首页 | 本学科首页   官方微博 | 高级检索  
相似文献
 共查询到18条相似文献,搜索用时 109 毫秒
1.
传统的基于形状信息目标定位的算法,对目标观测角度发生形变情况下的定位存在不少困难,针对该问题,提出了一种基于稀疏活动轮廓模型的感兴趣目标(OOI)检测算法.首先通过共同勾画算法学习到感兴趣目标的稀疏活动轮廓模型,它能够清晰地定义感兴趣目标模式;同时构成该模型的Gabor轮廓基元可以通过扰动进行局部的调整以适配图像,在一...  相似文献   

2.
真实场景下视频运动目标自动提取方法   总被引:17,自引:10,他引:7  
视频运动目标跟踪逐渐成为研究热点并应用到军事民用等领域,为了能够从真实场景中快速准确地提取视频跟踪单运动目标或多运动目标,提出了一种新的运动目标自动提取方法。首先通过自适应阈值获得滤波后的相邻帧差值图像。其次,为了消除差值图像中噪声的影响,标记此二值图像的连通像素来检测出运动目标所在的区域,并与边缘检测出的空间信息结果比较得到运动目标模型。最后,将图像分成若干区域,在每个分区域内依次连接每个运动目标模型的最外围轮廓点,由此构成目标闭合轮廓。利用得到的连续边界,对运动目标进行提取。实验结果表明,该算法能够有效地自动提取速度不同的单运动目标,同时能够提取多运动目标。  相似文献   

3.
Snake模型在医学图像分割中的应用已经越来越广泛,但在应用该模型时,如何选取合适的初始轮廓是一个难题;在对血管内超声医学图像的研究基础上,提出了一种基于灰度信息与ROI区域的初始轮廓获取方法,根据IVUS图像的灰度特征对其进行自适应阈值分割以及面积滤波,然后获得分割轮廓点集进而得到snake初始轮廓点集;在matlab7.0环境分别对不同类别的2种IVUS图像的中外膜边缘提取进行仿真实验,实验证明该方法获得的初始snake轮廓较为逼近目标真实轮廓且适合于snake模型进行迭代收敛,由于其初始轮廓已较为接近目标真实轮廓,因此节省了snake模型的迭代次数,算法运行效率也优于手工提取初始轮廓的snake方法,可以较为方便的应用于实际领域。  相似文献   

4.
基于线性不变矩的特征图象配准算法研究   总被引:7,自引:7,他引:0  
杨静  丘江  王岩飞  刘波 《光子学报》2003,32(9):1114-1117
针对图象差异较大的光学图象与合成孔径雷达图象(SAR)的特点,设计了一种基于特征的图象配准的算法,包括特征提取、匹配、控制点选取、变换系数计算和误差计算等步骤.在特征提取中,对于雷达图象需要进行预处理并采用LOG提取图象轮廓.在匹配算法中,以线性不变矩作为特征量,对现有基于图象轮廓特征的匹配算法进行了改进.采用上海浦东地区的Radarsat-1S2模式图象与LandsatTM-5波段图象作为待配准图象.实验结果表明,该方法可以较好地完成光学图象与SAR图象的配准,有着较高的应用价值.  相似文献   

5.
冯鑫  李川  胡开群 《物理学报》2014,63(18):184202-184202
为了克服红外与可见光图像融合时噪声干扰及易产生伪影导致目标轮廓不鲜明、对比度低的缺点,提出一种基于深度模型分割的图像融合方法.首先,采用深度玻尔兹曼机学习红外与可见光的目标和背景轮廓先验,构建轮廓的深度分割模型,通过Split Bregman迭代算法获取最优能量分割后的红外与可见光图像轮廓;然后再使用非下采样轮廓波变换对源图像进行分解,并针对所分割的背景轮廓采用结构相似度的规则进行系数组合;最后进行非下采样轮廓波反变换重构出融合图像.数值试验证明,该算法可以有效获取目标和背景轮廓均清晰的融合图像,融合结果不但具有较高的对比度,还能抑制噪声影响,具有有效性.  相似文献   

6.
基于红外图像的GVF Snake轮廓提取算法的研究   总被引:1,自引:0,他引:1  
针对红外图像目标具有边界模糊不清,区分效果较差的缺点,结合Ostu阈值法和梯度矢量流主动轮廓模型(GVF Snake),提出一种目标轮廓自动提取方法。采用Ostu法先对图像进行分割,然后将得到的边界作为Snake模型的初始边缘轮廓,利用 GVF Snake特性将初始轮廓准确地收敛到目标边界。由于Ostu算法具有将目标物体从复杂背景中分割开来的优点,使得在应用 GVF Snake模型对复杂图像进行分割时减少了人工的干预。实验证明:该方法运算速度快,能够快速地收敛到目标轮廓,并准确地跟踪目标,具有一定的抗噪能力。  相似文献   

7.
提出了一种基于分段轮廓平滑的目标识别算法.首先通过曲率将轮廓划分为特征区域和非特征区域;然后在不同区域内分别采用不同方差的高斯函数进行轮廓平滑;最后采用基于仿射不变矩的目标识别算法对平滑后的目标轮廓进行识别.结果表明,该算法不仅取得了更好的轮廓平滑效果,而且在强噪音条件下能够显著提高识别准确率.  相似文献   

8.
从复杂、动态多变的交通图像中准确提取障碍物的轮廓曲线是智能汽车的一个重要研究课题,它对行人保护起着十分重要的作用。Snake模型是用来自动提取物体边界的曲线模型。结合立体视觉技术和Snake模型以实现行人检测:运用基于稠密视差的立体分割方法查找并分割潜在行人目标区域,为了便于后期目标轮廓提取,基于边缘检索的立体匹配算法被进一步用于提取感趣区(ROIs)内的目标初始边界;在此基础上,用Snake模型提取目标的完整轮廓曲线;轮廓因子及目标高程被用于ROIs的验证,即行人识别。针对Snake模型易受噪声干扰及难以收敛到凹陷边界等缺陷,提出了改进的距离势能模型。以典型交通场景为分析对象,对所提出的方法进行了测试,得到了较为理想的结果。  相似文献   

9.
提出了一种基于分段轮廓平滑的目标识别算法.首先通过曲率将轮廓划分为特征区域和非特征区域|然后在不同区域内分别采用不同方差的高斯函数进行轮廓平滑;最后采用基于仿射不变矩的目标识别算法对平滑后的目标轮廓进行识别.结果表明,该算法不仅取得了更好的轮廓平滑效果,而且在强噪音条件下能够显著提高识别准确率.  相似文献   

10.
基于差分图像的运动目标跟踪与分割方法的应用研究   总被引:8,自引:2,他引:6  
提出了一种基于差分图像运动检测和轮廓提取的跟踪与分割图像的方法。该方法首先从初始图像中提取目标的轮廓,然后利用相邻帧之间的差分图像初步确定目标在每帧图像中的粗略位置,最后把从上一帧图像中得到的目标轮廓置于该位置,并作为轮廓提取的初始值,由此可得到对目标的准确分割。  相似文献   

11.
Active contour model also known as Snake, is a popular approach for contour extraction and image segmentation. However, some existing active contour models are sensitive to the initial curve, which is usually required to be placed near the true object contour. To address this problem, this paper proposes a novel active contour model with adaptive initial curve, namely Saliency Snake. Taking visual saliency into consideration, prior shape information of the interested object is incorporated explicitly into the energy functional of Saliency Snake. Such improvement facilitates active contour evolution and leads to fast and automatic segmentation. Experiment results demonstrate that the proposed Saliency Snake can achieve superior segmentation performance both in terms of accuracy and efficiency.  相似文献   

12.
A quick and efficient technique for a charge-coupled device (CCD)-based profilometric measurement system is proposed and verified by experiment. When a laser stripe is projected onto an object, the deformed laser line image is detected by a CCD camera with subpixel resolution and is automatically transferred to its space position using a least-squares mapping algorithm. Measurements are executed on each separate surface of a three-dimensional object and the multi-surface data are integrated through a process of coordinate transformation. Large arrays of data points are acquired on each surface by means of a scanning mechanism. Because the measuring data from each laser stripe position can provide two-dimensional information concerning the surface contour, the laser stripe and the CCD camera are integrated into a measurement system for obtaining the entire three-dimensional information. Theoretical analysis shows that this technique can determine the three-dimensional profile of an object with an accuracy below 0.2 mm.  相似文献   

13.
The model-to-image registration problem is a problem of determining the position and orientation (the pose) of a three-dimensional object with respect to a camera coordinate system. When there is no additional information available to constrain the pose of the object and to constrain the correspondence of object features to image features, the problem is also known as simultaneous pose and correspondence problem, or correspondenceless pose estimation problem. In this paper, we present a new algorithm, called extended gravitational pose estimation (EGPE), for determining the pose and correspondence simultaneously. The algorithm is based on gravitational pose estimation (GPE) algorithm. In our algorithm, the original GPE has been revised to deal with the problem with false image points. For problems with both occluded object points and false image points, we firstly applied single-link agglomerative clustering algorithm to pick out occluded object points when a local minimum has been found, then the revised GPE is applied again on the clustering result to update rotation and translation of the object model. EGPE has been verified on both synthetic images and real images. Empirical results show that EGPE is faster, more stable and reliable than most current algorithms, and can be used in real applications.  相似文献   

14.
Holographic contouring is a simple and flexible method of producing a contour map of the shape of a three-dimensional object with a contour interval from 1 to 300 μm. However, a limitation of the method is that precise information on the shape of the object is not available at points between the contours. This can sometimes result in ambiguities in the interpretation of the results. In addition, evaluation of the deviations of the shape of the object from a specified shape is not easy. This paper shows how these problems can be solved by the application of digital techniques of phase measurement. A simple technique by which the sensitivity of the method can be varied continuously over a wide range is also described.  相似文献   

15.
16.
2D MRSI suffers from the effect of the spatial response function due to the truncation of the sampling of k-space. Filtering of the k-space data-set is often used to suppress the side lobes caused by the effects of the SRF, where the sampled data-set is multiplied with a weighting function before inverse FT. Commonly used filters in MRSI are the cosine, Hanning and Hamming filters. The data-set is often interpolated into a larger image matrix size for analysis, where "Fourier interpolation" (FoI) and "cubic spline interpolation" (CSpI) are two common methods. In this work, the effects of k-space filtering in practical usage was examined, and the image representations of the object for the two interpolation methods were compared.This study showed that application of filtering improves the image representation of the structures in the object and should be used in MRSI. FoI correctly visualizes the information inherent in the data-set, while the features of the object were dependent on the position of the object relative the original matrix in the CSpI interpolated images. FoI should therefore be used for quantitative evaluation of MRSI images.  相似文献   

17.
18.
This paper proposes an active contour model (snake) based on image fusion, which is used in object detection and object tracking. Firstly, a multi-resolution image fusion with wavelet transform is applied to obtain the multi-resolution fused images. Secondly, in the low-frequency sub-image, a snake is applied with the Sobel operator to detect the object's contour; and in other high-frequency sub-images, a wavelet-based snake is applied. The two convergent snakes are fused in a multi-resolution scheme to obtain a fused snake in the fused image in an original resolution. Experiment results indicate this fused snake's detection or tracking accuracy is improved greatly.  相似文献   

设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号