistent with the requirement of forest resource change information extraction and a good method to extract the information of forest sub-compartment change.(6)The automatic method of extract changing forest sub-compartment is applied to the extraction of 2013-2014 images and the result show that the correct rate of vegetation coverage and reduction information extraction is 83.87% and 99.42%,the false detection rate is 6.45% and 5.78% and the detection rate is 12.9% and 0.58%.It can be seen that the method is suitable for the extraction of forest sub-compartment in Kaihua,Zhejiang based on ZY-1-02 C satellite image.