The Application Of Hyperspectral Imaging System In Biological Detection

In the field of biological detection, spectroscopic technology has been widely applied in the rapid, non-destructive and label-free detection of genes, molecules, proteins, organization, etc. As the increasing demand of the high throughput and high precision detection, the requirement for detection technique is essential. However, the traditional spectroscopic detection technology cannot meet the requirements, due to the limitation of their functions. Fortunately, hyperspectral imaging technology (HIT) can simultaneously measure the spectral information of all points throughout the sample area, overcoming the limitation of traditional spectroscopy, which can only measure the spectrum information of single point each time, This technique combined with the spatial information and spectral information of sample, meets the demands of high throughput and high precision biological detection.Since the photonic crystal (PC) with photonic band gap (PBG) has a unique optical modulation capability, it has been widely used in the field of biological detection. However, the conventional detection method of PC, such as spectrometer, could only capture the spectral information from one spot, limiting the use of PC. Fortunately, the introduction of HIT just makes up for the defects. In this thesis, we combined the PC and HIT, and developed the several detecting systems according to different structure of PCs for the biological detection. The main works are as follows:(1) This paper first simulated the light way of microscope kohler illumination, and calculated the incident angles between the incident light and the surface of PCBs, and then calculated the distribution of PBG on the different parts of PCBs, explaining the relationship between the PBG, the reflectance spectrum and fluorescence excitation spectrum. The results showed that the dark spots or dark rings appearing on the fluorescence image of PCBs resulted from the PBG structure, proving that the PBG could prevent the propagation of photon with specific frequency in PCs. This result helped the application of PCBs under microscope in the field of biologic detection,(b) A hyperspectral scanning system was developed to evaluate the structure of PC films. The analysis of illumination of this system was carried out to find the relationship between the reflection spectrum and the structure of PC films. Then, a preprocessing algorithm was proceeded to remove the effect of system noise on the spectral data. After that, a spectral peak extracting algorithm was conducted to calculate the distribution of PBG on PC films. The results showed that this system could evaluate the structural homogeneity of PC films, providing a tool for the fabrication and detection of large area and uniform PC films. This system provided a guarantee for the large-scale application of photonic crystal film sensors.(3) This paper had developed a set of hyperspectral imaging system, composed of a cooling CCD camera, a hyper-spectrometer, an inverted fluorescence microscope and a data processing software. This system designed a optimized local segmentation algorithm for the extraction of PCBs with weak fluorescent intensity, and obtained the coded information of PCBs, and then designed a decoding algorithm for the classification of different PCBs, and finally designed a fluorescence background correction algorithm for the calculation of fluorescence intensity. The practical application results showed that the classification accuracy is high and the calculation of fluorescence intensity is accurate. This paper showed that HIT could realize the multiple detection based on the PCBs array, laying the foundation for the detection of PCBs array based on the on-chip spectrometer.