The discovery of aggregation-induced emission (AIE) effect provides opportunities for the rapid development of fluorescence imaging-guided photodynamic therapy (PDT). In this work, a boron dipyrromethene (BODIPY)-based photosensitizer (ET-BDP-O) with AIE characteristics was developed, in which the two linear arms of BODIPY group were linked with triphenylamine to form an electron Donor–Acceptor–Donor (D–A–D) architecture while side chain was equipped with triethylene glycol group. ET-BDP-O was able to directly self-assemble into nanoparticles (NPs) without supplement of any other matrices or stabilizers due to its amphiphilic property. The as-prepared ET-BDP-O NPs had an excellent colloid stability with the size of 125 nm. Benefiting from the AIE property, ET-BDP-O NPs could generate strong fluorescence and reactive oxygen species under light-emitting diode light irradiation (60mW/cm2). After internalized in cancer cells, ET-BDP-O NPs were able to emit bright red fluorescence signal for bioimaging. In addition, the cell viability assay demonstrated that the ET-BDP-O NPs exhibited excellent photo-cytotoxicity against cancer cells, while negligible cytotoxicity under dark environment. Thus, ET-BDP-O NPs might be regarded as a promising photosensitizer for fluorescence imaging-guided PDT in future.The discovery of aggregation-induced emission (AIE) effect provides opportunities for the rapid development of fluorescence imaging-guided photodynamic therapy (PDT). In this work, a boron dipyrromethene (BODIPY)-based photosensitizer (ET-BDP-O) with AIE characteristics was developed, in which the two linear arms of BODIPY group were linked with triphenylamine to form an electron Donor–Acceptor–Donor (D–A–D) architecture while side chain was equipped with triethylene glycol group. ET-BDP-O was able to directly self-assemble into nanoparticles (NPs) without supplement of any other matrices or stabilizers due to its amphiphilic property. The as-prepared ET-BDP-O NPs had an excellent colloid stability with the size of 125 nm. Benefiting from the AIE property, ET-BDP-O NPs could generate strong fluorescence and reactive oxygen species under light-emitting diode light irradiation (60mW/cm2). After internalized in cancer cells, ET-BDP-O NPs were able to emit bright red fluorescence signal for bioimaging. In addition, the cell viability assay demonstrated that the ET-BDP-O NPs exhibited excellent photo-cytotoxicity against cancer cells, while negligible cytotoxicity under dark environment. Thus, ET-BDP-O NPs might be regarded as a promising photosensitizer for fluorescence imaging-guided PDT in future.

肿瘤是严重威胁人类健康和社会发展的重大公共卫生问题之一。针对肿瘤的治疗，外科手术切除仍是最普遍、最理想的策略。目前，外科医师在术中主要通过肉眼观察、超声等方法确定肿瘤边界、残余病灶以及微小转移病灶，然而，这些方式存在手术切缘肿瘤易残留、难以实时发现微小转移灶等问题，导致肿瘤术后复发、转移率高，严重影响病人的预后和远期生存。荧光手术导航的快速发展为解决这一问题提供了新的技术支撑。本文聚焦于荧光手术导航用荧光探针，重点对荧光手术导航系统和各类有机、无机近红外二区荧光分子探针进行系统阐述，分析其临床转化与应用面临的瓶颈问题，探讨可能的解决思路，以期为国内外相关领域的研究提供思路与参考。

在摄像机标定过程中,较差的光照条件和镜头畸变会造成棋盘格角点漏检和角点冗余。分析了角点的灰度分布特性,提出一种基于角点灰度分布特征的棋盘格角点检测算法。为确保在较差光照条件和镜头畸变下,棋盘格图像角点不漏检,算法首先利用角点的灰度分布特性提取候选角点;然后通过迭代的方式提高候选角点的精度并再次结合棋盘格角点的灰度分布特性剔除候选角点中非角点处的伪角点,避免棋盘格角点冗余;最后通过角点处的邻近点合并获得最终的棋盘格角点坐标。实验结果证明,在较差的光照条件和镜头畸变条件下,本算法角点无漏检和冗余。将该算法提取的棋盘格角点应用于摄像机标定,结果显示重投影误差的均方差在0.1 pixel范围以内,优于其他算法。

光响应纳米基因载体可借外界光源实现基因的定时、定点释放，已逐渐成为非病毒基因载体研究的热点。根据响应光波长的不同，可将光响应纳米基因载体分为紫外/可见光响应的纳米基因载体和近红外光响应的纳米基因载体。综述了这两类载体材料当前的研究进展，阐述了其响应机理和释放过程，并总结了光响应纳米基因载体在构建过程中亟待解决的问题。

针对单通道的同调制方式、同调制参数的时频重叠信号的盲分离问题，提出了一种新的卷积混合盲分离算法。首先，基于时间分集的过采样技术的信道多样性性质，将单通道混叠的MISO模型转化为多通道混叠的MIMO模型进行研究。然后，结合MIMO模型混叠矩阵的结构特点，讨论了混叠信号分量的时延差对MIMO模型混叠矩阵结构和混叠矩阵秩的影响。最后，数学推导证明：当信号分量时延差为符号速率的整数倍时，从信号时延的角度来讲混叠信号不具有盲可分离性，并通过仿真实例予以验证。

基于大场景合成孔径雷达（SAR）图像序列，研究了一种针对多类慢动车辆目标的识别与跟踪方法，采用先识别、再跟踪的思路。提出了一种图像目标局部多分辨分析与多核分类器相结合的识别方法，实现了多类目标的快速特征提取和准确分类。根据相邻帧之间目标的对应关系，利用无偏卡尔曼滤波对目标的运动参数进行估计，并用实际测量值不断进行修正，实时获取目标的坐标、类型等信息，实现了复杂背景下地面多类慢动目标的高效跟踪。通过构建大场景合成孔径雷达序列图像进行仿真实验，证实了该方法具有快速和稳定的收敛性能，实时性较好，具有较高的跟踪精度。

优化设计了一种双环变迹镜。该变迹镜由同心的，交替排列的透光圆环和不透光圆环组成，经过双环变迹镜调制的光波入射至光学接收系统，后经光学成像系统将所接收的光波成像聚焦，数据采集与处理系统对电流信号进行采集和处理，得到沿测量路径的等晕角。这种变迹镜相对于国内现有的单孔径变迹镜，能够在所有高度很好的模拟孔径滤波函数W(z)≡cz5/3，所得等晕角相对误差明显较小，可以实现对等晕角的高精度测量。

为了解决机器人同时定位，地图构建与目标跟踪(SLAMOT)过程中的多源，异构传感器空间一致性观测问题，提出了基于信息融合的摄像机与激光测距传感器联合标定优化方法。完成基于误差传播公式的激光扫描点图像平面投影不确定范围判定，并利用协方差交集算法实现基于运动物体检验方法和基于Camshift方法的图像坐标系下目标状态融合。在此基础上，利用目标图像平面投影方向误差构造目标函数，通过非线性优化方法实现摄像机与激光测距仪标定参数优化。实验验证了设计方法能有效提高目标跟踪以及多传感器参数标定的准确性。相关成果能够为基于多传感器信息融合的机器人同时定位，地图构建与目标跟踪滤波方法研究提供观测值支持。

为了提高景象匹配中基准图制备的效率和可靠性,对遥感影像图源的可匹配性进行分析,提出一种基于独立成分分析的遥感影像可匹配性度量方法.利用独立成分分析提取纹理基函数,并计算其概率分布,确定可匹配区域；然后定义可匹配区域所占面积比例指标、离散性指标和稳定性指标；最后,利用三个指标构建遥感影像可匹配性度量模型.实验表明,该遥感影像可匹配性指标与实际匹配概率相关性强,提高了基准图制备的可靠性,能够满足景象匹配图源筛选的实际需求.