基于形态学梯度的红外图像分割算法

提出了一种新的红外图像分割方法。该方法利用形态学方法来处理红外图像。首先进行形态学滤波,对红外目标图像中的噪声和微小的干扰区域进行滤除,接着提出了一种计算红外图像梯度的多尺度算法提取图像形态学梯度,而后分析了图像分形特征估计方法与形态学梯度的关系,提出了一种新的红外图像分形特征估计算法,在此基础上对图像进行分割。实验结果表明,该算法能较好地解决红外图像的分割问题。     

结合背景抑制技术的ELIR图像分割

针对前视红外（ＦＬＩＲ）图像的分割,提出采用去中值波滤器进行预处理抑制背景、增强目标,而进利用基于模型的ＦＬＩＲ图像分割（ＭＢＳ）算法完成图像分割,从相容性向量及初始概率计算两方面对ＭＢＳ算法进行了改进,对红外目标图结果证实该方法与ＭＢＳ算法相比,在低对比度、高噪声情况下能得到更为精确的分割结果,同时能极大地降低了背景干扰。     

结合背景抑制技术的FLIR图像分割

针对前视红外（ＦＬＩＲ）图像的分割,提出采用去中值波滤器进行预处理抑制背景、增强目标,进而利用基于模型的ＦＬＩＲ图像分割（ＭＢＳ）算法完成图像分割,从相容性向量及初始概率计算两方面对ＭＢＳ算法进行了改进．对实际红外目标图像分割结果证实该方法与ＭＢＳ算法相比,在低对比度、高噪声情况下能得到更为精确的分割结果,同时能极大地降低了背景干扰     

复杂背景的红外图像过渡区提取与分割

摘要：在基于个数判断噪声的中值滤波器进行平滑处理的基础上,提出了一种改进的过渡区提取方法,并将该方法应用于对比度低、宽谷直方图的复杂背景红外图像的分割。Matlab仿真实验表明,该方法减少了图像分割的运算量,并能将具有复杂背景的红外图像目标分割得较好。     

自适应权重的偏微分分割模型在变电设备红外图像中的应用

针对变电设备运维中的红外检测图像无法被准确分割的问题,本文应用了一种改进的自适应权重偏微分图像分割方法.通过分析红外图像信噪比低、边缘模糊、对比度低以及最常出现的灰度不均匀问题,在明确传统分割方法不足的基础上,对基于偏微分方程的分割模型开展改进.本文所提出的自适应权重的LGIF分割模型利用目标设备和背景灰度不均匀程度不...     

基于改进高低帽变换的钢轨裂纹红外图像增强

针对钢轨裂纹红外图像对比度低、信噪比低、纹理细节模糊而难以增强目标区域的问题,借助形态学高帽变换和低帽变换,提出了多尺度高帽低帽变换的钢轨裂纹红外图像增强优化算法。首先,用改进高帽变换、低帽变换分别提取多尺度明亮、暗淡图像区域;其次对多尺度的明亮与暗淡图像区域实施最大值的提取;然后操作其最大值以构建明亮和暗淡的图像区域;最后通过加权处理,实现图像增强。实验结果表明:本文算法在抑制噪声和突出了目标图像的边缘的基础上,有效地提高图像对比度,可应用于红外图像增强的场合,为后续图像信息处理奠定了必要的基础。     

结合多尺度循环卷积和多聚类空间的红外图像增强

红外图像低对比度、低亮度及轮廓细节模糊等问题限制了红外成像技术的应用.为了提高红外图像的质量,本文提出了一种结合多尺度循环卷积和多聚类空间红外图像增强方法,该方法分为五个阶段:多尺度循环卷积、多聚类空间混合均衡化、多角度组合滤波器、线性融合和图像层次化处理.其中,通过多尺度循环卷积和多角度组合滤波器增强图像细节及轮廓信...     

红外图像阈值分割算法的研究

目标分割是图像处理、计算机视觉及人工智能领域中的一项基本而关键的技术,其目的是为后续的识别、分类、跟踪和检索提供依据.由于红外图像具有对比度低、噪音大,图像模糊等特点,故将遗传算法应用到红外图像阚值分割中.本文对红外图像阈值分割算法进行了综述,对遗传算法在红外图像分割中应用的研究方法,研究现状及其新技术进行了介绍.该算法被用在红外图像处理中,并取得较好效果.     

基于形态学与二维Otsu相结合的红外图像分割方法

Otsu法是常用的基于阈值的图像分割方法之一,二维Otsu法利用图像的像素灰度值分布及邻域像素的平均灰度分布构成的二维直方图对图像进行阈值分割。由于红外图像具有低对比度、低信噪比、边缘模糊的特点,仅采用二维Otsu对其进行分割,由于噪声的影响,分割后的图像会有边缘信息不清晰以及误分割的问题。针对这个问题,本文提出的形态学与二维Otsu相结合的红外图像分割方法。实验证明,利用形态学可以保留图像基本形状,弥补分割图像细节,并使图像的轮廓更光滑的特点,达到较好的红外图像分割效果。     

一种热红外人脸分割的新方法

针对热红外人脸的边缘和细节特征模糊、对比度低、人脸和背景温度分布不同等特点,提出一种新的图像分割方法.该方法使用灰度投影粗定位人脸,使用模糊连接度分割背景和确定人眼眉毛的位置,根据眉毛的中心精确定位和归一化人脸.实验结果表明,该方法消除了背景干扰,保留了更多的人脸信息,能够有效解决热红外人脸图像的定位和分割问题.     

Communication emitter individual identification via 3D‐Hilbert energy spectrum‐based multiscale segmentation features

Specific emitter identification can detect emitters automatically by extracting and analyzing features. A novel specific emitter identification method based on 3D‐Hilbert energy spectrum‐based multiscale segmentation (3D‐HESMS) is proposed. First, the time‐frequency energy spectrum is derived via the Hilbert‐Huang transform, that is, a complicated curved surface in a 3D space, namely, the 3D‐Hilbert energy spectrum. The differential box dimension, multifractal dimension, lacunarity change rate, and 3D‐Hilbert energy entropy are extracted to compose the feature vector under multiscale segmentation using fractal theory. Subsequently, communication emitter individual identification is obtained using the 4 features. Finally, the performance and complexity of the 3D‐HESMS method are compared with those of 2 existing methods. Experiments show that the performance of the 3D‐HESMS method is better than those of the 2 other methods. The extracted features with high stability, sufficiency, and identifiability can overcome the negative effects of the changes in signal‐to‐noise ratio and the number of training samples.     

分形维数的多尺度形态估计与目标提取

本文研究了Minkowrki－Bouligand维数估计的数学形态学方法实现，给出了一种基于多尺度灰度形态运算的图像维数估计算法。将分形特征用于目标的检测，提出了一种具有较好稳健性的基于多分辨率分形特征的目标提取方法。     

Wavelet-based rotational invariant roughness features for texture classification and segmentation

We introduce a rotational invariant feature set for texture segmentation and classification, based on an extension of fractal dimension (FD) features. The FD extracts roughness information from images considering all available scales at once. In this work, a single scale is considered at a time so that textures with scale-dependent properties are satisfactorily characterized. Single-scale features are combined with multiple-scale features for a more complete textural representation. Wavelets are employed for the computation of single- and multiple-scale roughness features because of their ability to extract information at different resolutions. Features are extracted in multiple directions using directional wavelets, and the feature vector is finally transformed to a rotational invariant feature vector that retains the texture directional information. An iterative K-means scheme is used for segmentation, and a simplified form of a Bayesian classifier is used for classification. The use of the roughness feature set results in high-quality segmentation performance. Furthermore, it is shown that the roughness feature set exhibits a higher classification rate than other feature vectors presented in this work. The feature set retains the important properties of FD-based features, namely insensitivity to absolute illumination and contrast.     

基于分形理论的改进型二维最大熵红外图像分割算法

提出了一种基于分形理论的改进型二维最大熵红外图像阈值分割算法。该算法利用图像分形维数挖掘像素的空间分布信息,然后将原图像灰度及其分形维数映射图像灰度相结合组成二维随机向量,并构造出联合离散概率分布。在此基础上,以二维最大熵原则来确定一个最佳二维分割阈值,进而取得分割结果。实验结果表明,该算法在分割效果上优于传统的二维最大熵分割算法。     

基于两种神经网络的SAR图像多尺度分割与分析

基于多尺度回归技术和神经网络提出两种合成孔径雷达(SAR)图像分割的新方法.首先利用多尺度自回归模型(MAR)来描述SAR图像不同尺度间的统计相依性,以此提取SAR图像的多尺度统计特征;然后分别构造自组织特征映射网络和概率神经网络,并利用统计特征作为输入训练两种网络,实现SAR图像的分割.最后通过实验对这两种方法以及其他方法之间进行比较、分析,结果表明本文提出的两种方法的实验结果比较理想.     

Hierarchical Multiple Markov Chain Model for Unsupervised Texture Segmentation

In this paper, we present a novel multiscale texture model and a related algorithm for the unsupervised segmentation of color images. Elementary textures are characterized by their spatial interactions with neighboring regions along selected directions. Such interactions are modeled, in turn, by means of a set of Markov chains, one for each direction, whose parameters are collected in a feature vector that synthetically describes the texture. Based on the feature vectors, the texture are then recursively merged, giving rise to larger and more complex textures, which appear at different scales of observation: accordingly, the model is named Hierarchical Multiple Markov Chain (H-MMC). The Texture Fragmentation and Reconstruction (TFR) algorithm, addresses the unsupervised segmentation problem based on the H-MMC model. The “fragmentation” step allows one to find the elementary textures of the model, while the “reconstruction” step defines the hierarchical image segmentation based on a probabilistic measure (texture score) which takes into account both region scale and inter-region interactions. The performance of the proposed method was assessed through the Prague segmentation benchmark, based on mosaics of real natural textures, and also tested on real-world natural and remote sensing images.      

Communication emitter individual identification based on 3D-Hibert energy spectrum and multi-scale fractal features

For communication emitter identification,a novel method based on Hilbert-Huang transform (HHT) and multi-scale fractal features was proposed.First,the time frequency energy spectrum was derived via HHT,which was called a complicated curved surface in the three-dimension space,namely,3D-Hilbert energy spectrum.Then,the differential box dimension and the multi-fractal dimension was extracted to compose the feature vector under multi-scale segmentation using fractal theory.Finally,communication emitter individual identification was obtained using the two dimensions of features above and the support vector machine (SVM).Moreover,the novel method was compared with two existing methods to identify simulated and actual signals with different and the same modulation modes,respectively.Results show that the identification rate of the novel method is higher than that of the two other methods.The features extracted by the novel method have high stability,sufficiency,and identifiability,also outweigh the negative effects of the change of signal-to-noise ratio and the number of training samples and emitters.     

基于分形曲面尺度斜率特征的弱小目标检测

鉴于弱小目标检测所固有的难点及常用的检测方法不能准确、稳定地检测出目标,提出了运用分形曲面尺度斜率特征检测弱小目标的方法。通过实际数据分析可以得出:相比常用的分形维数和分形拟合误差等检测特征,分形曲面尺度斜率特征在表征人造目标与自然背景的差异上更加明显,在抗图像噪声干扰上也更为优异,有着更强的鲁棒性。该方法普遍适用于检测自然环境中的弱小目标,尤其在对空弱小目标方面,检测概率更高。无论背景、飞行姿态、目标类型发生怎样的变化,经本文算法运算后只需一步简易的分割就可以检测出微弱暗小目标。     

基于改进差分盒维数的烟雾分割方法

针对森林背景与烟雾的分形纹理结构的不同特性,提出了改进差分盒维数的烟雾分割方法。首先,在已有的差分盒维数算法基础上,扩大子窗口的选择范围,计算每个像素的分形维数值并分析得到的其分形特征数据;然后,选择合适的阈值对像素进行分类,筛选出符合烟雾特征的像素,从而实现烟与森林背景的分割;最后,应用形态学中膨胀算法进行连通处理。实验结果证明,基于改进的差分盒维数方法对以树木为背景的烟雾图像具有较好的分割效果。