Hyperspectral imaging (HSI) acquires a three-dimensional dataset called hypercube, with two spatial dimensions and one spectral dimension. HSI is an emerging imaging modality for various medical applications, especially in disease diagnosis of early and biomedical research based on their unique spectral signatures. A visible-near-infrared HSI for microspectroscopy is designed, the measured spectrum is from 450 nm to 900 nm, which is sampled by 256 spectral channels. The spatial resolution is 1.25μm.The modulation transfer function (MTF) value of full spectrum and full field of view is close to the diffraction limit. At last, this microspecroscopy have been fabricated and preliminary tests have been implemented. The results indicated this visible-near-infrared hyperspectral microscopy optical systems have excellent optical performances. This hyperspectral microscopy will be well developed and used in the life sciences fields.
The study of underwater in-situ detection is an important research trend in underwater detection. In view of the design requirements of the spectral data acquisition system for underwater in-situ detection, this paper used the software and hardware co-design method, from two aspects of software and hardware. A prototype system of spectral data acquisition system based on Xilinx Zynq chip and linear array CCD detector was designed and implemented. Through theoretical analyzing, experimental debugging and verification analyzing, the results shown that the system could collect and store the spectral data in real time. It also had the characteristics of low noise and had a small electronic structure, which laid a foundation for the spectral data acquisition of underwater in-situ detection.
In case of light absorption and diffusion, the clarity of the underwater images are degraded. The color of the underwater image is distortion. In order to improve the quality of the original underwater image quality, a method based on RGB channels histogram equalization is proposed. The method improves the contrast of the underwater image based on strength the histogram of the RGB channels respectively. The real-world experiment confirms the effectiveness of the method.
High concentration of nitrate will cause many problems, such as water eutrophication and compromise of human beings’ health. A fast and stable approach was applied to predict nitrate concentration in solutions using the dual optical active correction continuous spectrum analyzer designed by our research group. Firstly, standard normal variate (SNV) was used to pretreat the spectral data. Then characteristic wavelengths of spectral curve were selected by using successive projections algorithm (SPA) and genetic algorithm (GA) respectively. Finally, partial least-squares regression (PLSR) was applied to build the spectral prediction model to predict nitrate concentration, and coefficient of determination (R<sup>2</sup>) and root mean square error of prediction (RMSEP) were introduced as the evaluation indicators of prediction models. For SNV-GA-PLS model, R<sup>2</sup>=0.9966 and RMSEP=0.1712, which outperformed than SNV-UVE-SPA-PLS model (R<sup>2</sup>=0.9896, RMSEP=0.3952). It demonstrated that he model which selects spectral characteristic wavelengths by GA can decrease the complexity of prediction model building and ensure the accuracy as well. Hence, SNV-GA-PLS model can be used to predict nitrate concentration in water with quick and steady performance.
Spectral imaging technology has made great achievements in applications of earth observation and space target detection, with the further development of research, the requirement that People tend to get more material properties about target is also improving rapidly, so getting more characters of the target is continuous pursuit goal for the instruments of optical remote sensing. Polarization is one of the four main physical properties of light including intensity, frequency and phase . It has very important significance for remote sensing observations such as improving the accuracy of target recognition. This paper proposes on a spectropolarimeter system based on Sagnac interferometer, and introduces the main aspects related to System components, working principle, optical design, adaptive spectrum extraction algorithm, state of polarization extraction methods. Also get the data of polarization spectral imaging by using the instruments designed by the principle .By processing these data I have got the combined polarization image and target spectral curves, achieved a good result. It is a new attempt to obtain polarization spectral image by integrated measuring system. Then thoroughly solve the traditional shortcoming of spectropolarimeter, such as asynchronous detecting, poor stability and vibration, poor energy efficiency. It can be applied to many kinds of fields. Simultaneously the paper puts forward some relevant new points in the future research for this kind of principle.
We propose an approach to correct the data of the airborne large-aperture static image spectrometer (LASIS). LASIS is a kind of stationary interferometer which compromises flux output and device stability. It acquires a series of interferograms to reconstruct the hyperspectral image cube. Reconstruction precision of the airborne LASIS data suffers from the instability of the plane platform. Usually, changes of plane attitudes, such as yaws, pitches, and rolls, can be precisely measured by the inertial measurement unit. However, the along-track and across-track translation errors are difficult to measure precisely. To solve this problem, we propose a co-optimization approach to compute the translation errors between the interferograms using an image registration technique which helps to correct the interferograms with subpixel precision. To demonstrate the effectiveness of our approach, experiments are run on real airborne LASIS data and our results are compared with those of the state-of-the-art approaches.
Optical alignment machining is an effective method to ensure the co-axiality of optical system. The co-axiality accuracy
is determined by optical-centering accuracy of single optical unit, which is determined by the rotating accuracy of lathe
and the optical-centering judgment accuracy. When the rotating accuracy of 0.2um can be achieved, the leading error can
be ignored. An axis-determination tool which is based on the principle of auto-collimation can be used to determine the
only position of centerscope is designed. The only position is the position where the optical axis of centerscope is
coincided with the rotating axis of the lathe. Also a new optical-centering judgment method is presented. A system which
includes the axis-determination tool and the new optical-centering judgment method can enhance the optical-centering
accuracy to 0.003mm.