Local invariant algorithm applied in downward-looking image registration,usually computes the camera's pose relative to visual landmarks.Generally,there are three requirements in the process of image registration whe...Local invariant algorithm applied in downward-looking image registration,usually computes the camera's pose relative to visual landmarks.Generally,there are three requirements in the process of image registration when using these approaches.First,the algorithm is apt to be influenced by illumination.Second,algorithm should have less computational complexity.Third,the depth information of images needs to be estimated without other sensors.This paper investigates a famous local invariant feature named speeded up robust feature(SURF),and proposes a highspeed and robust image registration and localization algorithm based on it.With supports from feature tracking and pose estimation methods,the proposed algorithm can compute camera poses under different conditions of scale,viewpoint and rotation so as to precisely localize object's position.At last,the study makes registration experiment by scale invariant feature transform(SIFT),SURF and the proposed algorithm,and designs a method to evaluate their performances.Furthermore,this study makes object retrieval test on remote sensing video.For there is big deformation on remote sensing frames,the registration algorithm absorbs the Kanade-Lucas-Tomasi(KLT) 3-D coplanar calibration feature tracker methods,which can localize interesting targets precisely and efficiently.The experimental results prove that the proposed method has a higher localization speed and lower localization error rate than traditional visual simultaneous localization and mapping(vSLAM) in a period of time.展开更多
针对传统SURF算法(speeded up robust features)在拼接高分辨率无人机航拍图像时运行速度慢、特征匹配率低的特点,提出了一种基于IB-SURF(image block-SURF)技术的无人机图像拼接算法。结合无人机定位定姿系统(position and orientation...针对传统SURF算法(speeded up robust features)在拼接高分辨率无人机航拍图像时运行速度慢、特征匹配率低的特点,提出了一种基于IB-SURF(image block-SURF)技术的无人机图像拼接算法。结合无人机定位定姿系统(position and orientation system,POS)求取图像重叠区域;构造掩模在无人机图像重叠区域检测特征点,减少特征提取时间;借助图像分块(image block,IB)的思想对图像划分网格,精简筛选特征点;引入Neighborhood-KNN(neighborhood-K nearest neighbors)进行特征点匹配,提高图像匹配效率。实验结果表明,IB-SURF算法有较快的运行速度和较高的特征匹配率,平均特征匹配率达到84.3%,特征匹配正确率超过95.1%,为图像高质量拼接提供了技术基础。展开更多
类脑导航是模拟鼠类感知环境机制提出的一种同步定位与构图(Simultaneous localization and mapping,SLAM)的导航算法。针对复杂环境如室内光线变化导致类脑SLAM导航产生误差的问题,本文提出了基于特征匹配(Speeded up robust features,...类脑导航是模拟鼠类感知环境机制提出的一种同步定位与构图(Simultaneous localization and mapping,SLAM)的导航算法。针对复杂环境如室内光线变化导致类脑SLAM导航产生误差的问题,本文提出了基于特征匹配(Speeded up robust features,SURF)算法的优化类脑SLAM导航模型。该模型通过一套移动视觉系统采集环境信息,构建的局部场景细胞通过SURF特征匹配算法获取到载体在环境中的方向与位置信息;头朝向细胞与位置细胞通过连续吸引子神经网络共同表示载体当前的位姿。利用所获取的位姿与时间信息,通过路径积分计算当前载体在坐标系中所处的位置;最后,构建基于认知点的拓扑经验地图。此外,在局部场景细胞获取环境信息的同时,通过SURF特征匹配算法来进行闭环检测,判断是否需要对当前位置进行修正。本文提出的优化类脑SLAM模型很大程度改进了原有模型在有光线变化的室内情况下易产生场景误匹配的问题,并通过实验验证了本文提出方法的有效性。展开更多
针对传统图像匹配算法计算量大、耗时长等缺陷,提出一种基于SURF(speeded up robust features)的图像特征点快速匹配算法.首先对图像采用SURF算法提取特征点;然后通过Haar小波变换确定特征点的主方向和特征点描述子,使用优化的最近邻搜...针对传统图像匹配算法计算量大、耗时长等缺陷,提出一种基于SURF(speeded up robust features)的图像特征点快速匹配算法.首先对图像采用SURF算法提取特征点;然后通过Haar小波变换确定特征点的主方向和特征点描述子,使用优化的最近邻搜索算法(best bin first,BBF)进行特征点匹配;最后根据实际需要选取相似度最高的前n对匹配点进行对比实验.实验结果表明:该算法鲁棒性强,速度快,匹配准确性高,具有较大的应用价值.展开更多
针对传统车辆检索方法中存在准确性和区分度较低的问题,提出了一个基于改进SURF(speeded up robust features)算法的视频车辆检索方法。在车辆视频关键帧提取的基础上,根据改进SURF算法完成车辆图像的特征提取及匹配,其中包含改进FAST(f...针对传统车辆检索方法中存在准确性和区分度较低的问题,提出了一个基于改进SURF(speeded up robust features)算法的视频车辆检索方法。在车辆视频关键帧提取的基础上,根据改进SURF算法完成车辆图像的特征提取及匹配,其中包含改进FAST(features from accelerated segment test)特征点检测、SURF特征向量提取以及最近邻查询方法来进行特征点的匹配;通过计算比较待检索车辆图像与数据库车辆图像的相似度,算法完成图像筛选并反馈检索结果。实验结果表明:针对交通监控视频中待检索车辆,该方法能够较为准确地进行检索并反馈结果。展开更多
为从不同角度识别目标物体以及解决左右两幅图像中目标轮廓中心不匹配的问题,将SURF(Speeded Up Robust Features)算法与Grab Cut算法相结合,离线采集目标物体不同角度的图像,生成目标模板图片库。利用SURF算法完成目标物体的识别;利用S...为从不同角度识别目标物体以及解决左右两幅图像中目标轮廓中心不匹配的问题,将SURF(Speeded Up Robust Features)算法与Grab Cut算法相结合,离线采集目标物体不同角度的图像,生成目标模板图片库。利用SURF算法完成目标物体的识别;利用SURF算法自动初始化Grab Cut算法,实现目标轮廓的提取;利用基于灰度相关的区域匹配算法完成目标轮廓中心点的匹配,结合三维重建原理实现目标定位。实验结果表明,该方法可以成功识别目标物体并对目标物体进行准确定位。展开更多
基金supported by the National Natural Science Foundation of China (60802043)the National Basic Research Program of China(973 Program) (2010CB327900)
文摘Local invariant algorithm applied in downward-looking image registration,usually computes the camera's pose relative to visual landmarks.Generally,there are three requirements in the process of image registration when using these approaches.First,the algorithm is apt to be influenced by illumination.Second,algorithm should have less computational complexity.Third,the depth information of images needs to be estimated without other sensors.This paper investigates a famous local invariant feature named speeded up robust feature(SURF),and proposes a highspeed and robust image registration and localization algorithm based on it.With supports from feature tracking and pose estimation methods,the proposed algorithm can compute camera poses under different conditions of scale,viewpoint and rotation so as to precisely localize object's position.At last,the study makes registration experiment by scale invariant feature transform(SIFT),SURF and the proposed algorithm,and designs a method to evaluate their performances.Furthermore,this study makes object retrieval test on remote sensing video.For there is big deformation on remote sensing frames,the registration algorithm absorbs the Kanade-Lucas-Tomasi(KLT) 3-D coplanar calibration feature tracker methods,which can localize interesting targets precisely and efficiently.The experimental results prove that the proposed method has a higher localization speed and lower localization error rate than traditional visual simultaneous localization and mapping(vSLAM) in a period of time.
基金National Natural Sciences Foundation of China(Nos.61973281,51821003,51922009)Key Research and Development Project of Shanxi Province(No.202003D111003)+4 种基金Excellent Youngth Foundation of Shanxi Province(No.202103021222011)Foundation of Science and Technology on Electro-Optical Information Security control Laboratory(No.2021JCJQLB055010)Aviation Science Foundation(No.2018ZCU0002)Shanxi Province Key Laboratory of Quantum Sensing and Precision Measurement(No.201905D121001)1331 Project of Shanxi Province。
文摘类脑导航是模拟鼠类感知环境机制提出的一种同步定位与构图(Simultaneous localization and mapping,SLAM)的导航算法。针对复杂环境如室内光线变化导致类脑SLAM导航产生误差的问题,本文提出了基于特征匹配(Speeded up robust features,SURF)算法的优化类脑SLAM导航模型。该模型通过一套移动视觉系统采集环境信息,构建的局部场景细胞通过SURF特征匹配算法获取到载体在环境中的方向与位置信息;头朝向细胞与位置细胞通过连续吸引子神经网络共同表示载体当前的位姿。利用所获取的位姿与时间信息,通过路径积分计算当前载体在坐标系中所处的位置;最后,构建基于认知点的拓扑经验地图。此外,在局部场景细胞获取环境信息的同时,通过SURF特征匹配算法来进行闭环检测,判断是否需要对当前位置进行修正。本文提出的优化类脑SLAM模型很大程度改进了原有模型在有光线变化的室内情况下易产生场景误匹配的问题,并通过实验验证了本文提出方法的有效性。
文摘针对传统图像匹配算法计算量大、耗时长等缺陷,提出一种基于SURF(speeded up robust features)的图像特征点快速匹配算法.首先对图像采用SURF算法提取特征点;然后通过Haar小波变换确定特征点的主方向和特征点描述子,使用优化的最近邻搜索算法(best bin first,BBF)进行特征点匹配;最后根据实际需要选取相似度最高的前n对匹配点进行对比实验.实验结果表明:该算法鲁棒性强,速度快,匹配准确性高,具有较大的应用价值.
文摘针对传统车辆检索方法中存在准确性和区分度较低的问题,提出了一个基于改进SURF(speeded up robust features)算法的视频车辆检索方法。在车辆视频关键帧提取的基础上,根据改进SURF算法完成车辆图像的特征提取及匹配,其中包含改进FAST(features from accelerated segment test)特征点检测、SURF特征向量提取以及最近邻查询方法来进行特征点的匹配;通过计算比较待检索车辆图像与数据库车辆图像的相似度,算法完成图像筛选并反馈检索结果。实验结果表明:针对交通监控视频中待检索车辆,该方法能够较为准确地进行检索并反馈结果。