印制电路板图像对准算法研究

使用模板匹配法检测印刷电路板(PCB)质量,对标准图和待检测图之间的图像对准具有很高要求.在定位标志为圆的情况下,应用Hough变换虽然行之有效,但存在占用内存多,计算量大等问题.图像对准分为模糊对准和精确对准两个步骤,利用设置定位圆的先验知识实现模糊对准,并由模糊对准给出定位圆的相关参数,将这些参数运用于精确对准,能使图像的边缘提取和Hough变换变得简单、易行,减少了内存占用,提高了算法的处理速度.     

A new level set method for inhomogeneous image segmentation

Intensity inhomogeneity often appears in medical images, such as X-ray tomography and magnetic resonance (MR) images, due to technical limitations or artifacts introduced by the object being imaged. It is difficult to segment such images by traditional level set based segmentation models. In this paper, we propose a new level set method integrating local and global intensity information adaptively to segment inhomogeneous images. The local image information is associated with the intensity difference between the average of local intensity distribution and the original image, which can significantly increase the contrast between foreground and background. Thus, the images with intensity inhomogeneity can be efficiently segmented. What is more, to avoid the re-initialization of the level set function and shorten the computational time, a simple and fast level set evolution formulation is used in the numerical implementation. Experimental results on synthetic images as well as real medical images are shown in the paper to demonstrate the efficiency and robustness of the proposed method.     

An efficient alignment algorithm for masked sequences

We consider the alignment problem where sequences may have masked regions. The bases in masked regions are either unspecified or unknown, and they will be denoted by N. We present an efficient algorithm that finds an optimal local alignment by skipping such masked regions of sequences. Our algorithm works for both the affine gap penalty model and the linear gap penalty model. The time complexity of our algorithm is O((n−T)(m−S)+vm+wn) time, where n and m are the lengths of given sequences A and B, T and S are the numbers of base N in A and B, and v and w are the numbers of masked regions in A and B, respectively.     

一种改进的无人机影像拼接方法

SIFT算子在无人机影像拼接中有着广泛的应用,针对传统的基于SIFT的影像拼接处理速度慢的问题,通过分析原始影像与缩小影像之间单应矩阵的关系,提出了一种改进拼接速度的方法。分析了SIFT特征检测时各种参数的作用,设计了一种选择拼接参数的机制。实验结果表明,该方法可以根据影像数据的特点及应用需求设计出合适的拼接参数,有效地提高拼接速度,保证较好的拼接效果。     

基于多视角多监督网络的无人机图像定位方法

针对现有跨视角图像匹配算法精度低的问题,提出了一种基于多视角多监督网络（MMNet）的无人机（UAV）定位方法。首先,所提方法融合卫星视角和UAV视角,在统一的网络架构下学习全局和局部特征并以多监督方式训练分类网络并执行度量任务。具体来说,MMNet主要采用了重加权正则化三元组损失（RRT）学习全局特征,该损失利用重加权和距离正则化加权策略来解决多视角样本不平衡以及特征空间结构紊乱的问题。同时,为了关注目标地点中心建筑的上下文信息,MMNet对特征图进行方形环切割来获取局部特征。然后,分别用交叉熵损失和RRT执行分类和度量任务。最终,使用加权策略聚合全局和局部特征来表征目标地点图像。通过在当前流行的UAV数据集University-1652上进行实验,可知MMNet在UAV定位任务的召回率Recall@1 (R@1)及平均精准率（AP）上分别达到83.97%和86.96%。实验结果表明,相较于LCM、SFPN等方法,MMNet显著提升了跨视角图像的匹配精度,进而增强了UAV图像定位的实用性。     

A two level real-time path planning method inspired by cognitive map and predictive optimization in human brain

A biologically inspired two level method is proposed for real-time path planning in a complex and dynamic environment, employable in ground vehicles. This method takes the advantage of both global and local path finding procedures. In the first level, i.e., global level, the planner utilizes a neural network architecture as a sensory-motor map, similar to the cognitive map used by humans, and an optimization algorithm to produce a coarse path. In the second level, i.e., local level, the global path is improved by employing a model-based prediction method with a finite prediction horizon in a way that future information about the environment is involved in the planner's decision making. In the suggested method, the prediction horizon is variable and is adjusted in each step of the planning in agreement with the kinematic features of the closest obstacle in the visual field of the planner. We considered four different path planning tasks in a virtual dynamic environment to evaluate the performance of the proposed method against the human path planning strategy. The results demonstrate the ability of the method to plan a strategy comparable to the driving scenarios chosen by most subjects and to generate a real-time collision-free path in a dynamic environment with obstacles.     

A novel level set method for image segmentation by incorporating local statistical analysis and global similarity measurement

This paper presents a novel level set method for complex image segmentation, where the local statistical analysis and global similarity measurement are both incorporated into the construction of energy functional. The intensity statistical analysis is performed on local circular regions centered in each pixel so that the local energy term is constructed in a piecewise constant way. Meanwhile, the Bhattacharyya coefficient is utilized to measure the similarity between probability distribution functions for intensities inside and outside the evolving contour. The global energy term can be formulated by minimizing the Bhattacharyya coefficient. To avoid the time-consuming re-initialization step, the penalty energy term associated with a new double-well potential is constructed to maintain the signed distance property of level set function. The experiments and comparisons with four popular models on synthetic and real images have demonstrated that our method is efficient and robust for segmenting noisy images, images with intensity inhomogeneity, texture images and multiphase images.     

Iterative Grassmannian optimization for robust image alignment

Robust high-dimensional data processing has witnessed an exciting development in recent years. Theoretical results have shown that it is possible using convex programming to optimize data fit to a low-rank component plus a sparse outlier component. This problem is also known as robust PCA, and it has found application in many areas of computer vision. In image and video processing and face recognition, the opportunity to process massive image databases is emerging as people upload photo and video data online in unprecedented volumes. However, data quality and consistency is not controlled in any way, and the massiveness of the data poses a serious computational challenge. In this paper we present t-GRASTA, or “Transformed GRASTA (Grassmannian robust adaptive subspace tracking algorithm)”. t-GRASTA iteratively performs incremental gradient descent constrained to the Grassmann manifold of subspaces in order to simultaneously estimate three components of a decomposition of a collection of images: a low-rank subspace, a sparse part of occlusions and foreground objects, and a transformation such as rotation or translation of the image. We show that t-GRASTA is 4 × faster than state-of-the-art algorithms, has half the memory requirement, and can achieve alignment for face images as well as jittered camera surveillance images.     

A robust variable order facet model for image data

The underlying piecewise continuous surface of a digital image can be estimated through robust statistical procedures. This paper contains a systematic Monte Carlo study of M estimation and LMS estimation for image surface approximation, an examination of the merits of postprocessing and tuning various parameters in the robust estimation procedures, and a new robust variable order facet model paradigm. Several new goodness-of-fit measures are introduced and systematically compared. We show that the M estimation tuning parameters are not crucial, postprocessing is cheap and well worth the cost, and the robust algorithm for variable order facet models (using M estimation, new statistical goodness-of-fit measures, and postprocessing) manages to retain most of the statistical efficiency of M estimation, yet displays good robustness properties, and preserves the main geometric features of an image surface: step edges, roof edges, and corners.     

A new fast reversible method for image safe transfer

In this paper a novel reversible method for fast and safe image transfer is proposed. The method combines compression, data hiding and partial encryption of images in a single processing step. The proposed approach can embed data into the image according to the message size and partially encrypt the image and the message without changing the original image content. Moreover, during the same process the image is lossless compressed. Nevertheless, the compression rate depends on the upper bound of message size to embed in the image. The main idea is to decompose the original image into two sub-images and to apply various processes to each sub-image in order to gain space and increase the amount of embedded data. The two sub-images are then scrambled and partially encrypted. The most significant characteristic of the proposed method is the utilization of a single procedure to simultaneously perform the compression, the reversible data hiding and the partial encryption rather than using three separate procedures. Our approach reduces then the computational effort and the required computation time. This method is specially suited for medical images where one can associate the patient diagnostic to the concerned medical image for safe transfer purpose.     

基于SURF特征的航空序列图像位置估计算法

视觉传感器在航空无人机导航和定位任务中应用越来越广泛。针对无人机位置参数估计问题,提出了一种基于SURF特征的图像配准算法,该算法能够适应航空序列图像的旋转、尺度变换及噪声干扰,实现无人机位置的精确估计。构建了SURF尺度空间,运用快速Hessian矩阵定位极值点,计算出航空图像的64维SURF特征描述子;基于Hessian矩阵迹完成特征点匹配;使用RANSAC算法剔除出格点,实现位置参数的精确估计。通过航空图像序列实测数据位置估计实验,验证了该算法的有效性。     

ProMIS:概率图采样图像增广驱动的弱监督物体检测方法

目的 弱监督物体检测是一种仅利用图像类别标签训练物体检测器的技术。近年来弱监督物体检测器的精度不断提高,但在如何提升检出物体的完整性、如何从多个同类物体中区分出单一个体的问题上仍面临极大挑战。围绕上述问题,提出了基于物体布局后验概率图进行多物体图像增广的弱监督物体检测方法ProMIS（probability-based multi-object image synthesis）。方法 将检出物体存储到物体候选池,并将候选池中的物体插入到输入图像中,构造带有伪边界框标注的增广图像,进而利用增广后的图像训练弱监督物体检测器。该方法包含图像增广与弱监督物体检测两个相互作用的模块。图像增广模块将候选池中的物体插入一幅输入图像,该过程通过后验概率的估计与采样对插入物体的类别、位置和尺度进行约束,以保证增广图像的合理性;弱监督物体检测模块利用增广后的多物体图像、对应的类别标签、物体伪边界框标签训练物体检测器,并将原始输入图像上检到的高置信度物体储存到物体候选池中。训练过程中,为了避免过拟合,本文在基线算法的基础上增加一个并行的检测分支,即基于增广边界框的检测分支,该分支利用增广得到的伪边界框标注进行训练,原有基线算法的检测分支仍使用图像标签进行训练。测试时,本文方法仅使用基于增广边界框的检测分支产生检测结果。本文提出的增广策略和检测器的分支结构在不同弱监督物体检测器上均适用。结果 在Pascal VOC（pattern analysis, statistical modeling and computational learning visual object classes）2007和Pascal VOC 2012数据集上,将该方法嵌入到多种现有的弱监督物体检测器中,平均精度均值（mean average precision,mAP）平均获得了2.9%和4.2%的提升。结论 本文证明了采用弱监督物体检测伪边界框标签生成的增广图像包含丰富信息,能够辅助弱监督检测器学习物体部件、整体以及多物体簇之间的区别。     

A method of illumination compensation for human face image based on quotient image

Illumination variation is one of the critical factors affecting face recognition rate. A novel approach for human face illumination compensation is presented in this paper. It constructs the nine-dimension face illumination subspace based on quotient image. In addition, with the aim to improve algorithm efficiency, a half-face illumination image is proposed and the low-dimension training set of the face image under different illumination conditions are obtained by means of PCA and wavelet transform. After processing, two different illumination compensation strategies are given: one is adding light, and the other is removing light. Based on the illumination compensation strategy, we implement the typical illumination sample image synthesis and the standard illumination sample image synthesis on a PCA feature subspace and a wavelet transform subspace, respectively, and the illumination compensation of the gray images and the color images are further realized. Experimental results based on the Yale Face Database B, the Extended Yale Face Database B and the CAS-PEAL Face Database indicate that execution time after compensation is approximately half the time and face recognition rate is improved by 20% compared with that of the original images.     

海面图像单应的估算及数据归一化方法

根据海面图像的单应进行距离估算对于海洋工程具有重要意义。结合导管架入水参数观测问题,提出一种海面图像单应的估算方法。该方法以导管架上表面为参考平面,依据其单应矩阵及它与海平面之间的夹角计算出海面图像的单应。重点介绍了根据特征直线计算平面单应矩阵时的数据归一化方法。通过实验分析了这种归一化方法对平面远处距离估算的影响,并根据现场图片中的距离估算误差说明了在实际应用中的效果。该方法可以应用到多种海洋观测当中。     

A binarization method with learning-built rules for document images produced by cameras

In this paper, we propose a novel binarization method for document images produced by cameras. Such images often have varying degrees of brightness and require more careful treatment than merely applying a statistical method to obtain a threshold value. To resolve the problem, the proposed method divides an image into several regions and decides how to binarize each region. The decision rules are derived from a learning process that takes training images as input. Tests on images produced under normal and inadequate illumination conditions show that our method yields better visual quality and better OCR performance than three global binarization methods and four locally adaptive binarization methods.     

基于多层次分块的指纹图像二值化方法

基于块方向信息的指纹图像二值化方法,其效果主要取决于能否准确的估计指纹图像的块方向图.鉴于在指纹奇异点附近及方向变化较快的脊线、谷线区域,块方向信息很难精确得到,提出了一种多层次分块的指纹图像块方向的求取方法,并把这种方法用于指纹图像的二值化,取得了良好的效果.     

谱段交织预测的多光谱图像可逆信息隐藏算法

针对现有可逆信息隐藏算法嵌入容量不高并且不适合多光谱图像的问题,提出一种基于谱段交织预测的多光谱图像可逆信息隐藏算法.利用含秘图像像素值与原始图像像素值相差不大于1的特点,结合含秘谱段对多光谱图像进行多波段预测,通过预测误差直方图移位技术嵌入秘密信息.Landsat卫星和Terra卫星多光谱图像仿真结果表明,提出的算法和典型可逆信息隐藏算法相比具有更好的视觉质量和隐藏容量.     

A multi-scale 3D Otsu thresholding algorithm for medical image segmentation

Thresholding technique is one of the most imperative practices to accomplish image segmentation. In this paper, a novel thresholding algorithm based on 3D Otsu and multi-scale image representation is proposed for medical image segmentation. Considering the high time complexity of 3D Otsu algorithm, an acceleration variant is invented using dimension decomposition rule. In order to reduce the effects of noises and weak edges, multi-scale image representation is brought into the segmentation algorithm. The whole segmentation algorithm is designed as an iteration procedure. In each iteration, the image is segmented by the efficient 3D Otsu, and then it is filtered by a fast local Laplacian filtering to get a smoothed image which will be input into the next iteration. Finally, the segmentation results are pooled to get a final segmentation using majority voting rules. The attractive features of the algorithm are that its segmentation results are stable, it is robust to noises and it holds for both bi-level and multi-level thresholding cases. Experiments on medical MR brain images are conducted to demonstrate the effectiveness of the proposed method. The experimental results indicate that the proposed algorithm is superior to the other multilevel thresholding algorithms consistently.     

Feature‐based recursive observer design for homography estimation and its application to image stabilization

This paper presents a new algorithm for online estimation of a sequence of homographies applicable to image sequences obtained from robotic vehicles equipped with vision sensors. The approach taken exploits the underlying Special Linear group structure of the set of homographies along with gyroscope measurements and direct point‐feature correspondences between images to develop temporal filter for the homography estimate. Theoretical analysis and experimental results are provided to demonstrate the robustness of the proposed algorithm. The experimental results show excellent performance and robustness even in the case of very fast camera motions (relative to frame rate) and severe occlusions.     

Rectangular region coding for image data compression

A new data compression algorithm for a binary image is presented. This is fully 2-dimensional and has a hierarchical format for describing a macroscopic image structure.

An image is described as a set of uniform rectangles and a set of explicit point-by-point data which are outside of these rectangles. The basic problem involved is to find the largest possible rectangle in a given region. For this purpose an image pyramid is employed.

The simulation results of this algorithm show a very good compression ratio. This algorithm has many other interesting characteristics and can be applied for the image data storage, image edition and image transmission.