视频图像中的车辆检测跟踪和分类

介绍了一种用固定的单摄像头拍摄交通图像,并从图像序列中检测、跟踪、分类车辆的方法。该方法大致可分为3部分：抽取背景图像和图像分割;基于针孔模型的摄像机定标,计算透视投影矩阵;利用区域特征进行匹配跟踪,建立目标链。恢复目标三维信息,采用模型匹配法对车型分类。实验证明方法是简单可行的。     

结合分类与迁移学习的薄云覆盖遥感图像地物信息恢复

利用多源多时相遥感图像,给出一种结合分类与迁移学习的薄云覆盖遥感图像地物信息恢复算法.首先利用多方向非抽样对偶树复小波变换对多源多时相遥感图像进行多分辨率分解,对分解后的薄云图像的高频系数利用贝叶斯方法进行地物初分类;再对每类地物的低频系数通过迁移最小方差支持向量回归模型进行域自适应学习,获取模型参数;最后利用所获的迁移回归模型,用无云参考图像的低频系数预测薄云覆盖图像的低频系数,去除薄云,恢复薄云覆盖图像的地物信息.实验结果表明,本文算法恢复的地物细节清楚,光谱失真较小.特别对地物季节性变化的薄云覆盖遥感图像,本文算法能有效恢复薄云覆盖区域的地物信息.     

多尺度多特征融合的高分辨率遥感影像分类

针对高分辨率遥感影像多尺度、空间分布复杂以及特征繁多的特点,从遥感影像特征提取的尺度效应以及各类地物显著性特征各异入手,提出了基于多尺度多特征融合的高分辨率遥感影像分类的方法。该方法构建最优尺度分割函数模型,寻找出各地物的最优尺度,分别提取影像的纹理、颜色和形状特征。在此基础上,利用各地物特征的显著性差异实现多尺度下多特征的加权融合。该加权融合方法突破了常规的最优尺度分割算法未能充分考虑各类地物特征差异性的局限性,通过分析各类地物的显著性,建立了各个特征在分类中所占权重的模型。实验结果表明：相对传统无监督分类算法,该方法准确率提高约7%,且运行效率高。     

智能型目标识别系统

介绍了图像目标识别技术中的图像分割，不变性参数提取和目标分类，利用图像目标的均匀性和相应知识自适应地分割和提取图像目标，被提取的每个图像目标的不变性参数由归一化过程和Ｚｅｒｎｉｋｅ矩提取，并利用ＭＰＮＮ模型将图像目标分类，实验结果该识别系统能识别光照不均匀或复杂背景下的图像目标。     

2-D mesh-based video object segmentation and tracking with occlusion resolution

This paper integrates fully automatic video object segmentation and tracking including detection and assignment of uncovered regions in a 2-D mesh-based framework. Particular contributions of this work are (i) a novel video object segmentation method that is posed as a constrained maximum contrast path search problem along the edges of a 2-D triangular mesh, and (ii) a 2-D mesh-based uncovered region detection method along the object boundary as well as within the object. At the first frame, an optimal number of feature points are selected as nodes of a 2-D content-based mesh. These points are classified as moving (foreground) and stationary nodes based on multi-frame node motion analysis, yielding a coarse estimate of the foreground object boundary. Color differences across triangles near the coarse boundary are employed for a maximum contrast path search along the edges of the 2-D mesh to refine the boundary of the video object. Next, we propagate the refined boundary to the subsequent frame by using motion vectors of the node points to form the coarse boundary at the next frame. We detect occluded regions by using motion-compensated frame differences and range filtered edge maps. The boundaries of detected uncovered regions are then refined by using the search procedure. These regions are either appended to the foreground object or tracked as new objects. The segmentation procedure is re-initialized when unreliable motion vectors exceed a certain number. The proposed scheme is demonstrated on several video sequences.     

基于视频的车型分类算法研究

对视频图像处理中车型分类算法涉及的摄像机标定和目标分割算法进行研究。通过摄像机标定实现像素坐标到世界坐标的映射,进而准确获取车辆长度及宽度信息。建立车辆高度信息计算模型,获得车辆高度信息。对实时视频图像进行目标分割,并通过车辆长度、高度及长高比等信息实现高速公路中车型分类。经过验证,该方法能够比较准确地进行车型分类,满足实际应用要求。     

结合场景描述的文本生成图像方法

研究基于场景描述文本生成对应图像的方法,针对生成图像常常出现的对象重叠和缺失问题,提出了一种结合场景描述的生成对抗网络模型。首先,利用掩模生成网络对数据集进行预处理,为数据集中的对象提供分割掩模向量。然后,将生成的对象分割掩模向量作为约束,通过描述文本训练布局预测网络,得到各个对象在场景布局中的具体位置和大小,并将结果送入到级联细化网络模型,完成图像的生成。最后,将场景布局与图像共同引入到布局鉴别器中,弥合场景布局与图像之间的差距,得到更加真实的场景布局。实验结果表明,所提模型能够生成与文本描述更匹配的图像,图像更加自然,同时有效地提高了生成图像的真实性和多样性。     

Segmentation of Pedestrians with Confidence Level Computation

In this work, we propose a mechanism to segment groups of pedestrians with confidence level computation for intelligent surveillance systems. The goal is to specify the number of people and locate the position and size of each individual in groups of people. Human detection and clustering techniques are combined to achieve the segmentation purpose. The histogram of oriented gradients and curvelet features are extracted for full body detection using a support vector machine classifier. Modified Haar of Oriented Gradient features are constructed for upper body and lower body detectors. A clustering algorithm is then applied to the detected humans to eliminate the redundant detection responses. The proposed mechanism requires no prior assumptions of human sizes, human heights, camera distances, and other calibration parameters. Moreover, confidence level computation can provide valuable information for subsequent surveillance applications. The proposed approach is tested with pedestrian benchmark dataset and surveillance videos. The experimental results have demonstrated the effectiveness of the proposed pedestrian segmentation mechanism.     

Automatic objects segmentation with RGB-D cameras

Automatic object segmentation is a fundamentally difficult problem due to issues such as shadow, lighting, and semantic gaps. Edges play a critical role in object segmentation; however, it is almost impossible for the computer to know which edges correspond to object boundaries and which are caused by internal texture discontinuities. Active 3-D cameras, which provide streams of depth and RGB frames, are poised to become inexpensive and widespread. The depth discontinuities provide useful information for identifying object boundaries, which makes automatic object segmentation possible. However, the depth frames are extremely noisy. Also, the depth and RGB information often lose synchronization when the object is moving fast, due to different response time of the RGB and depth sensors. We show how to use the combined depth and RGB information to mitigate these problems and produce an accurate silhouette of the object. On a large dataset (24 objects with 1500 images), we provide both qualitative and quantitative evidences that our proposed techniques are effective.     

Towards 3-D scene reconstruction from broadcast video

Three-dimensional (3-D) scene reconstruction from broadcast video is a challenging problem with many potential applications, such as 3-D TV, free-view TV, augmented reality or three-dimensionalization of two-dimensional (2-D) media archives. In this paper, a flexible and effective system capable of efficiently reconstructing 3-D scenes from broadcast video is proposed, with the assumption that there is relative motion between camera and scene/objects. The system requires no a priori information and input, other than the video sequence itself, and capable of estimating the internal and external camera parameters and performing a 3-D motion-based segmentation, as well as computing a dense depth field. The system also serves as a showcase to present some novel approaches for moving object segmentation, sparse and dense reconstruction problems. According to the simulations for both synthetic and real data, the system achieves a promising performance for typical TV content, indicating that it is a significant step towards the 3-D reconstruction of scenes from broadcast video.     

一种快速红外图像分割方法

为准确地实现目标识别,提出了将二维最大熵图像分割方法应用于红外图像实行分割．利用图像的二维直方图,二维最大熵分割方法不仅考虑了象素的灰度信息,而且还充／矿利用了象素的空间领域信息,能取得较为理想的分割结果．然而该方法所需的巨大运算量限制了其实际应用．运用PSO算法代替穷尽搜索获得阈值向量,求解速度可提高300～400倍,提高了分割效率．通过对实际的红外图像分割表明,这种方法简单、有效．     

基于时空曲线演化的多视频运动对象分割算法

多视频对象由于其运动的复杂性,在分割提取过程中有较大的难度.本文提出了一种基于时空曲线演化的多视频对象自动分割方法,首先根据视频序列帧间(时间域)和帧内(空间域)信息的不同特点,建立基于全局和局部特征的能量模型,并由此导出基于level sets方法的曲线演化方程;然后用视频序列的连继两帧帧差得到初始的视频对象,分别进行时间和空间曲线演化跟踪,提取多个视频对象;当对象因运动而发生相互遮挡现象时,利用基于Bayes最小错误概率决策法则的判断方法,分割遮挡对象和显露对象.实验结果表明,本文提出算法的分割效果在空间准确度上比COST211算法提高30-50%,比最佳的帧差分割算法提高5-10%.     

一种基于相对模糊连通度的交互式序列图像快速分割算法

在基于相对模糊连通度的交互式图像分割方法的基础上提出了一种序列图像分割的快速算法。对于单幅图像,新算法在保持分割精度的同时,运行速度提高了3倍。作为该算法的扩展,一方面将原算法中单目标分割推广到多目标分割,另一方面将单帧图像的分割推广到序列图像的分割,实现了复杂背景下多目标的图像分割和图像序列的批处理分割。并进行了分割结果的后处理,提取出目标的单像素宽度的光滑边缘。用人造图像和实际的医学图像和图像序列所做的测试实验取得了令人满意的分割结果。     

Generation of high-quality depth maps using hybrid camera system for 3-D video

In this paper, we present a hybrid camera system combining one time-of-flight depth camera and multiple video cameras to generate multi-view video sequences and their corresponding depth maps. In order to obtain the multi-view video-plus-depth data using the hybrid camera system, we capture multi-view videos using multiple video cameras and a single view depth video with the depth camera. After performing a three-dimensional (3-D) warping operation to obtain an initial depth map at each viewpoint, we refine the initial depth map using segment-based stereo matching. To reduce mismatched depth values along object boundaries, we detect the moving objects using color difference between frames and extract occlusion and disocclusion areas with the initial depth information. Finally, we recompute the depth value of each pixel in each segment using pairwise stereo matching with a proposed cost function. Experimental results show that the proposed hybrid camera system produces multi-view video sequences with more accurate depth maps, especially along the boundary of objects. In addition, it is suitable for generating more natural 3-D views for 3-D TV than previous works..     

集成成像系统实现三维物体旋转不变分类识别

提出一种基于集成成像系统和综合判别函数(SDF)实现三维物体旋转不变分类识别的新方法。方法利用集成成像系统获取旋转三维物体的单元像阵列图像,应用图像中各个单元像记录的三维物体信息之间的高关联特性,结合SDF,实现了三维物体的旋转不变分类识别。与已有方法相比,本文方法减小了识别过程的数据量,降低运算复杂度,提高了识别效率,且不受物体旋转角度的限制。针对多类旋转三维物体,利用本文方法实现了旋转不变分类识别,验证了方法的有效性。     

基于遥感影像中电磁特性的地物提取方法

电磁特性数据是电子信息装备内场仿真试验不可或缺的数据资源,为了克服传统电子信息装备仿真试验中的电磁特性数据获取方法的不足,提出了从遥感影像中大批量获取大区域范围的电磁特性数据的新方法.首先依据不同地物的电磁特性规律对某一地域的地物进行了分类,从多尺度分割概念和基于异质性最小的区域合并算法两个方面.深入研究了针对地物分类的面向对象多尺度分割算法,并结合实践对算法进行了应用.由于遥感影像的真实性.保证了电磁特性数据提取的准确性与可靠性,实践证明此研究为大批量获取电磁特性数据提供了一种有效的方法和思路.     

基于直方图极大值准则可变参数图像阈值分割方法

提出了一种基于直方图极大值准更则图像阈值分割方法，通过调节参数可对多类图像进行分割。阉值分割是实时图像处理首选的分割算法。本文通过一维灰度直方图确定极大值，从而获得图像中多个目标区域和背景区域阉值，分割出多个区域，将每个区域赋为不同的灰度级，根据具体所要分割的目标再进行分割。实验证明该方法分割效果较好，分割速度快，实时性较好。     

显著性背景感知的多尺度红外行人检测方法

超大视场(U-FOV)红外成像系统探测范围大、不受光照限制,但存在尺度多样、小目标丰富的特点。为此该文提出一种具备背景感知能力的多尺度红外行人检测方法,在提高小目标检测性能的同时,减少冗余计算。首先,构建了4尺度的特征金字塔网络分别独立预测目标,补充高分辨率细节特征。其次,在特征金字塔结构的横向连接中融入注意力模块,产生显著性特征,抑制不相关区域的特征响应、突出图像局部目标特征。最后,在显著性系数的基础上构建了锚框掩膜生成子网络,约束锚框位置,排除平坦背景,提高处理效率。实验结果表明,显著性生成子网络仅增加5.94%的处理时间,具备轻量特性;超大视场(U-FOV)红外行人数据集上的识别准确率达到了93.20%,比YOLOv3高了26.49%;锚框约束策略能节约处理时间18.05%。重构模型具有轻量性和高准确性,适合于检测超大视场中的多尺度红外目标。     

A Method for 3D Scene Description and Segmentation in an Object Record

ＡＭｅｔｈｏｄｆｏｒ３ＤＳｃｅｎｅＤｅｓｃｒｉｐｔｉｏｎａｎｄＳｅｇｍｅｎｔａｔｉｏｎｉｎａｎＯｂｊｅｃｔＲｅｃｏｒｄ￥ＣｈｅｎＴｉｎｇｂｉａｏ（ＤｅｐａｒｔｍｅｎｔｏｆＲａｄｉｏＥｎｇｉｎｅｅｒｉｎｇ，ＮａｍｉｎｇＵｎｉｖｅｒｓｉｔｙｏｆＰｏｓｔｓ...