星点坐标辅助的全天区三角形星图识别算法

随着当前星敏感器视场（FOV）的增大,探测能力的提高,一帧图中拍摄到的恒星更多。但是受星敏感器光谱范围的限制及空间环境干扰影响,星等测试精度一般不高于0.2 mV。为了充分发挥当前星敏感器视场和探测能力的优势,并避免星等误差的影响,提高全天区星图识别算法在线应用的适用性,提出了一种星点坐标辅助的全天区三角形星图识别算法。该方法采用"全局初步搜索识别—局部精细匹配验证—最优结果选取"的算法思想。首先,根据星敏感器探测到的极限星等范围构建导航星表,选取亮星构建角距星表,既确保了星表的完备性,又有利于充分利用星敏感器的探测能力。然后,在三角形约束条件下进行角距匹配识别,得到一个或多个导航三角形,在该识别环节提出了非线性矢量法查找星表,既提高了定位精度,又能采用单精度数据类型降低存储空间。最后,提出局部天区星点坐标匹配算法进一步消除冗余匹配,同时又识别出视场内更多的观测星,有利于提高识别率和定姿精度。试验结果表明,与其他一些经典的星图识别算法相比,所提算法在识别率和星表容量方面更有优势。识别率可达99.9%,且随着星等的增加,存储容量增加的最少。所提算法更加适于大视场、高星等敏感范围的星敏感器在线应用。     

一种改进的三角形星图识别方法

首先提出在直角坐标系下将天球球面均匀地分成486个分区的方法——"内接正方体法",实现导航星的快速检索。然后按照星对角距的大小升序排列,直接存储星对角距,并通过设置星对角距状态标识,完成星对角距的快速匹配搜索。最后,引入验证识别环节有效地解决观测三角形与匹配的导航三角形产生的冗余问题。仿真实验表明,改进的三角形星图识别方法对星点位置噪声和星等(亮度)噪声均具有较强的抗干扰能力,且在识别时间和存储空间的需求方面均比传统的三角形算法也具有较大的优势。     

P向量实现快速星图识别的方法

 提出了一种快速的星图识别方法。该方法识别时提取出描述观测三角形的一维特征量P，根据该值快速检索到对应的导航三角形，减少了计算过程。由于获得星点位置的时候存在误差，所以将一定误差范围内的三角形均作为待选三角形考虑。上述识别后引入了验证识别环节，保证识别的准确性。随机实验统计表明，同等星点位置噪声方差下，该方法和栅格算法相比识别率提高约1%，识别时间约为后者的1/5。     

亮星辅助下基于坐标转换的快速星图识别方法

数字天顶仪作为一种地面使用的星敏感器，主要用于高精度定位。为提高仪器的工作效率需要对星图识别的快速性进行研究。通过对恒星像点理论坐标与图像坐标的分析，构建坐标转换模型。依据星表中恒星的分布及星等筛选出视场范围内的亮星，并构建导航星表。结合导航星表完成3颗亮星的准确识别，在识别亮星的基础上解算坐标转换模型的参数。通过构建的坐标转换模型对视场范围内的恒星进行坐标转换，将转换后的星点坐标与提取的星点图像坐标进行匹配完成星图的识别，这样能够提高星图识别的快速性。实验数据表明：在保证识别星点数量的基础上，采用亮星辅助下基于坐标转换的星图识别方法使时间缩短为改进三角形星图识别算法的五分之一。     

一种改进的星图降噪算法研究

在星敏感器的使用过程中，由于外界环境的影响及传感器自身的限制，拍摄出来的星图不可避免地存在一些噪声，因此对星图进行去噪处理是一项非常重要的工作。针对传统高斯模板滤波存在的引入邻域噪声、无法自行根据星图特性修正等造成去噪效果不好的问题，提出了一种改进的星图降噪算法。该方法在滤波前先进行坏点剔除工作，并采用高斯低通滤波与高通滤波结合的方式对图像进行处理，在抑制噪声的同时有效地保留了星点信号。通过阐述星敏感器的工作原理，分析星图的噪声特性，对星图滤波去噪算法进行研究，并进行模拟星图影像提取星点坐标实验。结果表明：使用该算法进行滤波比传统的高斯滤波算法提取的质心坐标精确度更高，较传统方法横坐标提高0.00538个像素点，纵坐标提高0.0077个像素点，证实了图像处理算法的有效性。     

基于帧间角距匹配的跟踪模式星图识别

为了提高星敏感器的跟踪匹配速率,提高星敏感器姿态更新率,本文提出了一种基于帧间角距匹配的跟踪模式星图识别方法.该方法充分利用上一时刻的角距匹配信息构建了一个实时跟踪星库,并把当前时刻的角距信息在实时跟踪星库中进行匹配识别,识别成功后利用识别时用到的信息对实时跟踪星库进行更新.仿真实验表明,该方法具有匹配时间短、匹配成功率高的优点.     

一种改进的快速全天星图识别算法

针对三角形算法中模式识别率低、易造成误识别的问题,提出了一种改进的三角形算法。 首先,该算法通过对全天区导航星进行筛选,缩小星库存储空间;同时,在识别过程中对找到的三 边角距集合采用设置状态标识的方法,加快对探测星三角形的识别速度;在三角形匹配识别结果 不唯一时,引入其他观测星进行验证,确定唯一的匹配对应;在识别结束后,应用投影验证检验识 别结果。仿真结果表明,该算法能够有效地节省存储空间,缩短全天星图识别时间,与传统的三角 形算法相比具有较大的优势。     

大动态干扰下基于光学图像的自主导航技术

针对高超声速飞行器在平流层内飞行过程中自主天文导航时,气动光学效应及运动模糊等大动态干扰严重影响观星和高精度导航的问题,提出一种基于正则化思想的高超星图半盲复原改进算法。该算法首先借助已有的高超星图模糊核提取算法获得先验模糊核,为复原提供便利条件。接着结合天文图像灰度及梯度的稀疏先验分布特性,设计了一种针对高超星图的正则化盲复原模型,并结合先验模糊核所提供的信息,构建了改进的半盲复原方法,能在先验模糊核精度较低或缺失时转化为盲复原方法,提高了算法的鲁棒性。将本算法与传统星图复原算法、其他最新正则化复原算法进行星图复原与导航效果比较,数值仿真结果显示本算法的复原效果更佳,峰值信噪比（PSNR）达到36.97,并且质心提取成功率达到99.23%,能够明显改善星图识别的成功率,从而大幅提升高超声速飞行器在平流层中的大动态干扰下的自主导航能力。     

一种基于Sobel算子的星敏感器星图预处理方法

星敏感器在太空环境下工作时容易受到杂散光的干扰,对星图进行二值化后直接采用质心法难以从星图中提取恒星星像坐标.针对上述问题,提出了一种基于Sobel算子的星敏感器星图预处理方法,用于提取星像坐标.首先采用Sobel算子计算星图中所有像素的梯度,保留所有大于阈值的像素梯度作为新的星图,然后以新的星图作为样板通过质心法从原始星图中提取恒星星像.由于杂散光干扰信号在星图中表现为缓慢信号,这些信号计算的梯度很小,因此通过Sobel算子计算后新的星图只有恒星星像.该方法能够消除星图中的杂散光干扰信号,使星图中只保留恒星星像信息,提高恒星星像坐标提取精度,具有较好的鲁棒性.     

寡星条件下的半球谐振陀螺与星敏感器 组合姿态测量算法

针对半球谐振陀螺与星敏感器松组合系统在寡星条件下无法正常工作的问题,采用新的星图识别算法和新的数据融合观测方程,使星敏感器在观测到的导航星数量为1或2颗的情况下完成星图识别,从而能够完成组合测姿。     

Modified Grid Algorithm for Noisy All-Sky Autonomous Star Identification

Grid algorithm is a classical algorithm for all-sky autonomous star identification. However its recognition rate decreases rapidly when the position and magnitude noise increases in the star image. In order to overcome this problem, a modified grid algorithm called the "elastic gray grid algorithm" is proposed. Gray pattern and elastic template matching are introduced into the grid algorithm to enhance its ability to tolerate the position and magnitude noise. Experimental results show that the modified grid algorithm performs better compared with the original one. The elastic gray grid algorithm reached up to a 100% recognition rate under random positional noise in our test bed. Furthermore it increases the original recognition rate from 69.86% to 97%, under the same condition, for both the position and the magnitude noise.     

Geometric voting algorithm for star trackers

We present an algorithm for recovering the orientation (attitude) of a satellite-based camera. The algorithm matches stars in an image taken with the camera to stars in a star catalogue. The algorithm is based on a geometric voting scheme in which a pair of stars in the catalogue votes for a pair of stars in the image if the angular distance between the stars of both pairs is similar. As angular distance is a symmetric relationship, each of the two catalogue stars votes for each of the image stars. The identity of each star in the image is set to the identity of the catalogue star that cast the most votes. Once the identity of the stars is determined, the attitude of the camera is computed using a quaternion-based method. We further present a fast tracking algorithm that estimates the attitude for subsequent images after the first algorithm has terminated successfully. Our method runs in comparable speed to state of the art algorithms but is still more robust than them. The system has been implemented and tested on simulated data and on real sky images.     

一种基于星座匹配的全天自主星图识别算法

讨论了一种新的无须任何先验知识的星图识别算法。通过对导航星座数据库进行调整和最大限度地利用了已识别基元所提供的匹配信息,使得算法的计算量和存储容量与传统的三角形匹配算法相当。ＭｏｎｔｅＣａｒｌｏ 模拟表明,在星等误差０．５ 星等（３σ）,位置误差１０ 角秒（１σ） 时,算法的识别成功率接近１００％ 。     

Rapid Star Tracking Algorithm for Star Sensor

A rapid star tracking algorithm is proposed. In order to speed up the tracking, three techniques includng parallel star centroiding, sorting, and star catalogue partition are designed for three time-consuming portions in tracking algorithms. The parallel star centroiding is implemented with Field Programmable Gate Array (FPGA) which avoid image storage and transmission. Update rate of star sensor is improved. Sorting star coordinates in star image plane, and then matching, which avoid matching between stars with a long distance in image plane. Star catalogue partition divides the celestial sphere into many small partitions. In star mapping, guide stars are searched in the partitions near the direction of star sensor's boresight is not in the whole celestial sphere and therefore reduced the total number of searched guide stars. The software and hardware performance of tracking algorithms are simulated. Tracking robustness and the tracking speed comparison are tested in the software simulation. In hardware tests, the tracking time in every step is obtained.     

Star recognition algorithm for APS star tracker: oriented triangles

An efficient star pattern recognition algorithm is presented. The purpose of this algorithm is to make sure of the compatibility of the software and the imaging sensor noise level. The new CMOS APS sensors have not currently reached the same accuracy as the former CCD sensors in position as well as in magnitude determination, especially in the dynamic stages. This algorithm allows the system to recognize the star pattern 20% faster than with reference algorithms. No false recognition has been noticed. Used databases have a size 5 to 10 times smaller, depending on other reference algorithms. Oriented triangles are used to compare the measured star pattern with the catalogue stars. The triangle's characterization criteria propose several solutions in a first time. A unique solution is selected by means of identification and validation methods in a second time. First results, presented hereinafter, are very encouraging, and this algorithm may be used in the future APS star trackers. APS star tracker robustness is significantly enhanced by this method during the critical navigation phases     

Star-configuration searching for satellite attitude computation

A star pattern recognition algorithm is described which recovers attitude information by processing data obtained from a modular star sensor (MOSS) mounted on a spacecraft information is obtained by matching the stars detected by the MOSS with those inside a stellar catalogue. The system performance has been evaluated both from the point of view of computation time and error probability. The effect of errors in the MOSS measurements and multiple matches cases are also discussed