A traffic surveillance system using dynamic saliency map and SVM boosting

This paper proposes a traffic surveillance system that can efficiently detect an interesting object and identify vehicles and pedestrians in real traffic situations. The proposed system consists of a moving object detection model and an object identification model. A dynamic saliency map is used for analyzing dynamics of the successive static saliency maps, and can localize an attention area in dynamic scenes to focus on a specific moving object for traffic surveillance purposes. The candidate local areas of a moving object are followed by a blob detection processing including binarization, morphological closing and labeling methods. For identifying a moving object class, the proposed system uses a hybrid of global and local information in each local area. Although the global feature analysis is a compact way to identify an object and provide a good accuracy for non-occluded objects, it is sensitive to image translation and occlusion. Therefore, a local feature analysis is also considered and combined with the global feature analysis. In order to construct an efficient classifier using the global and local features, this study proposes a novel classifier based on boosting of support vector machines. The proposed object identification model can identify a class of moving object and discard unexpected candidate area which does not include an interesting object. As a result, the proposed road surveillance system is able to detect a moving object and identify the class of the moving object. Experimental results show that the proposed traffic surveillance system can successfully detect specific moving objects.     

多标签传播和边缘局部分析的显著性检测

针对显著性检测中显著目标周围模糊的现象,提出了一种基于多标签传播和边缘局部分析的图像显著性检测算法。首先提取最信赖的目标和背景标签,通过基于元胞更新策略的多标签传播算法,准确定位显著目标;然后提出了一种新颖的分析边缘局部线索并完成分类标记的策略,以消除目标边缘的模糊现象。公开数据集上的实验结果显示,检测结果有效地抑制了目标周围的背景,解决了目标周围的模糊问题,清晰均匀地突出了显著目标。     

传感环境下基于局部Voronoi图的启发式反监控路径发现算法

针对移动对象通过传感区域时的安全问题,提出了一种基于局部Voronoi图(VT)的启发式反监控路径发现算法.首先,给出了一种基于局部Voronoi图的路径暴露风险近似估算模型.在该模型中,移动目标可依据当前探测到的传感器节点位置信息动态生成局部Voronoi图,并可依据定义的暴露风险计算公式近似估算出局部Voronoi图中各条边所对应路径的暴露风险.然后,在此基础上设计并实现了一种启发式的反监控路径发现算法.在该算法中,移动目标可首先基于局部Voronoi图确定自己的下一跳位置点候选集,然后再基于定义的启发式代价函数从候选集中选择一个风险代价最小的位置点作为其下一跳目标位置点.最后,沿着局部Voronoi图中对应的最小暴露风险路径移动到该目标位置点.理论分析和实验结果表明,所提算法具有良好的反监控性能,针对部署有n个传感器节点的区域,能够使得移动对象在不超过O(n log n)的时间内快速找到一条具有较低暴露风险的路径来穿越整个传感区域.     

一种基于轮廓特征点的图像检索方法

传统基于形状的图像检索方法检索效率较低,针对该问题,提出一种基于对象轮廓特征点的图像检索方法。利用Mean Shift算法提取感兴趣对象,以对象曲率的局部极值点作为特征点,并将对象表示为这些特征点的特征向量,定义检索对象与被检索对象特征向量间的距离匹配机制,实现对象的匹配或识别。实验结果表明,与传统方法相比,该方法具有较高的查全率和查准率。     

基于全局和局部约束直推学习的鲁棒跟踪研究

在目标跟踪中, 大部分算法都是假设目标亮度不变或者目标子空间不变, 然而, 这些假设在实际场景中并不一定满足, 特别是当目标和背景都发生较大变化时, 目标容易丢失. 针对这种情况, 本文从直推学习的角度重新描述跟踪问题, 并提出一种鲁棒的目标跟踪方法.为获得更好的跟踪效果, 目标当前状态估计不仅要逼近目标模型, 而且要与以前的结果具有相同的聚类. 本方法利用目标模型对跟踪问题进行全局约束, 利用以前的结果约束状态局部分布, 构造代价函数. 将以前的状态估计作为正样本, 当前的候选状态作为未标记样本, 以所有样本为顶点建立图, 同时学习目标的全局外观模型和所有状态的局部聚类结构. 最后利用图拉普拉斯, 通过简单的线性代数运算, 获得代价函数的最优解. 在实验中, 选取包含各种情形的视频, 如目标的姿势改变、表情变化、部分遮挡以及周围光照的变化等, 利用本文提出的方法测试, 并和其他算法比较. 实验结果表明, 本文方法能够很好处理这些情形, 实现对目标的鲁棒跟踪.     

Graph-based transductive learning for robust visual tracking

In object tracking problem, most methods assume brightness constancy or subspace constancy, which are violated in practice. In this paper, the object tracking problem is considered as a transductive learning problem and a robust tracking method is proposed under intrinsic and extrinsic varieties. The object not only fits the object model, but also has the same cluster with the previous objects, which are the labeled data. By constraining the global and local information, the cost function is constructed firstly. The solution for minimizing the cost function can be solved by a simple linear algebra with graph Laplacian. Moreover, a novel graph is constructed over the positive samples and candidate patches, which can simultaneously learn the object's global appearance model and the local intrinsic geometric structure of all the patches. Furthermore, a heuristic positive samples selection scheme is adopted to make the method more effective. The proposed method is tested on different videos, which undergo large pose, expression, illumination and partial occlusion, and compared with state-of-the-art algorithms. Experimental results and comparative studies are provided to demonstrate the efficiency of the proposed method.     

Dynamic obstacle identification based on global and local features for a driver assistance system

This paper proposes a novel dynamic obstacle recognition system combining global feature with local feature to identify vehicles, pedestrians and unknown backgrounds for a driver assistance system. The proposed system consists of two main procedures: a dynamic obstacle detection model to localize an area containing a moving obstacle, and an obstacle identification model, which is a hybrid of global and local information, for recognizing an obstacle with and without occlusion. A dynamic saliency map is used for localizing an area containing a moving obstacle. For the global feature analysis, we propose a modified GIST using orientation features with MAX pooling, which is robust to translation and size variations of an object. Although the global features are a compact way to represent an object and provide a good accuracy for non-occluded objects, they are sensitive to image translation and occlusion. Thus, a local feature-based identification model is also proposed and combined with the global feature. As such, for the obstacle identification problem, the proposed system mainly follows the global feature-based object identification. If the global feature-based model identifies a candidate area as background, the system verifies the area again using the local feature-based model. As a result, the proposed system is able to provide information on both the appearance of obstacles and the class of an obstacle. Experimental results show that the proposed model can successfully detect obstacle candidates and robustly identify obstacles with and without occlusion.     

Combinatorial manifold mesh reconstruction and optimization from unorganized points with arbitrary topology

In this paper, a novel approach is proposed to reliably reconstruct the geometric shape of a physically existing object based on unorganized point cloud sampled from its boundary surface. The proposed approach is composed of two steps. In the first step, triangle mesh structure is reconstructed as a continuous manifold surface by imposing explicit relationship among the discrete data points. For efficient reconstruction, a growing procedure is employed to build the 2-manifold directly without intermediate 3D representation. Local and global topological operations with ensured completeness and soundness are defined to incrementally construct the 2-manifold with arbitrary topology. In addition, a novel criterion is proposed to control the growing process for ensured geometric integrity and automatic boundary detection with a non-metric threshold. The reconstructed manifold surface captures the object topology with the built-in combinatorial structure and approximates the object geometry to the first order. In the second step, new methods are proposed to efficiently obtain reliable curvature estimation for both the object surface and the reconstructed mesh surface. The combinatorial structure of the triangle mesh is then optimized by changing its local topology to minimize the curvature difference between the two surfaces. The optimized triangle mesh achieves second order approximation to the object geometry and can serve as a basis for many applications including virtual reality, computer vision, and reverse engineering.     

A novel distribution-based feature for rapid object detection

The discriminative power of a feature has an impact on the convergence rate in training and running speed in evaluating an object detector. In this paper, a novel distribution-based discriminative feature is proposed to distinguish objects of rigid object categories from background. It fully makes use of the advantage of local binary pattern (LBP) that specializes in encoding local structures and statistic information of distribution from training data, which is utilized in getting optimal separating hyperplane. The proposed feature maintains the merit of simplicity in calculation and powerful discriminative ability to distinguish objects from background patches. Three LBP-based features are derived to adaptive projection ones, which are more discriminative than original versions. The asymmetric Gentle Adaboost organized in nested cascade structure constructs the final detector. The proposed features are evaluated on two different object categories: frontal human faces and side-view cars. Experimental results demonstrate that the proposed features are more discriminative than traditional Haarlike features and multi-block LBP (MBLBP) features. Furthermore they are also robust in monotonous variations of illumination.     

Object recognition using discriminative parts

The existing object recognition methods can be classified into two categories: interest-point-based and discriminative-part-based. The interest-point-based methods do not perform well if the interest points cannot be selected very carefully. The performance of the discriminative-part-base methods is not stable if viewpoints change, because they select discriminative parts from the interest points. In addition, the discriminative-part-based methods often do not provide an incremental learning ability. To address these problems, we propose a novel method that consists of three phases. First, we use some sliding windows that are different in scale to retrieve a number of local parts from each model object and extract a feature vector for each local part retrieved. Next, we construct prototypes for the model objects by using the feature vectors obtained in the first phase. Each prototype represents a discriminative part of a model object. Then, we establish the correspondence between the local parts of a test object and those of the model objects. Finally, we compute the similarity between the test object and each model object, based on the correspondence established. The test object is recognized as the model object that has the highest similarity with the test object. The experimental results show that our proposed method outperforms or is comparable with the compared methods in terms of recognition rates on the COIL-100 dataset, Oxford buildings dataset and ETH-80 dataset, and recognizes all query images of the ZuBuD dataset. It is robust enough for distortion, occlusion, rotation, viewpoint and illumination change. In addition, we accelerate the recognition process using the C4.5 decision tree technique, and the proposed method has the ability to build prototypes incrementally.     

基于核函数的高阶特征表示方法

为了解决高阶局部特征带来的计算复杂度提高问题,提出一种基于核函数的高阶局部特征表示方法。通过在两幅图像的局部特征之间进行比较,将特征空间映射到几何不变空间,统计高阶局部特征构建核函数,并结合支持向量机进行多类目标图像分类实验。实验结果分析表明,该方法在提高分类准确率的同时,所需的计算时间只与局部特征的个数呈线性增长。     

Image representation for generic object recognition using higher-order local autocorrelation features on posterior probability images

This paper presents a novel image representation method for generic object recognition by using higher-order local autocorrelations on posterior probability images. The proposed method is an extension of the bag-of-features approach to posterior probability images. The standard bag-of-features approach is approximately thought of as a method that classifies an image to a category whose sum of posterior probabilities on a posterior probability image is maximum. However, by using local autocorrelations of posterior probability images, the proposed method extracts richer information than the standard bag-of-features. Experimental results reveal that the proposed method exhibits higher classification performances than the standard bag-of-features method.     

Target tracking for mobile robot platforms via object matching and background anti-matching

This study proposes a novel method for target tracking based on the combination of object matching and background anti-matching which take account of both the global property of covariance matching and local property of mean shift tracking synthetically. In the background anti-matching phrase, a certain number of background regions are extracted based on the feature of color orientation codes via an entropy filter, and the covariance matrix is adapted to match these regions to get the global motion of the background; further, the object matching is carried out by a mean-shift tracking algorithm. The proposed method is evaluated in various datasets in comparison with their counterpart algorithms; experimental results sufficiently demonstrate the effectiveness of the method proposed in this study.     

多模型协作的分块目标跟踪

为解决复杂场景下,基于整体表观模型的目标跟踪算法容易丢失目标的问题,提出一种多模型协作的分块目标跟踪算法.融合基于局部敏感直方图的产生式模型和基于超像素分割的判别式模型构建目标表观模型,提取局部敏感直方图的亮度不变特征来抵制光照变化的影响;引入目标模型的自适应分块划分策略以解决局部敏感直方图算法缺少有效遮挡处理机制的问题,提高目标的抗遮挡性;通过相对熵和均值聚类度量子块的局部差异置信度和目标背景置信度,建立双权值约束机制和子块异步更新策略,在粒子滤波框架下,选择置信度高的子块定位目标.实验结果表明,本文方法在复杂场景下具有良好的跟踪精度和稳定性.     

基于粒子滤波的红外运动目标跟踪

提出一种基于粒子滤波及Mean Shift算法的红外运动目标跟踪方法。该方法首先利用目标区域的灰度分布,建立了一种基于统计直方图的系统观测概率模型,并针对红外目标机动性强,需要大量粒子才能保证算法鲁棒性的问题,将Mean Shift算法引入到粒子更新的过程中,使粒子分布在观测的局部区域内,在利用少量粒子实现分布多样性的同时,有效克服了粒子退化现象。序列图像的实验表明：该算法能够在目标高速运动或发生遮挡的情况下稳健跟踪目标,其总体性能优于传统的粒子滤波算法。     

2.5D face recognition using Patch Geodesic Moments

In this paper, we propose a novel Patch Geodesic Distance (PGD) to transform the texture map of an object through its shape data for robust 2.5D object recognition. Local geodesic paths within patches and global geodesic paths for patches are combined in a coarse to fine hierarchical computation of PGD for each surface point to tackle the missing data problem in 2.5D images. Shape adjusted texture patches are encoded into local patterns for similarity measurement between two 2.5D images with different viewing angles and/or shape deformations. An extensive experimental investigation is conducted on 2.5 face images using the publicly available BU-3DFE and Bosphorus databases covering face recognition under expression and pose changes. The performance of the proposed method is compared with that of three benchmark approaches. The experimental results demonstrate that the proposed method provides a very encouraging new solution for 2.5D object recognition.     

Automatic sub-category partitioning and parts localization for learning a robust object model

This paper introduces a novel topic model for learning a robust object model. In this hierarchical model, the layout topic is used to capture the local relationships among a limited number of parts when the part topic is used to locate the potential part regions. Naturally, an object model is represented as a probability distribution over a set of parts with certain layouts. Rather than a monolithic model, our object model is composed of multiple sub-category models designed to capture the significant variations in appearance and shape of an object category. Given a set of object instances with a bounding box, an iterative learning process is proposed to divide them into several sub-categories and learn the corresponding sub-category models without any supervision. Through an experiment in object detection, the learned object model is examined and the results highlight the advantages of our present method compared with others.