Road Traffic Monitoring from Aerial Images Using Template Matching and Invariant Features

Road traffic monitoring is an imperative topic widely discussed among researchers.Systems used to monitor traffic frequently rely on cameras mounted on bridges or roadsides.However,aerial images provide the flexibility to use mobile platforms to detect the location and motion of the vehicle over a larger area.To this end,different models have shown the ability to recognize and track vehicles.However,these methods are not mature enough to produce accurate results in complex road scenes.Therefore,this paper presents an algorithm that combines state-of-the-art techniques for identifying and tracking vehicles in conjunction with image bursts.The extracted frames were converted to grayscale,followed by the application of a georeferencing algorithm to embed coordinate information into the images.The masking technique eliminated irrelevant data and reduced the computational cost of the overall monitoring system.Next,Sobel edge detection combined with Canny edge detection and Hough line transform has been applied for noise reduction.After preprocessing,the blob detection algorithm helped detect the vehicles.Vehicles of varying sizes have been detected by implementing a dynamic thresholding scheme.Detection was done on the first image of every burst.Then,to track vehicles,the model of each vehicle was made to find its matches in the succeeding images using the template matching algorithm.To further improve the tracking accuracy by incorporating motion information,Scale Invariant Feature Transform(SIFT)features have been used to find the best possible match among multiple matches.An accuracy rate of 87%for detection and 80%accuracy for tracking in the A1 Motorway Netherland dataset has been achieved.For the Vehicle Aerial Imaging from Drone(VAID)dataset,an accuracy rate of 86%for detection and 78%accuracy for tracking has been achieved.

A Robust Template Matching Algorithm Based on Reducing Dimensions

Template matching is a fundamental problem in pattern recognition, which has wide applications, especially in industrial inspection. In this paper, we propose a 1-D template matching algorithm which is an alternative for 2-D full search block matching algorithms. Our approach consists of three steps. In the first step the images are converted from 2-D into 1-D by summing up the intensity values of the image in two directions horizontal and vertical. In the second step, the template matching is performed among 1-D vectors using the similarity function sum of square difference. Finally, the decision will be taken based on the value of similarity function. Transformation template image and sub-images in the source image from 2-D grey level information into 1-D information vector reduce the dimensionality of the data and accelerate the computations. Experimental results show that the computational time of the proposed approach is faster and performance is better than three basic template matching methods. Moreover, our approach is robust to detect the target object with changes of illumination in the template also when the Gaussian noise added to the source image.

Robust multi-sensor image matching based on normalized self-similarity region descriptor

Multi-modal image matching is crucial in aerospace applications because it can fully exploit the complementary and valuable information contained in the amount and diversity of remote sensing images.However,it remains a challenging task due to significant non-linear radiometric,geometric differences,and noise across different sensors.To improve the performance of heterologous image matching,this paper proposes a normalized self-similarity region descriptor to extract consistent structural information.We first construct the pointwise self-similarity region descriptor based on the Euclidean distance between adjacent image blocks to reflect the structural properties of multi-modal images.Then,a linear normalization approach is used to form Modality Independent Region Descriptor(MIRD),which can effectively distinguish structural features such as points,lines,corners,and flat between multi-modal images.To further improve the matching accuracy,the included angle cosine similarity metric is adopted to exploit the directional vector information of multi-dimensional feature descriptors.The experimental results show that the proposed MIRD has better matching accuracy and robustness for various multi-modal image matching than the state-of-the-art methods.MIRD can effectively extract consistent geometric structure features and suppress the influence of SAR speckle noise using non-local neighboring image blocks operation,effectively applied to various multi-modal image matching.

基于相似度加权的自适应HD算法

Hausdorff距离(Hausdorff distance,HD)是一种点集与点集之间的距离测度,常用于目标物体的匹配、跟踪和识别等.本文在分析经典HD及改进算法的基础上,提出了一种基于相似度加权的自适应HD(Adaptive Hausdarff distance,AHD)算法.AHD算法利用不同点到点集的最小距离的个数作为匹配相似度的测量,并舍弃对判断匹配几乎没有作用的较大的点到点集的最小距离值;同时根据点到点集的最小距离自适应选择权值,从而得到一种基于相似度测量加权系数;通过利用部分点到点集的最小距离和基于相似度的加权平均,既增强了算法的鲁棒性,又尽可能地保证了算法的精度.实验结果显示,AHD算法在匹配准确性、抵抗噪声和遮挡干扰等方面性能良好.

A novel dense descriptor based on structure tensor voting for multi-modal image matching

Automatic and robust matching of multi-modal images can be challenging owing to the nonlinear intensity differences caused by radiometric variations among these images.To address this problem,a novel dense feature descriptor and improved similarity measure are proposed for enhancing the matching performance.The proposed descriptor is built on a voting scheme of structure tensor that can effectively capture the geometric structural properties of images.It is not only illumination and contrast invariant but also robust against the degradation caused by significant noise.Further,the similarity measure is improved to adapt to the reversal of orientation caused by the intensity inversion between multi-modal images.The proposed dense feature descriptor and improved similarity measure enable the development of a robust and practical templatematching algorithm for multi-modal images.We verify the proposed algorithm with a broad range of multi-modal images including optical,infrared,Synthetic Aperture Radar(SAR),digital surface model,and map data.The experimental results confirm its superiority to the state-of-the-art methods.

A Configurable Circuit for Cross-Correlation in Real-Time Image Matching

Cross-correlation （CC） is the most time-consuming in the implementation of image matching algorithms based on the correlation method. Therefore, how to calculate CC fast is crucial to real-time image matching. This work reveals that the single cascading multiply-accumulate （CAMAC） and concurrent multiply-accumulate （COMAC） architectures which have been widely used in the past, actually, do not necessarily bring about a satisfactory time performance for CC. To obtain better time performance and higher resource efficiency, this paper proposes a configurable circuit involving the advantages of CAMAC and COMAC for a large amount of multiply-accumulate （MAC） operations of CC in exhaustive search. The proposed circuit works in an array manner and can better adapt to changing size image matching in real-time processing. Experimental results demonstrate that this novel circuit which involves the two structures can complete vast MAC calculations at a very high speed. Compared with existing related work, it improves the computation density further and is more flexible to use.

Template Matching from 2-D into 1-D

The following article has been retracted due to special reason of the authors. This paper published in Vol.5 No. 2, 2014, has been removed from this site. Title: Template Matching from 2-D into 1-D Author: Yasser

Adaptive template filter method for image processing based on immune genetic algorithm

To preserve the original signal as much as possible and filter random noises as many as possible in image processing,a threshold optimization-based adaptive template filtering algorithm was proposed.Unlike conventional filters whose template shapes and coefficients were fixed,multi-templates were defined and the right template for each pixel could be matched adaptively based on local image characteristics in the proposed method.The superiority of this method was verified by former results concerning the matching experiment of actual image with the comparison of conventional filtering methods.The adaptive search ability of immune genetic algorithm with the elitist selection and elitist crossover(IGAE) was used to optimize threshold t of the transformation function,and then combined with wavelet transformation to estimate noise variance.Multi-experiments were performed to test the validity of IGAE.The results show that the filtered result of t obtained by IGAE is superior to that of t obtained by other methods,IGAE has a faster convergence speed and a higher computational efficiency compared with the canonical genetic algorithm with the elitism and the immune algorithm with the information entropy and elitism by multi-experiments.

工业机械臂辅助机器视觉的工件识别与定位

传统的工业机器人系统由于缺乏感知判断能力,很难适应目标种类和位置的变化,而现代工业需要生产不同种类和规格的产品。为此,以多品种小批量生产的零件为背景提出了一种机械臂辅助机器视觉系统,该系统能够实现对工件的图像采集并进行预处理,通过改进的Canny轮廓的提取算法与基于Hausdorff距离的模版匹配完成对不同种类工件进行识别和定位。最后经实验的验证该系统的定位精度在0.35 mm内,基本满足了在实际生产中对小批量生产、不同形状且没有固定夹具的目标工件的抓取、搬运与检测等要求。

基于图像处理的电路板缺陷检测系统设计

针对传统电路板缺陷检测多为人工检测、速度较慢且成本较高的问题,本文研究设计了一款以图像处理为基础、利用现场可编程门阵列(FPGA)实现对电路板缺陷准确、高速的检测系统。在传统图像增强算法的基础上提出一种针对不同图像信息采用不同感兴趣区间的方法,增强效果显著;为减少电路板上标识字样对匹配算法计算速度的影响,提出一种去除丝印算法,将电路板上多余的标识字样取消,减少图像匹配的计算量,加快检测的速度;在传统绝对误差和算法(SAD)模板匹配算法的基础上采用去平均值法计算图像信息,减小光照变化带来的影响;将传统2算子Sobel边缘检测扩展到8算子边缘检测,边缘信息更加明显清晰。采用FPGA作为硬件平台,在Vivado开发环境下实现Verilog HDL硬件逻辑语言,下载到FPGA中实现。实验结果表明,系统的平均检测精度为98.53%,检测单张电路板的时间为8.204 s。本系统设计在检测精度和速度上都有明显提升,且造价成本低。

基于FPGA的小目标识别分类系统的设计与实现

为了提高小目标识别和分类的实时性,同时降低识别系统的资源消耗,本文提出了一种简易、高效的现场可编程门阵列(Field Programmable Gate Array,FPGA)小目标识别分类系统。该系统首先通过图像预处理消除图像噪点,并采用并行计算提升系统实时性。然后将处理后的图像与模板进行匹配计算得到识别结果,设计的模板匹配电路具有较小的硬件复杂度和较快的处理速度。实验结果表明,本文所提出的识别系统在680×480图像分辨下,可达137.5帧/s的处理速度,实时性强,同时仅消耗了9个块随机存储器(Block Random Access Memory,BRAM)和2个数字信号处理器(Digital Signal Processor,DSP),硬件资源消耗较少,在处理小目标识别和分类问题上有较好的实用价值。

基于模板匹配的接缝宽度变化测量方法

为了解决传统接缝宽度变化测量算法在倾斜条件下拍摄出来的图像检测结果精度不足的问题,提出一种适应倾斜拍摄条件下的接缝宽度变化测量算法。分别采用三角形以及正方形靶标作为左右模板。首先选取合适大小的模板并通过模板匹配对拍摄图像进行匹配,然后获取左右模板并对其进行二值化处理,以获得左右两个模板的角点,利用交比不变性获得靶标的距离,以得到接缝的宽度变化。经过试验证明,该方法适用于远距离、倾斜拍摄下的接缝宽度变化测量,测量精度在0.2 mm以内,检测时间小于10 s,且减少了接缝宽度变化监测的成本,为工程应用提供了一定的价值。

曲率定位的图像区域化SSDA模板匹配方法

提出了一种结合区域化SSDA方法进行模板匹配的方法,基于灰度曲线的微血管定位和检测,并利用定位的特征区域的伪色差灰度图像形成图像模板,用于微血管显微图像的匹配和纠偏。在微血管的显微图像序列配准中进行了应用和测试,实验结果表明该方法可以有效应用于显微血管图像的快速匹配分析。

SSDA图像匹配算法的研究及实现

传统的模板图像匹配算法,匹配速度较慢,在实际应用中受到限制.为了加快图像匹配速度,因此人们提出了一种快速高效的图像匹配算法———序贯相似性检测算法(SSDA).对该算法进行了深入地分析,并在此基础上,采用VisualC++6·0对该算法进行了编程实现.

基于改进SSDA的彩色目标识别方法

提出了基于改进SSDA算法的机器人视觉彩色目标识别方法,利用颜色分量的权重系数对SSDA算法进行了改进,同时,在图像特征提取时引入目标的形状和大小信息。实验表明,这些措施有效地减少了运算量,提高了目标识别的准确性,具有较好的实时性和鲁棒性。

多模态遥感图像模板匹配Log-Gabor滤波方法

针对多模态遥感图像匹配难的问题,本文提出了一种基于Log-Gabor滤波的高精度匹配方法。该方法采用由粗到细的多层级密集匹配框架,无须进行特征点检测,避开了多模态图像特征点检测重复率低的问题,能够提取大量高精度匹配点对。本文方法主要分为两步:首先,利用多尺度多角度Log-Gabor滤波器构建对图像间非线性辐射差异稳健的特征金字塔;然后,利用粗尺度的底层特征图进行密集模板匹配,提取大量粗粒度的特征匹配点对,在此基础上再利用特征金字塔,实现粗匹配点自下而上的逐层优化,完成高精度特征匹配点对的提取。同时,针对模板匹配滑窗运算效率不高的问题,提出了一种密集模板匹配的快速实现方式,有效减少了密集模板匹配的运算时间。本文使用多组不同模态的遥感图像进行试验,结果表明,本文方法能够克服图像间非线性辐射差异的影响,在正确匹配数目、匹配准确率与匹配精度上均优于现有多模态图像特征匹配方法。

基于NCC与SSDA的快速特征点匹配融合算法

快速、正确的匹配从两幅或多幅图像中提取出来的特征点是基于特征点图像配准问题的关键。传统的只使用归一化互相关匹配(NCC)算法进行的特征点粗匹配,虽然具有较强的抗噪声能力,但是匹配的速度很慢,而且错误率也比较高。因此,在研究了NCC算法与序贯相似度检测(SSDA)算法基础上,并改进了SSDA算法的非相似度计算方法,提出将两种算法融合在一起形成一种快速的特征点匹配算法,改进算法充分利用了两种算法的优点,大大提高了特征点的匹配速度,而且减少了错误匹配的个数。通过实验证明,该算法是一种有效的特征点匹配算法,比只使用NCC算法进行特征点匹配所需的时间降低了70%以上,正确匹配率也有所提高。

钢拱架铰接孔视觉定位方法研究

针对钢拱架自动化拼接过程中需要对钢拱架铰接孔进行定位的实际问题,设计了一种基于双目视觉的钢拱架铰接孔定位方法。首先通过模板匹配对钢拱架铰接孔进行初定位,找到感兴趣区域;采用一种新的自适应阈值算法计算阈值,对感兴趣区域进行二值化处理;利用边缘检测和最小二乘法提取得到铰接孔椭圆轮廓特征;然后找到椭圆轮廓上特殊的两条弦,根据空间几何关系,计算空间圆平面的法向量;利用消影线和椭圆参数矩阵,计算铰接孔圆心在图像上的实际投影点;最后基于双目视觉成像模型计算铰接孔圆心的精确空间坐标。实验证明,本方法可实现钢拱架铰接孔高精度定位,满足实际工业需求,为钢拱架自动化拼接提供技术支撑。

基于改进多目标模板匹配在铝材计数的应用

针对铝材入库计数过程存在效率低和耗时长的问题,提出了一种基于机器视觉和改进的图像金字塔的铝材计数方法。该方法应用Sobel边缘检测提取模板轮廓,采用加权最小二乘法进行直线拟合实现模板轮廓的优化处理,进而得到最优模板,改进后的图像金字塔算法实现了对图像模板匹配的快速计数。实验结果表明,所提出的方法应用于铝材入库计数,具有较高的召回率和较快的匹配速度,测试数据集的平均检测准确率达到98.4%,匹配效率相较于传统图像金字塔算法提升了37.0%。

基于模板匹配的车牌字符识别算法研究

基于模板匹配的字符识别是车牌识别中重要的研究方法,该方法下组成的车牌识别系统涵盖了图像预处理、车牌定位、字符分割以及模板匹配等主要内容。该算法首先对拍照获取的车牌图像进行灰度化、边缘检测、阈值分割等预处理,其目的是为得到只含有车牌信息的图像;然后将车牌图像进行腐蚀处理,确定车牌区域,并且通过二值化得到突出车牌字符信息的二值图像,以便于进行字符分割;最后再将分割出的字符与字符模板进行匹配得到准确的识别结果。实验结果表明:基于该方法所实现的车牌识别系统,其识别过程简单且便捷,对于车牌的定位较为精准,能够较好地满足车牌识别的基本要求。