An Effective Wavelength Detection Method Based on Echelle Spectra Reduction
YIN Lu1, 2, Bayanheshig1*, CUI Ji-cheng1, YANG Jin1, 2, ZHU Ji-wei1, YAO Xue-feng1
1. Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun 130033, China 2. University of Chinese Academy of Sciences, Beijing 100049, China
Abstract：Echelle spectrometer with high dispersion, high resolution, wide spectral coverage, full spectrum transient direct-reading and many other advantages, is one of the representative of the advanced spectrometer. In the commercialization trend of echelle spectrometer, the method of two-dimension spectra image processing is becoming more and more important. Currently, centroid extraction algorithm often be used first to detect the centroid position of effective facula and then combined with echelle spectrum reduction method to detect the effective wavelength, but this method is more difficult to achieve the desired requirements. To improve the speed, accuracy and the ability of imaging error correction during detecting the effective wavelength, an effective wavelength detection method based on spectra reduction is coming up. At the beginning, the two-dimension spectra will be converted to a one-dimension image using echelle spectra reduction method instead of finding centroid of effective facula. And then by setting appropriate threshold the one-dimension image is easy to be dealing with than the two-dimension spectra image and all of the pixel points stand for effective wavelength can be detected at one time. Based on this new idea, the speed and accuracy of image processing have been improved, at the same time a range of imaging errors can be compensated. Using the echelle spectrograph make a test applying this algorithm for data processing to check whether this method is fit for the spectra image processing or not. Choosing a standard mercury lamp as a light source during the test because the standard mercury lamp have a number of known characteristic lines which can be used to examine the accuracy of wavelength detection. According to experimental result, this method not only increase operation speed but improve accuracy of wavelength detection, also the imaging error lower than 0.05 mm (two pixel) can be corrected, and the wavelength accuracywould up to 0.02 nm which can satisfy the requirements of echelle spectrographfor image processing.
