Compact Offner Imaging Pectrometer And Spectral Data Processing

The spectral imaging technology, by combining the imaging and spectroscopy technologies, can provide both spatial and spectral information of objects simultaneously. Through data processing, the acquired data can be constructed to a three-dimensional data cube which contains two-dimensional spatial information and one-dimensional spectral information. This technology has been widely used in satellite remote sensing, forestry, geology, oceanography, medicine, manufacturing, ecology and military, etc.In this paper, technical specifications for imaging spectrometer are figured out according to the application requirements of precision agriculture. Then a compact Offner imaging spectrometer and its information system is developed. The calibration, test, noise analysis, noise improvement and unmanned aerial vehicles remote sensing experiments are also performed. The imaging spectrometer consists of four parts: imaging optical system, entrance slit, Offner-type optical system with a convex grating and image sensor. The software is developed for spectrometer control, data acquisition and data processing. The data acquisition and storage software has the characteristics of high speed transmition and storage of massive data. Besides, the software provides the function of system response compensation, image stacking and display.After the Offner imaging spectrometer is finished the optical commissioning, its performance evaluation, wavelength calibration and radiometric calibration are performed in the laboratory. The results show that the spectrometer satisfies the design requirements. Then the spatial distribution of Hg hollow cathode lamp is measured. An equivalent electron method is applied in order to estimate the SNR precisely, and the theoretical SNR is obtained. Overall noise in the Offner imaging spectrometer consist of noise in the time domain and noise in the spatial domain, they are irrelevant to each other, the total noise powrer of system is equal to mean root of time domain noise and spatial domain noise. The various noise models are analyzed and the key parameters of image sensor are tested. The SNR calculation result is verified by the radiometric calibration experiment. The SNR can be improved by data denoise processing, which can be tested by experiment results.The spectrometer is tested by the unmanned aerial vehicles for remote sensing experiment. The results on the helicopter and fixed-wing UAV are given. The influence of stability of UAV to image quality is analyzed. The spectral image improvement for a clearer image is proposed.