Font Size: a A A

Research On Hadamard Transform Coded Signal Reconstruction And Noise Reduction Technology

Posted on:2016-12-16Degree:MasterType:Thesis
Country:ChinaCandidate:L JiFull Text:PDF
GTID:2208330461979213Subject:Optical Engineering
Abstract/Summary:PDF Full Text Request
Hadamard transform encoding detection is a multichannel technology based on signal modulation, which has high SNR, high light efficiency and high spatial resolution for detection of weak radiation signals. However, the detection and reconstruction of the complex object scene are complicated, and in the condition of low detection sensitivity, the reconstructed object images remain obvious noise. Aimed at above issues the main researches in this paper can be divided into the following aspects:In traditional direct detection, the imaging quality of weak radiation is comparatively inexactitude with the image signals totally besieged by noise. In this paper, the Hadamard transform is applied into the detection of weak radiation signals, and the signal reconstruction model Hadamard decoding is optimized, by this signal reconstruction method, the target scenes which is unavailable to image by the traditional direct way can be re-imaged.In the condition of low detection sensitivity, moderate encoding matrixes are applied to balance the performance and speed of encoded detection, which will lead to residual noise in the reconstructed object images and impair the images. Aimed at the blurred reconstructed images, the particular characteristics of noise in the reconstructed signals: noise characteristic of signal independence and noise relevance between adjacent signals are analyzed in this paper, and a de-noising method based on regional averaging noise reduction is proposed after studying the pre-existing de-nosing method based on independent component analysis (ICA), the feasibility of the proposed de-noising method is dominated.Based on above theoretical research works, in the condition of weak radiation and low detection sensitivity, we reconstruct the object images detected by Hadamard transform encoding; blurred reconstructed object images are taken as experimental research targets, the proposed method based on RA are applied to de-noise in this paper. The experimental results show that compared to the traditional direct detection, the optimized signal reconstruction method can greatly enhance the SNR of weak radiation detection,and the proposed de-noising method based on RA has better performance and higher efficiency than the de-noising method based on ICA, which can further improve the SNR of reconstructed target images.
Keywords/Search Tags:Weak Radiation Detection, Hadamard Transform, Reconstruct of Signals, Regional Averaging
PDF Full Text Request
Related items