Small target detection using bilateral filter and temporal cross product in infrared images

We introduce a spatial and temporal target detection method using spatial bilateral filter (BF) and temporal cross product (TCP) of temporal pixels in infrared (IR) image sequences. At first, the TCP is presented to extract the characteristics of temporal pixels by using temporal profile in respective spatial coordinates of pixels. The TCP represents the cross product values by the gray level distance vector of a current temporal pixel and the adjacent temporal pixel, as well as the horizontal distance vector of the current temporal pixel and a temporal pixel corresponding to potential target center. The summation of TCP values of temporal pixels in spatial coordinates makes the temporal target image (TTI), which represents the temporal target information of temporal pixels in spatial coordinates. And then the proposed BF filter is used to extract the spatial target information. In order to predict background without targets, the proposed BF filter uses standard deviations obtained by an exponential mapping of the TCP value corresponding to the coordinate of a pixel processed spatially. The spatial target image (STI) is made by subtracting the predicted image from the original image. Thus, the spatial and temporal target image (STTI) is achieved by multiplying the STI and the TTI, and then targets finally are detected in STTI. In experimental result, the receiver operating characteristics (ROC) curves were computed experimentally to compare the objective performance. From the results, the proposed algorithm shows better discrimination of target and clutters and lower false alarm rates than the existing target detection methods.     

基于双边滤波的弱小目标背景抑制

 为解决红外弱小目标检测技术中结构化背景抑制的难题,利用双边滤波集成了图像几何、光度和局部结构相似性等信息并以非迭代、局部操作的优点,提出了一种基于双边滤波的红外弱小目标背景抑制算法,并引入了局部梯度的统计特性来抑制背景细节、增强目标信息,从而达到更好抑制图像中的背景,突出目标图像,提高图像整体对比度、信噪比的目的。实验结果显示,与小波滤波算法比较,该算法对含有弱小目标的复杂背景从主观视觉和数值指标都具有良好抑制效果。     

Infrared small target detection based on directional zero-crossing measure

Infrared small target detection under complex background and low signal-to-clutter ratio (SCR) condition is of great significance to the development on precision guidance and infrared surveillance. In order to detect targets precisely and extract targets from intricate clutters effectively, a detection method based on zero-crossing saliency (ZCS) map is proposed. The original map is first decomposed into different first-order directional derivative (FODD) maps by using FODD filters. Then the ZCS map is obtained by fusing all directional zero-crossing points. At last, an adaptive threshold is adopted to segment targets from the ZCS map. Experimental results on a series of images show that our method is effective and robust for detection under complex backgrounds. Moreover, compared with other five state-of-the-art methods, our method achieves better performance in terms of detection rate, SCR gain and background suppression factor.     

基于双边滤波的弱小目标背景抑制

为解决红外弱小目标检测技术中结构化背景抑制的难题,利用双边滤波集成了图像几何、光度和局部结构相似性等信息并以非迭代、局部操作的优点,提出了一种基于双边滤波的红外弱小目标背景抑制算法,并引入了局部梯度的统计特性来抑制背景细节、增强目标信息,从而达到更好抑制图像中的背景,突出目标图像,提高图像整体对比度、信噪比的目的。实验结果显示,与小波滤波算法比较,该算法对含有弱小目标的复杂背景从主观视觉和数值指标都具有良好抑制效果。     

深空背景弱小运动目标检测算法研究

在深入分析监测设备CCD图像特点的基础上,提出了一种深空背景弱小运动目标检测新方法。该算法使用"列高通滤波器"进行背景抑制;采用序列图像多帧累加增强目标与恒星的对比度,用交叉投影法确定星点区域,提取局部星图,利用局部星图匹配剔除恒星干扰;结合候选目标的特征,采用基于逻辑的最近邻关联方法完成目标检测。结果表明,该算法可满足深空背景弱小运动目标实时检测的要求。     

红外背景抑制与弱小目标的检测算法

强噪声背景下红外图像中弱小目标的检测一直是研究的重点和难点。根据弱小目标、背景干扰和噪声在红外图像中的差异,研究了三种低信噪比条件下红外图像中弱小目标的检测算法:小波变换、数学形态学、Top—hat算子,分别给出了处理的图像和相应的数据。仿真实验表明:这三种检测算法能十分有效地提高信噪比、增强目标、抑制背景杂波和去除噪声干扰,对信噪比约为2的弱小目标检测能得到很好的结果。三种算法所得结果一致,而且处理速度快,适合于实时图像处理和目标探测。     

基于全变差的弱小目标背景抑制

结构化背景抑制是红外弱小目标检测技术的一个难题。根据红外图像中目标和背景信号的特性,在定义Gabor函数为点扩展函数的基础上,提出了一种基于全变差滤波的背景抑制算法,该算法将图像的背景抑制转化为求解全变差滤波模型最小化,以便最终实现对红外图像的背景抑制。对真实的红外图像序列进行实验,并与小波域滤波算法进行了比较。几组实验结果表明,对在结构化背景下的红外弱小目标背景来说,从主观视觉和数值指标来看,该算法具有良好的抑制效果,且运算量较小,便于实时实现。     

红外搜索系统中弱小目标检测算法研究

 复杂背景下低信噪比弱小目标的检测是红外搜索系统中的重点和难点,为解决红外搜索系统中杂波干扰多、目标信噪比低等问题,提出一种模板匹配滤波的目标检测方法。该算法在预测背景的同时,通过对图像背景灰度值进行动态的阈值处理,自适应地进行背景抑制。当背景包含较多复杂因素时,采用模板匹配滤波的目标检测方法,消除背景抑制后的残留杂波,实现弱小目标的提取。试验结果表明：当场景较复杂且图像信噪比较低时,使用该算法处理后可使图像信噪比达到4 dB以上,从而提高了弱小目标的检测概率。     

Infrared small target detection via line-based reconstruction and entropy-induced suppression

This paper proposes a novel infrared small target detection method which is composed of two stages. The first stage is implemented by line-based reconstruction for suppressing the background clutter, and the second stage is induced by information entropy for further standing out the targets. Compared with the state-of-the-art approaches, the proposed approach is able to achieve better performance in terms of efficiency and accuracy.     

Small target detection based on accumulated center-surround difference measure

Small target detection is a critical problem in the Infrared Search And Track (IRST) system. Although it has been studied for years, there are some difficulties remained due to the clutter environment such as the cloud edge and the horizontal line. In the homogeneous area such as sky, cloud-inner area and sea surface area, target can easily be detected, but in heterogeneous area which contains cloud edge, sky-sea line the target may be falsely detected. This paper proposes a novel method called accumulated center-surround difference measure to detect infrared small target in heavy clutter. Each pixel’s accumulated center-surround difference measure is computed by using sliding window manner. The measure can effectively distinguish target region and heterogeneous region. Experimental results show our method achieves better performance.     

Bilateral two-dimensional least mean square filter for infrared small target detection

The basic TDLMS (Two-Dimensional Least Mean Square) filter fails to detect infrared small targets consistently, especially under conditions of heavy noise and distinct cloud edges. This paper proposes a robust and efficient small-target detection method based on the basic TDLMS filter. The method first smooths the input image with a Gaussian filter of adaptive variance, and then employs TDLMS with a selected step size to filter the image with rightward and leftward iterations. Two prediction error images are obtained by subtracting the prediction images of the bilateral filtering from the original input image. Each prediction error image is separated into positive and negative prediction error images. That is, four images are generated in the bilateral filtering. The final image is obtained by fusing these four images. Experimental results show that the proposed method achieves significant improvement in background suppression and detection performance over the basic TDLMS filter and other improved TDLMS filters.     

基于复滤波器组的红外弱小目标检测算法

为了有效抑制复杂背景的干扰,降低复杂背景所带来的虚警,提高目标检测的信噪比,提出了一种基于复滤波器组的红外弱小目标检测算法。分析了复杂背景下带有弱小目标的红外图像中复杂背景和弱小目标图像各自的频谱特性,并引入了分频段处理的思想。比较了各种滤波器的性能,并选用了基于复小波的滤波器组,用该滤波器组将红外弱小目标图像分解到各个子频域;对分解后的各频段图像分别进行基于罗宾逊滤波的目标检测处理,提取各频段图像中的奇异点;根据目标图像和背景图像的频谱特性的定量分析结果,选取合适的权值,将各频段检测的结果进行加权融合,得到最终的处理效果。实验结果表明:弱小目标检测方法较之于传统的不分频段的高通滤波处理方式可以获得更高的信噪比,目标得到明显的增强,背景杂波得到更有效的抑制,各项探测指标均更优。     

红外装甲目标检测中背景抑制技术研究

鉴于红外装甲目标检测中红外图像对比度低、背景复杂,导致图像的信噪比低而难于进行目标的检测,提出一种基于小波和改进的分形理论相结合的背景抑制方法。针对红外图像呈现的相关性强的特点,利用小波分析将图像中的低频缓变背景滤除,得到包含目标和强边缘杂波的图像;又由于目标分形维数对尺度的敏感程度高于边缘杂波的分形维数,提出通过计算图像在不同尺度内不规则因子的变化率来进一步抑制背景中的边缘杂波。实验表明:该算法能显著提高图像的信噪比(信噪比增益在2左右),对背景边缘有很好的抑制效果。     

基于边缘特征的中值滤波的小目标检测法

提出了一种基于边缘特征调整滤波尺寸的中值滤波算法用于小目标的检测。以图像像素的四个方向的边缘分量确定阈值,以此阈值判断像素是否为小目标区域,在此基础上调整滤波尺寸,采用中值滤波方法检测出小目标。结果表明,该算法对于小目标检测有很好的效果。     

Small target detection based on difference accumulation and Gaussian curvature under complex conditions

Small target detection is a significant subject in infrared search and track and other photoelectric imaging systems. The small target is imaged under complex conditions, which contains clouds, horizon and bright part. In this paper, a novel small target detection method is proposed based on difference accumulation, clustering and Gaussian curvature. Difference accumulation varies from regions. Therefore, after obtaining difference accumulations, clustering is applied to determine whether the pixel belongs to the heterogeneous region, and eliminate heterogeneous region. Then Gaussian curvature is used to separate target from the homogeneous region. Experiments are conducted for verification, along with comparisons to several other methods. The experimental results demonstrate that our method has an advantage of 1–2 orders of magnitude on SCRG and BSF than others. Given that the false alarm rate is 1, the detection probability can be approximately 0.9 by using proposed method.     

各向异性SUSAN滤波红外弱小目标检测

为了解决SUSAN滤波算子不能自适应调整滤波系数的问题,采用Geusebroek提出的各向异性高斯滤波器替代SUSAN滤波算子中的高斯滤波部分。由局部图像的方差和像素的邻域平滑度决定长短轴的方差,由该点的梯度方向决定滤波器的长轴方向,由局部图像的灰度值与均值差的一阶范数确定SUSAN滤波器的阈值,从而构造出各向异性SUSAN滤波器。将其用于红外弱小目标检测中,实验结果表明：各向异性SUSAN滤波器能够很好地保留图像中的边缘信息,使残差图像中弱小目标的信噪比增益和信杂比增益极大地提高,目标大小得到较好的保留,虚警率下降。     

深空背景弱小目标实时检测算法研究

根据深空背景的特点,提出一种基于FPGA和DSP的检测算法,并设计相应的硬件系统,能对大小为512×512像素、帧频率为40帧/s的视频图像进行实时处理。先在FPGA中用最大中值滤波抑制空间成像噪声,解决了图像预处理运算量大对处理速度要求高的难题,再在DSP中用使用改进的双差分法进行运动目标分割,克服了单差分提取目标轮廓不准的缺点,最后结合候选目标的特征,采用基于逻辑的最近邻关联方法提取目标航迹。试验结果证明该算法满足了深空背景弱小目标实时检测的要求。     

Edge directional 2D LMS filter for infrared small target detection

In this paper, we introduce an edge directional 2D least mean squares (LMSs) filter for small target detection in infrared (IR) images. Generally, the 2D LMS filter functions as a background prediction to apply to IR small target detection field. In order to accurately predict background objects as well as regions covered by small targets, the proposed 2D LMS filter take full advantage of edge information of prediction pixels corresponding to surrounding blocks around current filter window. And, to adjust adaptively its step size in the background and small target region, the adaptive region-dependent nonlinear step size is calculated by using the variance of the prediction pixels of the surrounding blocks. This prediction structure and adaptive step size of the proposed 2D LMS filter is applied to the background region including objects such as cloud edge and small target region differently. Through this way, the proposed 2D LMS filter predicts the background excluding small targets. Then, by subtracting the predicted background from the original IR image, small targets can be extracted. Experimental results show that the proposed 2D LMS filter has stronger target extraction and better background suppression ability compared to the existing 2D LMS filters.     

基于序列图像的小目标检测

 针对序列图像中小目标的运动具有运动的连续性和轨迹的一致性等特征,提出了采用单帧检测和多帧检测相结合的处理方法。对单帧图像进行预处理和图像分割,得到去掉大部分背景和噪声点的图像,并在图像中选取候选目标点进行标记。然后对处理后的序列图像进行N帧叠加。对叠加图中的标定点利用8邻域搜索法进行目标筛选,进而检测出目标。实验结果表明,该方法实用性强,能简单、快速、有效地检测出序列图像中的运动小目标。