Fluorescence imaging in the second near-infrared window (NIR-II, 900–1880nm) with less scattering background in biological tissues has been combined with the confocal microscopic system for achieving deep in vivo imaging with high spatial resolution. However, the traditional NIR-II fluorescence confocal microscope with separate excitation focus and detection pinhole makes it possess low confocal efficiency, as well as difficultly to adjust. Two types of upgraded NIR-II fluorescence confocal microscopes, sharing the same pinhole by excitation and emission focus, leading to higher confocal efficiency, are built in this work. One type is fiber-pinhole-based confocal microscope applicable to CW laser excitation. It is constructed for fluorescence intensity imaging with large depth, high stabilization and low cost, which could replace multiphoton fluorescence microscopy in some applications (e.g., cerebrovascular and hepatocellular imaging). The other type is air-pinhole-based confocal microscope applicable to femtosecond (fs) laser excitation. It can be employed not only for NIR-II fluorescence intensity imaging, but also for multi-channel fluorescence lifetime imaging to recognize different structures with similar fluorescence spectrum. Moreover, it can be facilely combined with multiphoton fluorescence microscopy. A single fs pulsed laser is utilized to achieve up-conversion (visible multiphoton fluorescence) and down-conversion (NIR-II one-photon fluorescence) excitation simultaneously, extending imaging spectral channels, and thus facilitates multi-structure and multi-functional observation.

乳腺组织自体荧光光谱法通常采用420～512 nm波段的光源，本次实验采用波长为405 nm的激光源获取乳腺组织切片的自体荧光光谱，避免了活体检测时脂肪和血液对光谱的干扰。荧光光谱是多种发光基团的叠加光谱，基于荧光光谱对乳腺癌细胞的生化信息进行定性和定量分析时，光谱特征峰不易分辨。本文提出了高斯函数拟合乳腺组织切片荧光光谱的方法，该方法可有效分离重叠峰。与正常组织相比，癌变组织的荧光特征峰出现了明显的红移现象。通过面积比值法分析了正常与癌变乳腺组织中荧光物质的差异，癌变组织在517 nm和635 nm处的峰面积与492 nm处的峰面积之比(A₅₁₇/A₄₉₂和A₆₃₅/A₄₉₂)分别是正常组织的2.4～5.5倍和5.4～8.5倍，可作为诊断乳腺癌的标准。最后，本文分析了支持向量机(SVM)算法对乳腺组织荧光光谱进行分类的可行性，分类准确率为87.50%，为乳腺癌的快速诊断提供了新方向。

为了有效复原雾霾天气下退化的图像，文章提出了一种自适应线性透射率估计去雾算法。建立有雾图像与无雾图像最小值通道之间的线性变换模型；利用有雾图像的混合通道得到自适应参数，结合自适应参数和线性变换模型估计出透射率，通过有雾图像的最小值通道构造高斯函数来补偿估计明亮区域透射率，提升该区域透射率的准确度，再使用交叉双边滤波器消除纹理效应得到优化透射率；最后，结合大气散射模型复原出无雾图像。实验结果表明，该方法有效降低了时间复杂度，且复原的图像细节明显，明亮度适宜。

Betatron radiation from laser wakefield accelerated electrons and X-rays scattered off a counter-propagating relativistic electron bunch are collimated and hold the potential to extend the energy range to hard X-ray or gamma ray band. The peak brightness of these incoherent radiations could reach the level of the brightest synchrotron light sources in the world due to their femtosecond pulse duration and source size down to a few micrometers. In this article, the principle and properties of these radiation sources are briefly reviewed and compared. Then we present our recent progress in betatron radiation enhancement in the perspective of both photon energy and photon number. The enhancement is triggered by using a clustering gas target, arousing a second injection of a fiercely oscillating electron bunch with large charge or stimulating a resonantly enhanced oscillation of the ionization injected electrons. By adopting these methods, bright photon source with energy over 100 keV is generated which would greatly impact applications such as nuclear physics, diagnostic radiology, laboratory astrophysics and high-energy density science.

针对通过局部相似假设估计透射率，景深突变边缘出现Halo效应问题，提出一种基于边缘保持的自适应高斯衰减图像去雾算法。该方法从大气散射模型出发，引入大气幕亮度，将场景透射率的估计等效为大气幕亮度的估计。通过边缘检测算子提取边缘信息，分离边缘区域与非边缘区域，利用邻域内像素点的空间邻近度构建自适应高斯函数对非边缘区域进行平滑衰减，从而获得最优效果。通过大量实验对所提方法进行验证，结果表明复原的图像整体平滑，细节明显，有效地消除景深突变处的Halo效应,并且在客观评价中也体现出了优势。