The spatially modulated snapshot imaging polarimeter (SMSIP) can measure two-dimensional state of polarization through using the spatial carrier to encode the Stokes vectors in a single image. The traditional reference calibration is not suitable for dynamic environment (such as temperature change), and the dynamic calibration of linear shear spatial modulation snapshot imaging is proposed in this paper. In the dynamic calibration used are two modified Savart polariscopes with shear along the same direction as the core modulation device. In addition, the thickness ratio of the two modified Savart polariscopes is 1∶2. The spatial modulation phase factor generated by the core modulation device can be deduced from the interferogram and thickness ratio so as to demodulate all the polarization information of the target. The significant advantage of the dynamic calibration is that the target detection and the system calibration are carried out simultaneously. The reference target is not required in the process. In this work, the detailed theoretical analysis of the dynamic calibration is presented. In addition, a laboratory experiment and numerical simulation are conducted to demonstrate the validity of the proposed dynamic calibration. The present study provides a new idea for calibrating the spatially modulated snapshot imaging polarimeter, and effectively promotes its applications in engineering under dynamic environment.
目前由于非均匀性技术的限制,红外焦平面阵列具有条纹非均匀性,这种非均匀性一般是一种固定的模式噪声,它会在原有非均匀性的基础上随着时间和温度的变化而变化.传统的残差计算忽略了非均匀性的特点,场景细节的变化会被误认为非均匀性而被过滤掉.提出了一种基于场景的条纹非均匀性校正算法.该方法采用非线性滤波的方法滤除单柱图像的非均匀性,并计算与原始图像的实际残差,然后利用前一帧的预测残差和实际残差得到当前残差.最后,根据场景鬼影抑制参数自适应校正计算增益系数和偏置系数.实验结果表明,提出的算法在200帧测试图像中滤除了61.3%的非均匀性,与其他自适应校正算法相比,非均匀性校正效果更明显,并且在一定程度上有效地保护了图像边缘,收敛速度快,质量高,能有效去除列条纹和非均匀随机噪声.
目前使用回音壁模式增强稀土离子发光的研究,多是采用先制备回音壁模式腔,后引入稀土离子的方式.由于稀土离子是后添加的,这种工艺常常会有表面粗糙和稀土元素引入不均匀等问题,使回音壁模式的增强效果降低.本文提出一种新的稀土掺杂回音壁模式微球腔的制备工艺,将稀土元素的引入过程提到了微腔的制备前.实验表明:相比于此前报道的先进行微腔制备,而后进行稀土掺杂的微球,所制微球的表面更光滑自然,稀土掺杂更均匀,回音壁模式对于发光强度的增强效果也更明显.该工艺可推动稀土掺杂回音壁模式微腔在传感和上转换激光器等领域的应用.
针对工业内窥镜在缺陷检测时遇到较大、较长的缺陷无法一次性完成识别、测量的问题,提出一种基于改进SIFT算法的图像拼接方法.在建立尺度空间的过程中,使用灰度阈值的方法,缩小了检测范围,解决SIFT算法在使用过程中计算量过大的问题.使用YOLOv3算法完成对缺陷区域的识别,并通过像素尺寸与实际尺寸的关系,实现对拼接后缺陷真实长度的测量.借助工业内窥摄像头,开展了缺陷拼接、识别和测量实验.实验结果表明,利用改进的SIFT算法对缺陷图像的拼接识别准确率可达 92.83%;角点检测速度提高了 44.3%;拼接后的缺陷测量相对误差可达0.63%.所提的方法具有较高的识别准确率和测量精度,可广泛应用于工业内窥镜的缺陷测量中.
随着元宇宙概念的兴起,虚拟现实头戴显示器得到了快速发展.基于偏振折反射的折叠光路可以减小系统总长和镜片数量,使虚拟现实头戴显示器更加轻薄.在不增大系统尺寸和重量的同时,为进一步增大虚拟现实头戴显示器的视场角,将菲涅耳透镜和折叠光路相结合,得到出瞳距为14 mm,出瞳直径为(4±3)mm,视场角为120°,系统总长小于21 mm的大视场角轻薄型光学系统.设计的系统满足虚拟现实头戴显示器大视场角和轻薄化的需求.
