检索结果-内蒙古大学图书馆

An image-tracking algorithm based on object center distance-weighting and image feature recognition

Frontiers of Electrical and Electronic Engineering in China 2007年第1期2卷 1-7页

作者： JIANG Shuhong WANG Qin ZHANG Jianqiu HU Bo Department of Electronic Engineering Fudan UniversityShanghai 200433China

Areal-time image-tracking algorithm is proposed,which gives small weights to pixels farther from the object center and uses the quantized image gray scales as a *** identifies the target’s location by the mean-shift iteration method and arrives at the target’s scale by using image feature *** improves the kernel-based algorithm in tracking scale-changing targets.A decimation method is proposed to track large-sized targets and real-time experimental results verify the effectiveness of the proposed algorithm.

关键词： object tracking video processing mean-shift algorithm feature recognition

来源：评论

学校读者我要写书评

暂无评论

Two-level automatic multiple target joint tracking and classification

Two-level automatic multiple target joint tracking and class...

引用

Conference on Signal and Data Processing of Small Targets 2010

作者： He, Xiaofan Tharmarasa, Ratnasingham Pelletier, Michel Kirubarajan, Thia McMaster Univ Dept Elect & Comp Engn Hamilton ON Canada ICx Radar Syst Laval PQ Canada

ISBN: (纸本)9780819481627

Target classification is of great importance for modern tracking systems. The classification results could be fed back to the tracker to improve tracking performance. Also, classification results can be applied for target identification, which is useful in both civil and military applications. While some work has been done on Joint Tracking and Classification (JTC), which can enhance tracking results and make target identification feasible, a common assumption is that the statistical description of classes is predefined or known a prior. This is not true in general. In this paper, two automatic multiple target classification algorithms, which can automatically classify targets without prior information, are proposed. The algorithms learn the class description from the target behavior history. The input to the algorithm is the noisy target state estimate, which in turn depends on target class. Thus, class description is learnt from the target behavior history rather than being predefined. This motivates the proposed two-level tracking and classification formulation for automatic multiple target classification. The first level consists of common tracking algorithm such as the Joint Probability Data Association (JPDA), the Multiple Hypothesis Tracking (MHT) or the Probability Hypothesis Density (PHD) filter. In the second level, a mean-shift (MS) classifier and a PHD classifier are applied to learn the class descriptions respectively based on the state estimations from the first level tracker. The proposed algorithms only require the kinematic measurements from common radar. However, feature information can be easily integrated. Besides theoretical derivations, extensive experiments based on both simulated and real data are performed to verify the efficiency of the proposed technique.

关键词： Classification mean-shift algorithm Probability Hypothesis Density tracking multiple targets

来源：评论

学校读者我要写书评

暂无评论

Estimation of Two Illuminant Spectral Power Distributions from Highlights of Overlapping Illuminants

Estimation of Two Illuminant Spectral Power Distributions fr...

引用

9th International Conference on Signal-Image Technology and Internet-Based Systems (SITIS)

作者： Hang, Nguyen Thuc Dieu Horiuchi, Takahiko Hirai, Keita Tominaga, Shoji Chiba Univ Grad Sch Adv Integrat Sci Inage Ku Chiba 2638522 Japan

ISBN: (纸本)9781479932115

A method is proposed for estimating two illuminant spectral power distributions from the highlights of overlapping illuminants on an object. It was assumed in most previous studies that specular highlights on object surfaces are caused by a single light source, or by separate multiple light sources. In this paper, we assume that two highlights from different light sources are overlapped on object surfaces, and estimate the spectral power distributions of both. A multiband camera system is used for capturing spectral images of dielectric objects in a scene. First, we detect specular highlight areas from the spectral image. Then, the illuminant spectra of two light sources are simultaneously estimated based on cluster classification of particular pixel distribution in the highlight area. The feasibility of the proposed method is examined in experiments on real-world scenes.

关键词： illuminant estimation mean-shift algorithm multiple light sources specular highlight area

来源：评论

学校读者我要写书评

暂无评论

A Target Tracking algorithm for vision based Sea Cucumber Grabbing

A Target Tracking Algorithm for vision based Sea Cucumber Gr...

引用

IEEE International Conference on Information and Automation (ICIA)

作者： Wei, Honglei Peng, Danni Zhu, Xufei Wu, Dongdong Dalian Polytech Univ Sch Mech Engn & Automat Dalian Peoples R China

ISBN: (纸本)9781509041022

In order to accurately track sea cucumber on the assembly line to realize automatic grabbing using mechanical arm, an object tracking method based on mean-shift algorithm was proposed. Firstly, the contours of the objects was extracted from the original image to select tracking target, and then the local image was cropped at the same position and local area in the second frame, and mean-shift algorithm was used to search the center location until the searching process converged. Finally, tracking was realized by searching the center location in the next frames one by one. Compared with the color histogram mean-shift algorithm(CHMS) algorithm, - the proposed algorithm have good tracking performance and the average consuming time is only 11. 04% of the average consuming time of the CHMS algorithm, which prove that the proposed algorithm can satisfy the automatic grabbing requirement of real-time and robustness.

关键词： Target tracking mean-shift algorithm Sea Cucumber Automatic Grabbing

来源：评论

学校读者我要写书评

暂无评论

KERNEL-BASED STORM TRACKING IN RADAR DATA

KERNEL-BASED STORM TRACKING IN RADAR DATA

引用

2008 IEEE Radar Conference

作者： Picus, C. Beleznai, C. Nowak, C. Ramoser, H. Mitterhuber, S. Austrian Res Ctr GmbH A-1220 Vienna Austria Adv Comp Vis GmbH A-1220 Vienna Austria

ISBN: (纸本)9781424415380

The tracking of storm centres in radar data is of particular importance for short term weather prediction and specifically thunderstorm prediction. This paper presents a method to track storm centres in terrestrial radar images. mean shift segmentation is used to outline storm centres and mean shift tracking to locate the storm in the consecutive images. Re sults demonstrate the ability of the method to deal with deformable objects such as storm centres. Moreover the method is able to handle the splitting and merging of the convective cells.

关键词： nowcasting thunderstorm tracking mean-shift algorithm

来源：评论

学校读者我要写书评

暂无评论

Target tracking of the scale and rotation problem research

Target tracking of the scale and rotation problem research

引用

第25届中国控制与决策会议

作者： Ren Hongwei Zhao Qinjun Cheng Xingong Wang Chengyou Schools of Electrical Engineer University of Jinan

ISBN: (纸本)9781467355322

In order to solve the scale and rotation problem in target tracking, a target tracking algorithm based on scale invariant feature transform and mean shift is proposed. First, template area and the target area in the extreme value point of scale space are detected ,then accurate position and scale of the feature point are detected through the simulation of three dimensional quadratic function,after that feature points of the target area and template area are matched,and error matching is removed according to the continuity of scale and angle between adjacent frame. The experimental results show that the algorithm has good tracking performance when targets angle and scale changed.

关键词： target tracking mean-shift algorithm SIFI algorithm scale space

来源：评论

学校读者我要写书评

暂无评论

Robust Object Tracking using Local Oriented Energy Features and its Hardware/Software Implementation

Robust Object Tracking using Local Oriented Energy Features ...

引用

11th International Conference on Control, Automation, Robotics and Vision (ICARCV 2010)

作者： Norouznezhad, Ehsan Bigdeli, Abbas Postula, Adam Lovell, Brian C. NICTA POB 6020 St Lucia Qld 4067 Australia Univ Queensland Sch ITEE Brisbane Qld 4072 Australia

ISBN: (纸本)9781424478132

This paper presents the use of local oriented energy features for real-time object tracking on embedded vision systems. Local oriented energy features are extracted using complex Gabor filters. Filtering is carried out across multiple channels with different frequencies and orientations. The effectiveness of the chosen feature set is tested using a mean-shift tracker. Our experiments show that adding local oriented energy features can significantly enhance the performance of the tracker in presence of photometric variations and geometric transformation. The real-time implementation of the system is also described in this paper. To achieve the desired performance, a hardware/software co-design approach is pursued. Multi-channel Gabor filtering, Local Oriented Energy Feature and Feature histogram Computation is implemented on hardware while mean-shift vector calculation is performed on a processor. The system was synthesized onto a Xilinx Virtex-5 XC5VSX50T using Xilinx ML506 development board and the implementation results are presented.

关键词： Object Tracking Gabor Filters Local Oriented Energy Features mean-shift algorithm Field Programmable Gate Array (FPGA)

来源：评论

学校读者我要写书评

暂无评论

Probabilistic Primitive Refinement algorithm for Colored Point Cloud Data

Probabilistic Primitive Refinement Algorithm for Colored Poi...

引用

European Conference on Mobile Robots

作者： Ekekrantz, Johan Thippur, Akshaya Folkesson, John Jensfelt, Patric KTH Royal Inst Technol Ctr Autonomous Syst SE-10044 Stockholm Sweden

ISBN: (纸本)9781467391634

In this work we present the Probabilistic Primitive Refinement (PPR) algorithm, an iterative method for accurately determining the inliers of an estimated primitive (such as planes and spheres) parametrization in an unorganized, noisy point cloud. The measurement noise of the points belonging to the proposed primitive surface are modelled using a Gaussian distribution and the measurements of extraneous points to the proposed surface are modelled as a histogram. Given these models, the probability that a measurement originated from the proposed surface model can be computed. Our novel technique to model the noisy surface from the measurement data does not require a priori given parameters for the sensor noise model. The absence of sensitive parameters selection is a strength of our method. Using the geometric information obtained from such an estimate the algorithm then builds a color-based model for the surface, further boosting the accuracy of the segmentation. If used iteratively the PPR algorithm can be seen as a variation of the popular mean-shift algorithm with an adaptive stochastic kernel function.

关键词： Gaussian distribution image colour analysis iterative methods probability stochastic processes PPR algorithm adaptive stochastic kernel function color-based model colored point cloud data iterative method mean-shift algorithm measurement noise probabilistic primitive refinement algorithm sensitive parameters selection sensor noise model Estimation Histograms Image color analysis Noise measurement Robot sensing systems Surface fitting Three-dimensional displays Noise measurement image colour analysis Gaussian distribution iterative methods three-dimensional displays Manipulator sensing systems surface fitting histograms Noise models Markov chain surface modeling Primitive

来源：评论

学校读者我要写书评

暂无评论

An Automatic Food Recognition algorithm with both Shape and Texture Information

An Automatic Food Recognition Algorithm with both Shape and ...

引用

Conference on PIAGENG - Image Processing and Photonics for Agricultural Engineering

作者： Deng, Yu Qin, Shiyin Wu, Yunjie BeiHang Univ Sch Automat Sci & Elect Engn Beijing 100191 Peoples R China

ISBN: (纸本)9780819478009

Automatic food classification with digital images has played an important role in modern agricultural and food engineering. For this purpose, a kind of recognition algorithm for food is presented based on their shape and texture information in this paper.. By using a combination of shape and texture feature, improved mean-shift procedure is a state-of-the-art learning algorithm for multi-classification of food. The proposed method has four steps: (1) computation of a high contrast monochrome image from an optimal linear combination of RGB components of the food colour image;(2) a morphological shape detection operation is applied to detect the actual food shape from the high contrast monochrome image, some structural elements that have special forms are utilized to eliminate noise and improve detection precision;(3) a food texture is modeled by co-occurrence matrix;(4) a feature combination method is specified by food shape and texture information synthetically, then an improved mean-shift algorithm is proposed to achieve automatic food classification and recognition. The algorithm was implemented in Matlab and tested with 180 images (512x512) taken for various food with big differences. The algorithm can be applied to recognize food categories at the speed of 1.13s per image with the approval recognition rate of 97.6%. The result shows that our algorithm fully satisfies the requests of real application.

关键词： Food recognition Morphological shape detection Co-occurrence matrix Feature combination mean-shift algorithm

来源：评论

学校读者我要写书评

暂无评论

Augmentation in Object Tracking Performance with Hue Preservation conditioned on Frequency domain characteristics of video

Augmentation in Object Tracking Performance with Hue Preserv...

引用

IEEE International WIE Conference on Electrical and Computer Engineering (WIECON-ECE)

作者： Karamchandani, S. H. Thomas, Chris Phadke, G. Univ Mumbai DJ Sanghvi Coll Engn Bombay 400056 Maharashtra India Univ Mumbai Ramrao Adik Inst Technol Bombay 400706 Maharashtra India

ISBN: (纸本)9781467387866

Tracking of any given object forms integral part in surveillance, control and analysis applications. The video tracker presented here works on the principle of mean shift. mean shift is an iterative algorithm which is extended to the field of object tracking. However mean shift tracker losses track of the object when there are variations in illumination. In order to improve the performance of the tracker the poorly illuminated video frames are pre-processed and enhancement is provided to only those frames based on hue preservation algorithm. Low illumination in a frame leads to low value DC coefficient. Enhancement is provided to frames with low DC coefficient. The frames with corrected illumination are used to track object of interest using the mean shift algorithm. The proposed method is tried on a variety of video sequences. The results exhibit much improved tracking. The graph plotted for Bhattacharyya coefficient versus frame index exhibits significant increase in its values when the frames are pre-processed in comparison with the basic mean shift algorithm.

关键词： mean-shift algorithm illumination hue reserving

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：