中低分辨率SAR图像桥梁识别方法研究

中低分辨率SAR图像中河流、桥梁目标常存在断裂现象,且桥梁边缘线性特征不明显,现有桥梁目标识别方法难以解决上述问题.为提高中低分辨率SAR图像桥梁识别稳健性,提出桥梁在图像中呈现断裂、河流灰度一致性不高情况下的桥梁目标识别算法.对水域分割结果进行河流粗提取,将河流轮廓外扩部分的相交区域作为桥梁兴趣区,在保证兴趣区完整性的同时减少无效兴趣区;采用目标分割与Radon变换结合的方法检测断裂的桥梁主体;对候选桥梁附近的河流轮廓进行形状验证,消除山体等阴影造成的桥梁虚警.实验结果表明,该方法能够在复杂场景中有效地识别出多个水上桥梁目标,桥梁虚警率大大降低.     

利用规则化热扩散方程的SAR图像桥梁检测

提出了一种利用规则化各向异性热扩散方程SAR图像分割的桥梁检测算法.该算法在Pemna和Malik提出的各向异性热扩散方程的基础上构造了一个新的扩散函数,利用数值逼近理论,得到一个新的规则化扩散模型,用此模型对图像初始分割的最大后验概率矩阵进行多尺度各向异性平滑,得到图像中河流的精确分割结果,最后在分割后的图像中按累加方向能量最小准则进行桥梁目标检测.真实数据实验结果表明,该方法能有效地抑制强斑点噪声,快速、精确地检测出SAR图像桥梁目标,同时保持桥梁的结构信息.     

基于水平集的SAR图像预处理方法

为了降低SAR图像中相干斑噪声和背景杂波的影响,针对SAR图像舰船目标切片,本文提出一种基于水平集的SAR图像预处理方法。首先对SAR图像切片进行水平集分割,将分割后的图像二值化处理,再将二值化后的图像采用区域消除的方法滤除杂波,从而获得预处理后的目标图像。为验证此方法的实验效果,将几种典型的SAR图像预处理方法与本文采用的方法进行了比较,实验结果表明,该方法能够较好地滤除杂波噪声并保留目标的边缘轮廓,降低了特征提取的难度,可以提高后续目标识别的精度,是一种有效的SAR图像预处理方法。     

基于SLIC超像素分割的SAR图像海陆分割算法

海陆分割在合成孔径雷达(SAR)图像的海面目标检测以及海岸线提取等海洋应用方面具有非常重要的意义。针对合成孔径雷达图像的特点,提出了基于SLIC超像素分割的SAR图像海陆分割算法。首先为抑制SAR图像固有相干斑噪声并较好地保留图像的边缘信息,采用精致Lee滤波对图像进行预处理。然后对图像进行SLIC超像素分割,再将分割后的图像进行FT区域显著性检测以及显著值相似度聚类。最后将处理后的图片二值化得到海陆分割结果。实验结果表明,本文所提海陆分割算法具有很高的处理精度以及较高的处理效率。     

一种基于极化圆周SAR图像的陆上桥梁提取方法

桥梁作为重要的人造目标,一直都是合成孔径雷达(SAR)图像解译的重要对象之一.目前针对桥梁检测问题已开展了较多研究,核心思想是:首先提取出河流水体,然后再根据河流与桥梁的位置关系检测桥梁.然而,已有的桥梁检测方法依赖于河流提取,很难实现陆上桥梁检测.因为陆上桥梁下方的背景不再是河流,而是陆地,其散射特性、形状分布与河流...     

基于全卷积网络与凹点搜索的重叠葡萄分割算法

为解决自然环境下重叠葡萄难以分离的问题,提出了一种基于改进的FCN全卷积网络和凹点搜索的重叠葡萄分割方法.主要步骤如下:第一,对重叠葡萄图像进行H分量提取,然后通过改进的FCN全卷积网络得到二值图,对二值图进行质心位置求解和轮廓凸包检测;第二,根据矢量夹角最大法进行单串葡萄轴线检测,分离前的轮廓凸包与二值图像相减得到凹...     

基于显著轮廓特征的SAR图像轮廓匹配新方法

在以星载SAR图像作为基准图、机载/弹载SAR图像作为实时图的匹配导航和精确制导研究中,传统基于点特征的匹配方法存在特征点数目过多, 误匹配率较高,容易受噪声及灰度变化影响等问题。该文提出一种基于显著轮廓特征的SAR图像“由粗到精”的匹配新方法。该方法在对SAR图像进行预处理的基础上,采用改进的模糊C均值聚类(FCM)的图像分割方法来提取闭合轮廓特征;采用归一化轮廓中心距离描述符进行双向匹配,获得强鲁棒性的粗匹配轮廓对;在粗匹配轮廓上采用改进的局部二值模式(LBP)算子得到精匹配结果。试验结果表明,该方法在图像旋转、空间变化以及噪声干扰较大的情况下,具有精确性高、鲁棒性强的优势,适宜遥感SAR图像匹配。     

基于显著轮廓特征的SAR图像轮廓匹配新方法

在以星载SAR图像作为基准图、机载/弹载SAR图像作为实时图的匹配导航和精确制导研究中,传统基于点特征的匹配方法存在特征点数目过多,误匹配率较高,容易受噪声及灰度变化影响等问题.该文提出一种基于显著轮廓特征的SAR图像"由粗到精"的匹配新方法.该方法在对SAR图像进行预处理的基础上,采用改进的模糊C均值聚类(FCM)的图像分割方法来提取闭合轮廓特征;采用归一化轮廓中心距离描述符进行双向匹配,获得强鲁棒性的粗匹配轮廓对;在粗匹配轮廓上采用改进的局部二值模式(LBP)算子得到精匹配结果.试验结果表明,该方法在图像旋转、空间变化以及噪声干扰较大的情况下,具有精确性高、鲁棒性强的优势,适宜遥感SAR图像匹配.     

典型Otsu算法阈值比较及其SAR图像水域分割性能分析

遥感图像中水域呈现低灰度、灰度起伏小特性。该文从实时性出发,采用1维Otsu算法分割SAR图像。证明了两种典型Otsu分割算法的阈值关系,提出了分割性能评估方法,并通过实测SAR图像对比分析了多阈值分割与单阈值递归分割的水域分割性能。结果表明,与单阈值递归分割相比,多阈值分割能够更好地提取水域轮廓,且漏警率低,可应用于桥梁自动目标识别和景象匹配等方面。     

基于深度学习的机载SAR典型目标识别算法

在多分辨率场景下基于合成孔径雷达(SAR)图像进行多类典型目标识别,是SAR图像信息解译的重要环节。基于YOLO-v4网络模型,针对目前机载SAR图像及目标信息的特点,提出一种应用于真实机载平台下多场景跨分辨率的实时检测处理架构。文中通过对多类目标进行双重检测,对样本数据量低的训练集进行数据增强,并将图像分割后的同类型目标信息进行合并,解决了多分辨率SAR场景下目标尺度跨度较大的问题。实验结果表明：该方法能够在相关机载SAR数据集上达到六类目标(机场、桥梁、立交桥、汽车、装甲车、飞机)82.8%的mAP值,对后续机载SAR复杂场景下更多类型目标的检测识别具有重要的借鉴意义。     

Knowledge-based segmentation of Landsat images

A knowledge-based approach for Landsat image segmentation is proposed. The image segmentation problem is solved by extracting kernel information from the input image to provide an initial interpretation of the image and by using a knowledge-based hierarchical classifier to discriminate between major land-cover types in the study area. The proposed method is designed in such a way that a Landsat image can be segmented and interpreted without any prior image-dependent information. The general spectral land-cover knowledge is constructed from the training land-cover data, and the road information of an image is obtained through a road-detection program     

红外图像目标快速分割方法

提出一种红外图像阈值分割方法。为了减少计算量,首先给出一种提取包含待分割目 标的感兴趣区域方法;然后利用Bezier 曲线法平滑感兴趣区域直方图的噪声,利用所提出的判据来确定最佳分割阈值并进行分割。最后结合上述分割结果和目标的边缘信息得到最终的分割效果。实验结果表明,提出的方法能快速有效的将红外目标从复杂的背景中分割出来。     

Minimal shape and intensity cost path segmentation

A new generic model-based segmentation algorithm is presented, which can be trained from examples akin to the active shape model (ASM) approach in order to acquire knowledge about the shape to be segmented and about the gray-level appearance of the object in the image. Whereas ASM alternates between shape and intensity information during search, the proposed approach optimizes for shape and intensity characteristics simultaneously. Local gray-level appearance information at the landmark points extracted from feature images is used to automatically detect a number of plausible candidate locations for each landmark. The shape information is described by multiple landmark-specific statistical models that capture local dependencies between adjacent landmarks on the shape. The shape and intensity models are combined in a single cost function that is optimized noniteratively using dynamic programming, without the need for initialization. The algorithm was validated for segmentation of anatomical structures in chest and hand radiographs. In each experiment, the presented method had a significant higher performance when compared to the ASM schemes. As the method is highly effective, optimally suited for pathological cases and easy to implement, it is highly useful for many medical image segmentation tasks.     

Irregular Segmented Region Compression Coding Based on Pulse Coupled Neural Network

An irregular segmented region coding algorithm based on pulse coupled neural network（PCNN） is presented. PCNN has the property of pulse-coupled and changeable threshold, through which these adjacent pixels with approximate gray values can be activated simultaneously. One can draw a conclusion that PCNN has the advantage of realizing the regional segmentation, and the details of original image can be achieved by the parameter adjustment of segmented images, and at the same time, the trivial segmented regions can be avoided. For the better approximation of irregular segmented regions, the Gram-Schmidt method, by which a group of orthonormal basis functions is constructed from a group of linear independent initial base functions, is adopted. Because of the orthonormal reconstructing method, the quality of reconstructed image can be greatly improved and the progressive image transmission will also be possible.     

基于混沌蚁群算法的快速红外图像分割

提出了一种基于混沌蚁群算法优化二维模糊划分最大熵的红外图像分割方法。二维模糊划分最大熵分割方法不仅利用了灰度信息以及空间邻域信息,并且兼顾图像自身的模糊性,能取得很好的分割效果,然而最大熵的最优参量组合却很难快速准确地获得。本文将混沌蚁群优化算法应用到二维模糊划分最大熵分割方法当中,充分利用混沌蚁群算法快速寻找最优解的特点,来搜索二维模糊划分最大熵的最优参量组合。实验仿真结果表明,该方法比传统的图像分割方法有更好地分割效果,有效抑制了图像噪声对目标区域分割的干扰。     

Compressing the background layer in compound images,using JPEG and data filling

To efficiently compress rasterized compound documents, an encoder must be content-adaptive. Content adaptivity may be achieved by employing a layered approach. In such an approach, a compound image is segmented into layers so that appropriate encoders can be used to compress these layers individually. A major factor in using standard encoders efficiently is to match the layers’ characteristics to those of the encoders by using data filling techniques to fill-in the initially sparse layers. In this work we present a review of methods dealing with data filling and propose also a sub-optimal non-linear projections scheme that efficiently matches the baseline JPEG coder in compressing background layers, leading to smaller files with better image quality.     

多分辨率下的彩色图像分割方法

通过对彩色图像分割方法的研究,提出一种在多分辨率条件下彩色图像的分割方法。通过离散小波变换获得原始图像的低分辨率近似,并利用分水岭算法对低分辨率近似图像进行初始分割。对初始分割区域根据颜色信息和空间关系按照一定的先后顺序进行合并,直到形成理想的分割结果。对合并后的图像应用小波逆变换得到原始分辨率下的分割结果。在低分辨率下进行的分割和合并大大减小了计算量,提高了分割速度。     

Controlling Familywise Error Rate for Matched Subspace Detection in Dynamic FDG PET

Detection of small lesions in fluorodeoxyglucose (FDG) positron emission tomography (PET) is limited by image resolution and low signal to noise ratio. We have previously described a matched subspace detection method that uses the time activity curve to distinguish tumors from background in dynamic FDG PET. Applying this algorithm on a voxel by voxel basis throughout the dynamic image produces a test statistic image or ldquomaprdquo which on thresholding indicates the potential locations of secondary or metastatic tumors. In this paper, we describe a thresholding method that controls familywise error rate (FWER) for the matched subspace detection statistical map. The method involves three steps. First, the PET image is segmented into several homogeneous regions. Then, the statistical map is normalized to a zero mean unit variance Gaussian random field. Finally, the images are thresholded at a fixed FWER by estimating their spatial smoothness and applying a random field theory maximum statistic approach. We evaluate this thresholding method using digital phantoms generated from clinical dynamic images. We also present an application of the proposed approach to clinical PET data from a breast cancer patient with metastatic disease.     

基于脉冲耦合神经网络和施密特正交基的一种新型图像压缩编码算法

自从脉冲耦合神经网络(PCNN)被提出以来,在图像处理、模式识别、人工智能等领域得到了广泛应用.由于其生物学背景的特性,使得其能够对灰度图像进行完美的分割:PCNN局部连接域的作用及阈值指数衰减特性,使得具有近似灰度特性的邻近像素能够同时处于激活状态,这就构成了PCNN分割特性的基础,使得图像分割结果既能较好地包含原始图像细节信息,又能避免一些无意义的小分割块的产生.借鉴施密特正交化思想,利用自然初始基对每一分割区域进行变换,得到一组正交基的变换系数,相对于分割前图像的数据量大为减少,存储空间需求小,从而实现了压缩.相对于JPEG算法,该方法使重建图像的质量得到显著提高,同时也使得逐步重建图像成为可能.     

Registration and real-time visualization of transcranial magnetic stimulation with 3-D MR images

This paper describes a method for registering and visualizing in real-time the results of transcranial magnetic stimulations (TMS) in physical space on the corresponding anatomical locations in MR images of the brain. The method proceeds in three main steps. First, the patient scalp is digitized in physical space with a magnetic-field digitizer, following a specific digitization pattern. Second, a registration process minimizes the mean square distance between those points and a segmented scalp surface extracted from the magnetic resonance image. Following this registration, the physician can follow the change in coil position in real-time through the visualization interface and adjust the coil position to the desired anatomical location. Third, amplitude of motor evoked potentials can be projected onto the segmented brain in order to create functional brain maps. The registration has subpixel accuracy in a study with simulated data, while we obtain a point to surface root-mean-square error of 1.17+/-0.38 mm in a 24 subject study.