Application Of Improved Wavelet Transform In Noise Reduction Of Remote Sensing Images

Remote sensing technology belongs to an advanced technology that has gradually emerged with the development of science and technology.Its definition refers to the perception of electromagnetic waves reflected by a target or radiation,as well as visible signals,near-infrared,infrared,and other signals over long distances.During the acquisition and transmission of remote sensing information,it is often influenced by various reasons(such as: air atmosphere,ground water vapor reflection,refraction,scattering and other optical phenomena on imaging imaging spectrometer imaging and the Earth’s magnetic field on the transmission process.Electromagnetic interference,etc.)and generate a lot of noise.These noises will make the key information of the remote sensing image such as edge texture and important feature details fuzzy,and thus the key information contained in the remote sensing image will be lost,resulting in the degradation of the overall image quality and affecting the subsequent application of the processed remote sensing image.Therefore,in order to obtain higher definition,signalto-noise ratio and improve the quality of remote sensing images,denoising of remote sensing images containing noise is a crucial preprocessing step that affects the application of remote sensing images.In this paper,the existing common wavelet analysis denoising algorithm and independent component analysis(ICA)are studied in depth and a large number of literatures and books are combined to perform the above two independent denoising algorithms.Combine and apply this improved wavelet analysis-ICA algorithm to practical remote sensing image denoising applications to test the feasibility and denoising effect of the new denoising algorithm.The main conclusions of this paper are as follows: By using the traditional wavelet transform de-noising algorithm and the improved wavelet analysis-ICA de-noising algorithm,the Aba Tibetan and Qiang Autonomous Prefecture in Sichuan Province was subjected to a series of pre-processing such as radiation calibration,atmospheric correction,image fusion,and image mosaic.The LandSat 8 OLI remote sensing image of Wenchuan County was denoised and the SNR/SNR and RMSE of the remote sensing image obtained by analyzing the two denoising methods were verified.The superiority of the noise method.Using the traditional wavelet transform de-noising algorithm and the improved wavelet analysisICA de-noising algorithm to de-noise the remote sensing image to use the support vector machine(SVM)supervised classification classifier in ENVI software to perform the features respectively.Classification,by comparing the classification accuracy factors such as total classification accuracy(%),Kappa coefficient(%),Error(%),etc.,the improved wavelet analysis-ICA denoising algorithm proposed in this paper is used to denoise the actual remote sensing image.Feasibility in practical application.