复杂工况下视觉AGV导航标识带中心线的提取研究

针对光照不均匀、导航标识带破损、少量杂物干扰的复杂工况条件下,对视觉AGV采集到的路面导航标识带图片进行特征提取,采用灰度化、中值滤波进行了图像预处理,研究了Otsu法和迭代法等动态阈值分割算法以及形态学在图像分割中的应用,并对预处理图像作了分割对比实验。然后对分割后的图像进行边缘提取,分析了基于最小二乘法和Hough变换算法的直线拟合原原理,提出了Hough算法下基于边缘线的中心线拟合算法,并作了直线拟合对比试验。实验表明,复杂工况下,采用基于形态学和Otsu算法相结合的方法对图像进行分割,得到的二值图像边界更完整,效果更好,基于Hough变换算法较最小二乘法能更精确有效地提取出导航标识带中心线及其方程。     

基于Shape Context和尺度不变特征变换的多模图像自动配准方法

提出了基于修正的尺度不变特征变换（SIFT）特征提取和Shape Context特征描述算子相结合的多模图像自动配准算法,该算法利用修正的SIFT算法提取多模图像中的特征点,然后采用Shape Context算子描述特征点,利用特征点周围区域边缘点的梯度方向形成特征向量。采用欧氏距离作为匹配标准对多模图像中特征点进行初始匹配,然后通过RANSAC算法消除误匹配的特征点对,并采用最小二乘法计算仿射变换参数,最后通过仿射变换和双线性插值实现图像配准。对红外图像和可见光图像的配准实验结果表明了本算法的有效性和稳定性。     

双目视觉测量系统特征点提取与匹配技术研究

自主研发了一种光笔式双目立体视觉大工件尺寸测量系统,对测量系统中特征点的提取及其匹配技术进行了研究。测量系统采用 Canny 算子和 Zernike 矩相结合的算法实现椭圆光斑的亚像素边缘提取,根据得到的亚像素边缘点,采用基于最小二乘的曲线拟合法得到椭圆光斑的中心坐标。针对特征点的匹配问题,提出了一种基于位置约束的快速匹配方法。实验结果显示:所提方法能提取到椭圆光斑亚像素边缘,可精确计算出椭圆光斑中心坐标,匹配率达到95%以上。     

基于TDLAS技术的天然气中痕量硫化氢分析的PLS算法应用

基于可调谐二极管激光吸收光谱技术的光谱分析仪测量天然气中硫化氢浓度的过程中,各种轻烃和二氧化碳等背景成分对光谱产生干扰,导致多个吸收峰叠加,对提取吸收光谱真实特征造成影响.应用偏最小二乘法消除背景成分的干扰,设计了检测天然气中硫化氢气体的可调谐二极管激光吸收光谱实验系统,采用偏最小二乘法和最小二乘法模型,分别检测了天然气中0~50ppm硫化氢成分的直测光谱和差分光谱.偏最小二乘法算法的测量结果均优于最小二乘法算法,且偏最小二乘法算法对于直测谱的测量误差保持在±1ppm范围内,满足分析仪器2%的准确度要求.利用偏最小二乘法算法避免了最小二乘法所必须的大量参考光谱数据的存储,分析仪可省掉复杂的差分光谱系统,从而达到降低成本、提高系统鲁棒性和实时性的效果.     

局部牛顿插值法提高多狭缝自准直仪准确度

根据多狭缝自准直仪目标物形状特征及光电探测器成像特点,通过边缘检测算法确定经过高斯滤波处理的CCD图像的像元级边界.在此基础上应用局部牛顿插值法对CCD图像边缘位置附近进行亚像元细分,实现亚像元边缘检测;再结合最小二乘直线边缘拟合法进一步提高图像边缘检测准确度.经验证,该算法可达到0.13″的定位准确度.     

多尺度边缘检测方法提取近红外光谱信息特征的研究

利用小波变换中的多尺度边缘检测方法对甲硝唑药片近红外光谱进行了信息特征的提取。通过偏最小二乘法建立所提取的特征与甲硝唑浓度之间的数学模型,并进行精度分析。结果表明,小波变换多尺度边缘检测能够有效分离并提取光谱信息特征,降低噪声,为一种有效的光谱信息特征提取方法,为解决近红外光谱中光谱叠加严重和信息率低等难题提供了一种思路。     

光学测量中椭圆圆心定位算法研究

提出了一种椭圆圆心定位算法。先对图像中的椭圆目标进行粗定位,获取椭圆的粗定位信息,再用Canny算子对图像中的感兴趣区域进行边缘检测。根据椭圆的粗定位信息以及图像的边缘灰度分布特征,用高斯曲线拟合的方法求得精确的亚像素边缘点的坐标。针对亚像素边缘点中出现的“孤立点”以及噪声点,分别用曲率滤波和均值滤波的方法加以滤波,将滤波后的亚像素边缘点用最小二乘法加以拟合,求得精确的椭圆圆心数据。实验证明了该算法的定位精度和抑制噪声能力,通过测试算法运行时间,证明该算法具有很好的实时性。     

一种复杂海天背景下红外弱小目标检测算法

红外弱小目标的检测是红外搜索跟踪系统(IRST)中的一项关键技术,常用的目标检测算法存在受海杂波严重、虚警率较高等问题,分析了海天背景下红外图像的背景、小目标的特征,提出了一种海杂波背景下的红外小目标检测算法。首先统计图像的行均值和梯度,用最小二乘法拟合出海天线,然后利用形态学算子抑制图像背景,并采用自适应阈值将图像二值化,最后分析图像的梯度特征,抑制海杂波和云层的干扰。实验结果表明,该方法能精确地提取海天线,稳定地提取海天背景下的弱小目标,虚警率低于5%,目标检测概率超过97%。     

局部牛顿插值法提高多狭缝自准直仪准确度

根据多狭缝自准直仪目标物形状特征及光电探测器成像特点,通过边缘检测算法确定经过高斯滤波处理的CCD图像的像元级边界.在此基础上应用局部牛顿插值法对CCD图像边缘位置附近进行亚像元细分,实现亚像元边缘检测;再结合最小二乘直线边缘拟合法进一步提高图像边缘检测准确度.经验证,该算法可达到0.13″的定位准确度.     

基于ARM的视觉导航AGV图像处理方法研究

为了实现更加简单高效的标识线图像的检测与处理,提出了一种优化的基于ARM的视觉导航AGV标识线图像处理方法。首先对采集到的图像进行灰度化处理并使用Otsu算法对图像进行阈值分割;然后采用优化的中值滤波算法进行图像滤波并使用高效的边缘提取策略获取路径边缘特征;最后采用角度判断的方法剔除错误点并使用最小二乘法拟合成路径的中心线。实验结果表明,该方法有较高的准确率和较好的实时性,可以满足工业生产中的实际需求,适用于基于嵌入式系统开发的视觉导航AGV。     

Human detection based on motion object extraction and head–shoulder feature

Aimed at the shortcomings of the traditional video monitoring system, human detection method in intelligent video monitoring system was researched. This paper proposed a human detection method based on motion object extraction and head–shoulder feature to complete human detection and statistics in video image sequences. Firstly, background subtraction based on adaptive threshold was used to extract foreground moving object information, then image erosion and image dilation were used to bypass the object shade and remove false object in order to optimize the results of motion object extraction. And finally, for realizing human moving object detection, we proposed the object discrimination algorithm based on human head–shoulder feature to complete human detection and statistics. Experimental results show that the method can successfully realize human detection and statistics. The method is highly accurate and has good real-time and extensive applications. The identification rate is 86% through human video sequences to test. This method can detect human automatically and provide the theoretical and technological base for object detection in the intelligent surveillance system.     

A non-rigid cardiac image registration method based on an optical flow model

According to non-rigid medical image registration, new method of classification registration is proposed. First, Feature points are extracted based on SIFT (Scale Invariant Feature Transform) from reference images and floating images to match feature points. And the coarse registration is performed using the least square method. Then the precise registration is achieved using the optical flow model algorithm. SIFT algorithm is based on local image features that are with good scale, rotation and illumination invariance. Optical flow algorithm does not extract features and use the image gray information directly, and its registration speed is faster. The both algorithms are complementary. SIFT algorithm is used for improving the convergence speed of optical flow algorithm, and optical flow algorithm makes the registration result more accurate. The experimental results prove that the algorithm can improve the accuracy of the non-rigid medical image registration and enhance the convergence speed. Therefore, the algorithm has some advantages in the image registration.     

Adaptive bistable stochastic resonance and its application in mechanical fault feature extraction

Stochastic resonance (SR) is an important approach to detect weak vibration signals from heavy background noise. In order to increase the calculation speed and improve the weak feature detection performance, a new bistable model has been built. With this model, an adaptive and fast SR method based on dyadic wavelet transform and least square system parameters solving is proposed in this paper. By adding the second-order differential item into the traditional bistable model, noise utilization can be increased and the quality of SR output signal can be improved. The iteration algorithm for implementing the adaptive SR is given. Compared with the traditional adaptive SR method, this algorithm does not need to set up the searching range and searching step size of the system parameters, but only requires a few iterations. The proposed method, discrete wavelet transform and the traditional adaptive SR method are applied to analyzing simulated vibration signals and extracting the fault feature of a rotor system. The contrastive results verify the superiority of the proposed method, and it can be effectively applied to weak mechanical fault feature extraction.     

基于相机阵列的光学组件缺陷在线检测方法

利用调焦方式可以实现焦距的连续变化从而对不同物距下的光学组件进行在线检测,但是调焦过程操作复杂且对调焦位移精度要求较高,景深内光学元件缺陷无法区分,难以实现真正意义上的在线检测。因此,本文提出了基于相机阵列的光学组件缺陷在线检测方法。首先建立了相机阵列的成像模型并给出了数字重聚焦表达式以及空间分辨率的表达式。接着利用MATLAB模拟相机阵列成像过程和数字重聚焦过程。最后进行实验验证,通过二维位移台带动相机对不同物距下的多个光学元件表面缺陷进行成像获得阵列相机图像,通过数字重聚焦算法得到不同物距下的光学元件表面缺陷分布信息。实验结果表明,基于相机阵列的光学组件缺陷在线检测技术能够同时对位于景深范围内的光学组件进行在线检测。该方法在光学元件缺陷在线检测方面有着一定的应用价值。     

Particle filter-based relative rolling estimation algorithm for non-cooperative infrared spacecraft

The issue of feature point mismatching among infrared image sequence would bring big challenge to estimating the relative motion of non-cooperative spacecraft for it couldn’t provide the prior knowledge about its geometric structure and motion pattern. The paper introduces particle filter to precisely match the feature points within a desired region predicted by a kinetic equation, and presents a least square estimation-based algorithm to measure the relative rolling motion of non-cooperative spacecraft. The state transition equation and the measurement update equation of non-cooperative spacecraft are represented by establishing its kinetic equations, and then the relative pose measurement is converted to the maximum posteriori probability estimation via assuming the uncertainties about geometric structure and motion pattern as random and time-varying variables. These uncertainties would be interpreted and even solved through continuously measuring the image feature points of the rotating non-cooperative infrared spacecraft. Subsequently, the feature point is matched within a predicted region among sequence infrared image using particle filter algorithm to overcome the position estimation noise caused by the uncertainties of geometric structure and motion pattern. Finally, the position parameters including rotation motion are estimated by means of solving the minimum error of feature point mismatching using least square estimate theory. Both simulated and real infrared image sequences are induced in the experiment to evaluate the performance of the relative rolling estimation, and the experimental data show that the rolling motion estimated by the proposed algorithm is more robust to the feature extraction noise and various rotation speed. Meanwhile, the relative rolling estimation error would increase dramatically with distance and rotation speed increasing.     

地球同步轨道碎片观测图像分级检测方法

为了快速检测出地球同步轨道带光学观测图像中的空间碎片,提出了一种基于信噪比的分级检测方法.首先分析实测数据,得出观测图像中不同信噪比碎片运动相似的特性.对于高信噪比碎片,采取帧差法快速检测,并利用多帧图像计算速度信息;对于低信噪比碎片,根据高信噪比碎片检测结果得到的速度信息,采用改进的动态规划检测方法.实际图像检测结果表明:当递归方程的搜索窗口为5pixel×5pixel时,分级检测方法检测性能最好;当累积帧数大于5帧时,检测时间比传统动态规划方法减少了87%以上.分级检测方法具有检测率高、虚警率低、计算量小的优势,适合地球同步轨道带光学观测图像中空间碎片的检测.     

基于星图识别的空间目标检测算法研究

从恒星背景中检测出空间目标是空间目标天基光学测量相机所要解决的关键技术难题之一。提出了一种基于星图识别的空间目标检测算法,利用背景星图的平移、旋转、比例伸缩不变性,应用星图识别原理,根据识别恒星在不同帧图像中的坐标变化估计背景运动参数,进行探测图像序列的配准,然后在配准后的图像序列中检测出空间目标。通过仿真验证,该方法对于运动恒星背景中运动小目标的检测具有比较好的效果。     

车削零件表面粗糙度图像法检测优选方法

为实现对车削零件表面粗糙度检测,提出一种基于机器视觉表面粗糙度检测图像处理的新方法。该方法先按相应算法对所采集图像剔出受光衍射影响严重区域,然后再按其灰度分布情况进行区域优化,获得的图像灰度特征参数能反映表面粗糙度量值的有效特征区域。用该方法对表面粗糙度Ra标称值为0.8 μm~12.5 μm的五种车削样件测试,处理后图像灰度的均值、方差、能量和熵等特征参数与Ra标称值单调关系显著,各特征曲线的非线性误差均在1.5%以内。对比实验显示,这种特征提取和区域优化方法可应用于表面粗糙度的区分与检测。     

基于分块速度域改进迭代运动目标检测算法的红外弱小目标检测

提出一种基于分块速度域的迭代红外运动目标检测算法来解决传统算法计算量巨大这一难题.首先,采用二维最小均方差滤波器对红外序列图像进行滤波,获得包含弱小目标以及残差的红外序列图像.然后,通过在序列图像块的速度域上应用改进的迭代运动目标检测算法进行能量累积,从而将弱小目标的运动速度在速度域进行累积增强,达到检测弱小运动目标的目的.最后在解算出的速度值附近进行搜索,得到弱小目标运动的精确速度.利用此速度进行空域能量累积,得到叠加图像,在此图上进行目标检测.与传统方法相比较,几组实验结果显示,本文提出的方法大大缩短了检测的时间,而且本文方法的检测效果也较好.