一种改进的角点探测方法

提出了一种改进的角点提取算法。该算法是基于角点的性质，即在角点的某个邻域内，图像灰度变化在任意一条通过该点的直线上都很大。本文利用此性质对象素点直线方向上灰度变化值求取极值，建立角点候选点。再通过SUSAN特征检测原则进一步剔除虚假角点。测试表明改进后的算法具有运算量小，定位精度高，抗噪能力强的特点，易于实时实现。     

基于边缘细化的角点提取算法

图像的角点提取在计算机视觉领域中具有重要的应用价值。本文基于角点一定是边缘点且是两条边缘或多条边缘交点的这一几何特性，提出了一种基于边缘细化的角点提取算法。算法先用Canny算子提取图像的边缘，然后进行细化并填充边界线上的断点，再利用角点的几何特性来确定角点的位置。实验结果表明，这种算法提高了角点提取的正确率、定位精确、实用性强。     

基于USAN的改进的角点检测算法

提出了一种基于USAN的改进的角点检测算法。该算法在原有SUSAN算法的基础上做了如下改进：使用一个3×3的方形预检测窗口对图像的像素进行预检测，在精确检测角点前剔除掉大部分的背景点、边界点及脉冲噪声点，提高了算法的效率；根据图像不同区域对比度不同的特性，采用根据对比度自动调节核心点与其邻域像素的灰度差值门限的方法，使所检测出的角点分布均匀；利用基于USAN定义的角点所应具有的特征(角的边缘及USAN的连续性)来剔除伪角点，降低了角点虚报和漏检的发生率。仿真实验证明了该文所提出的算法具有抗噪能力强、运算量小的特点，适于实时实现。     

视频帧中提取文字区域的算法

提出了一种在视频帧中提取文字区域的算法，该算法基于文字的纹理特征——角点，即文字，特别是汉字，含有丰富的角点。首先，在视频帧中进行角点检测，得到视频帧的角点分布图。第2步是进行角点滤波，即在角点分布图中剔除一些比较分散的，可以认为不是从文字中检测到的角点。第3步对角点分布图中余下的角点聚类形成候选文字区域，采用水平方向和竖直方向的投影对这些区域进行分化。第4步文字区域的最后确认，除去噪声。实验表明该算法计算速度快，提取文字区域准确。     

基于边界的角点无阈值识别算法

提出了局部特征和全局特征相结合的角点无阈值识别的思想和算法，同时提出了4个重要的概念及其算法：边界局部角度、相似度、高度、线性度，作为实现的基础。角度和相似度描述角点的局部特征；高度和线性度描述角点的全局特征，4者的拟合函数为F。F变化率的最大值点自动确定角点和非角点的分界。实验结果证实了该算法角点定位的准确性、旋转不变性，具有一定的尺度不变性和抗噪声能力。     

一种新的快速鲁棒性角点检测算法

对现有的角点检测算法进行了综述，着重分析了SUSAN算法和MIC算法的实现原理以及检测性能的优劣，提出了一种基于像素亮度变化的快速角点检测算法。对模拟图像和真实图像的角点进行了实验，并与目前应用比较广泛的Harris、MIC以及SUSAN角点检测子进行了比较。实验结果和数据分析表明了算法在计算效率、定位精度和鲁棒性等到方面具有良好性能。     

数字影像中目标角点定位的无阈值算法

在分析现有角点定位算子的基础上，针对现有大部分角点定位算子在角点定位过程中需要选择适当的阈值，而阈值的选择本身比较困难的这一关键问题，提出了一种基于移动小窗口内最大值的无阈值的角点定位算法的基本思想。并通过实验来证明算法的有效性，实验结果表明，这种算法在理想的情况下具有非常好的角点定位效果，而对于复杂的非均质图像来说也有较好的定位效果。     

一种便携式葡萄剪枝机器设计与实现

为进一步提升葡萄枝条修剪效果，提出一种便携式的葡萄剪枝机器。其中，在构建葡萄剪枝机主要架构基础上，使用Rosenfied细化算法作为图像骨架提取的主要方法，然后结合Harris角点算法进行角点检测，最后使用本研究提出的剪枝定位算法进行剪枝定位。仿真结果表明，与其他的细化算法相比，本研究使用的Rosenfied细化算法能够实现效果更好的图像骨架提取；将细化算法与角点检测算法相结合，对芽点、伪芽点和角点进行检测的正检率分别为79.5%、80%、94.4%;使用本研究提出的剪枝定位算法进行剪枝定位取得了良好的效果。以上仿真结果表明，本研究提出的便携式葡萄剪枝机器能够进行效果良好的稳定完成葡萄剪枝工作，进一步优化剪枝效果，具有一定的实际工程价值。     

基于角点跟踪的光流场计算

在综合视觉运动分析中的两类处理方法，选取图像中的角点作为特征点，并检测和跟踪图像序列中的角点。记录检测到的角点在图像序列中的位移，在理论上证明了时变图像的光流场可以近似地用角点的位移场代替，同时给出这种替代的两个前提条件。本文用真实图像序列验证提出的算法，实验结果表明该算法取得了较好的效果。     

一种快速自适应RSUSAN角点检测算法

根据图像边缘灰度的渐变性，我们重新定义SUSAN(Small Univalue Segment Assimilating Nucleus)算法中小核值相似区；并找到一种更为有效和简便的计算小核值相似区面积的方法；在此基础上提出了RSUSAN(Redefined SUSAN)角点检测算法。与经典的角点检测算法SUSAN、MIC(Minimum Intensity Change)相比，RSUSAN具有角点检测准确性高，计算简单，运算速度大为提高等优点。对于模糊、噪声大的图像本文还进一步提出了采用自适应平滑和RSUSAN相结合的方法，称为自适应RSUSAN算法。实验证明，相比较SUSAN、MIC算法而言，自适应RSUSAN算法没有显著地增加计算量，而且在对模糊、噪声大的图像进行角点检测时，虚报及漏检概率大大减少，对噪声的鲁棒性好，角点检测位置精确。     

Scale-based detection of corners of planar curves

A technique for detecting and localizing corners of planar curves is proposed. The technique is based on Gaussian scale space, which consists of the maxima of absolute curvature of the boundary function presented at all scales. The scale space of isolated simple and double corners is first analyzed to investigate the behavior of scale space due to smoothing and interactions between two adjacent corners. The analysis shows that the resulting scale space contains line patterns that either persist, terminate, or merge with a neighboring line. Next, the scale space is transformed into a tree that provides simple but concise representation of corners at multiple scales. Finally, a multiple-scale corner detection scheme is developed using a coarse-to-fine tree parsing technique. The parsing scheme is based on a stability criterion that states that the presence of a corner must concur with a curvature maximum observable at a majority of scales. Experiments were performed to show that the scale space corner detector is reliable for objects with multiple-size features and noisy boundaries and compares favorably with other corner detectors tested     

Space corner smoothing of CNC machine tools through developing 3D general clothoid

Tool paths defined by G01/G02/G03 commands need to be smoothed to eliminate the discontinuities of velocity, acceleration and jerk at the junction points. Because of inherent problems of curve fillets, traditional corner smoothing strategies are limited to the smoothing of corners in plane. This article presents a method to smooth the space corners through blending the position, tangent, curvature and sharpness of the adjacent trajectory segments based on 3D general clothoid splines, which are analytically developed by proposing 3D clothoid with G³ continuity. The 3D general clothoid realizes extending the traditional clothoid from 2-dimension to 3-dimension, and reserves some good properties of the traditional clothoid, i.e. the curve length parameterization and the analytically expressed curvature. It can also achieve higher degree of continuity compared to the traditional 2D clothoid. Based on the proposed 3D general clothoid, a corner smoothing algorithm, which is suitable for the smoothing of any planar or space line-line, line-arc or arc-arc pairs with G³ continuity, is proposed. At the same time, a smoothing-error-constraining-algorithm is developed to constrain the smoothing error within the tolerance. Simulation and experimental results, which are obtained from the smoothing of trajectories containing space corners, trajectories containing planar corners with G02/G03 commands, and trajectories containing planar corners with only G01 commands, demonstrate that the proposed corner smoothing algorithm can effectively increase the machining quality and efficiency.     

Direct curvature scale space: theory and corner detection

The curvature scale space (CSS) technique is considered to be a modern tool in image processing and computer vision. direct curvature scale space (DCSS) is defined as the CSS that results from convolving the curvature of a planar curve with a Gaussian kernel directly. In this paper we present a theoretical analysis of DCSS in detecting corners on planar curves. The scale space behavior of isolated single and double corner models is investigated and a number of model properties are specified which enable us to transform a DCSS image into a tree organization and, so that corners can be detected in a multiscale sense. To overcome the sensitivity of DCSS to noise, a hybrid strategy to apply CSS and DCSS is suggested     

一种基于曲率尺度空间的自适应角点检测方法

角点是图像处理中的重要特征,基于曲率尺度空间技术,提出一种自适应角点检测方法.首先提取图像的轮廓,采用一个固定的低尺度计算所有轮廓点的曲率,并根据曲率得到候选角点集,然后用由自适应支持区域确定的角点角度和一个动态曲率阙值代替固定的阙值筛选出正确角点.实验结果证实该方法应用于复杂图像的精确性和稳定性.     

曲率尺度空间与链码方向统计的角点检测

针对曲率尺度空间角点检测中,由于选择的尺度不同,会造成角点的漏检测,以及检测到错误角点的问题。提出一种基于曲率尺度空间与链码方向统计的角点检测方法。首先在较低的曲率尺度空间上检测出候选角点集;再通过自适应阈值及链码方向统计的方法删除错误角点。该方法采用较低的曲率尺度可检测出更多的角点,降低了角点漏检测率;通过计算椭圆角点自适应阈值可删除椭圆角点;采用Freeman链码方向统计可剔除伪角点;从而提高角点检测精度。通过实验充分验证了本文提出的角点检测算法比其他角点检测算法具有的高效性和准确性。     

Robust Image Corner Detection Based on the Chord-to-Point Distance Accumulation Technique

Many contour-based image corner detectors are based on the curvature scale-space (CSS). We identify the weaknesses of the CSS-based detectors. First, the “curvature” itself by its “definition” is very much sensitive to the local variation and noise on the curve, unless an appropriate smoothing is carried out beforehand. In addition, the calculation of curvature involves derivatives of up to second order, which may cause instability and errors in the result. Second, the Gaussian smoothing causes changes to the curve and it is difficult to select an appropriate smoothing-scale, resulting in poor performance of the CSS corner detection technique. We propose a complete corner detection technique based on the chord-to-point distance accumulation (CPDA) for the discrete curvature estimation. The CPDA discrete curvature estimation technique is less sensitive to the local variation and noise on the curve. Moreover, it does not have the undesirable effect of the Gaussian smoothing. We provide a comprehensive performance study. Our experiments showed that the proposed technique performs better than the existing CSS-based and other related methods in terms of both average repeatability and localization error.      

保护角点的非线性扩散滤波

基于偏微分方程的非线性扩散滤波可以有效去除噪声,同时很好地保护边缘信息,但角点在其滤波过程中逐渐变得圆滑甚至消失。通过设计扩散系数为梯度模及曲率的函数,提出了一种保护角点的非线性扩散滤波模型。该模型在区域内部为普通的高斯滤波,而在边缘处滤波效果降低,在角点处滤波效果进一步降低,从而在滤波过程中角点信息可以得到较好的保留,还可能出现角点及边缘相对增强的效果。实验结果的视觉效果及数据分析都表明,新模型可以同时保护边缘及角点信息。     

二值图像中基于陡变度的拐角点集检测方法

文章认为目前流行的拐角点检测方法有三个缺点：一是在数字图像空间中曲率存在计算误区,因而用曲率定位拐角点不合适;二是导致图像边缘产生拐角的不仅有单点,还有点集,因此不求拐角点集是不妥的;三是大曲率点不等价于拐角点。鉴于此,文章提出一种基于陡变度的拐角点集检测方法,其思想是：在二值图像中,图像边缘可看成是一维流形,陡变点集将一维流形分割成大小不等的光滑流形段,如果光滑流形通过该陡变点集后方向发生急剧改变,则此陡变点集是拐角点集。通过实验对比,文章中提出的算法检测结果优于目前流行算法的检测结果。