Conceptional design of the laser ion source based hadrontherapy facility

A laser ion source (LIS), which can provide a carbon beam with highly stripped state (C⁶⁺) and high intensity (several tens mA), would significantly change the overall design of the hadrontherapy facility. The proposed LIS based hadrontherapy facility has the advantages of short linac length, simple injection scheme, and small synchrotron size. With the experience from the DPIS and HITFiL projects that have been conducted in IMP, a conceptional design of the LIS based hadrontherapy facility will be presented, with special attention given to APF type IH DTL design and simulation.     

液-固扩散法制备可交联聚乙烯

基于小分子在聚合物中的扩散原理,采用液-固扩散法,将液态小分子交联剂、抗氧剂从固态低密度聚乙烯(LDPE)颗粒表面扩散到内部,实现小分子助剂与聚合物的均匀、温和混合,得到品质优良的可交联聚乙烯.采用DSC、FTIR等手段,分析不同扩散条件时LDPE颗粒表层和芯层的交联热焓、特征吸收峰强度的变化规律,探索小分子助剂向聚合...     

基于斑马鱼图像特征的鱼群检测算法北大核心CSCD

计算机视觉方法越来越多地应用于斑马鱼的群体行为研究;但是,由于斑马鱼游动过程形体变化大,遮挡多,准确与鲁棒地检测出斑马鱼仍然是一件非常具有挑战性的问题。为了解决该问题,提出一种基于斑马鱼图像特征的鱼群检测算法。首先通过分析目标特性,提出使用鱼头和鱼尾替代全鱼的检测方法,解决了传统整鱼检测在鱼群交叉遮挡时失效的难题;然后基于斑马鱼图像特征自动构建训练集,避免了深度学习手动标注的费时费力问题。通过对实际斑马鱼视频进行处理验证,与现有的算法相比,本文提出的方法在标注率、召回率(recall,R)与遮挡检测率(occlusion detection rate,ODR)等性能指标上有更好的实验效果。其中,在标注性能方面,本文提出的自动标注方法在总标注率上达到87.40%;在训练集效果方面,本文自动标注算法结合人工校正在标注时间上相比于人工标注方法减少93.11%,均值平均精度(mean average precision,mAP)达到79.80%;在目标检测方面,在目标遮挡率为42.72%的情况下,本文检测算法能够获得82.0%的召回率及58.02%的遮挡检测率。