Research On The Signal Detection Of Optical Encoding Image And Its Image Processing

Hadamard transform is a multiplexed optics modulation technique which is similar to Fourier transform. This technique uses binary orthonormal transform based on Hadamard matrix, has multiplexing and high input advantages. Hadamard transform multiplexed imaging technique provides some advantages for analytical instrument: Fellgett's advantage, by which an improvement in the signal-to-noise ratio can be obtained, the multiplexed spatial imaging ability and so on. The non-probe near-field optical scanning microscope based on the theory of Hadamard transform imaging technique , provides another detecting way for non-radiation field. That is encoding detection. On the principle of encoding detection, the thesis discusses the basic theory of encoding detection and the mathematical model of encoding detection. At the same time, as viewed from the practical application, the thesis makes certain that the cycle S matrix is feasible in the encoding detection. Based on the theory of encoding detection, a set of experiment system is designed. This experiment system consists of the optical lighting , the design and manufacturing of the encoding mask, the encoding driving of the encoding mask and the signal collection and so on. For the software systems, the fast decoding program is compiled by studying the fast Hadamard transform(FHT) and the πtransform in the thesis. In view of the existing noises in the encoding detection and the practical applications of the non-probe near-field optical scanning microscope, the processing of imagine software is preliminarily built to achieve the following result: interpolation amplification of image, average noise reduction of spatial area, filter of low-pass of frequency domain, filtering of middle value, grade sharpening of image, Laplace sharpening of image, filter of high-pass of frequency domain, enhancement of image's gray transformation and enhancement of the adjustment of image's gray histogram, edge detection and profile picking-up of image, finally saving the image file by "BMP"format.