Description: 人脸检测源代码. The souce demonstrates face detection SSE optimized C++ library for color and gray scale data with skin detection, motion estimation for faster processing, small sized SVM and NN rough face prefiltering, PCA/LDA/ICA/any dimensionality reduction/projection and final NN classification Platform: |
Size: 366639 |
Author:huangtyu |
Hits:
Description: 人脸检测源代码. The souce demonstrates face detection SSE optimized C++ library for color and gray scale data with skin detection, motion estimation for faster processing, small sized SVM and NN rough face prefiltering, PCA/LDA/ICA/any dimensionality reduction/projection and final NN classification -Face detection source code. The souce demonstrates face detection SSE optimized C++ Library for color and gray scale data with skin detection, motion estimation for faster processing, small sized SVM and NN rough face prefiltering, PCA/LDA/ICA/any dimensionality reduction/projection and final NN classification Platform: |
Size: 366592 |
Author:huangtyu |
Hits:
Description: 跟踪相关论文，《基于彩色图像的机器人视觉跟踪》《基于扩展卡尔曼滤波的主动视觉跟踪技术》《基于视觉和扩展卡尔曼滤波的位姿和运动估计新方法》《一种运动目标检测与跟踪快速算法的研究》-Tracking the relevant papers, based on the color image of the robot visual tracking , Extended Kalman Filter based on active vision tracking technology, Based on the visual and extended Kalman filter and the position and orientation of new motion estimation method , a moving target detection and fast algorithm for tracking of Study Platform: |
Size: 1433600 |
Author:王云 |
Hits:
Description: 这是运动补偿编码的运动估计检测，非常的常用，而且很实用-This is the motion compensation coding motion estimation detection, is very common, and very practical Platform: |
Size: 2048 |
Author:xp |
Hits:
Description: 自适应核密度估计运动检测方法
提出一种自适应的核密度(kernel density estimation, KDE)估计运动检测算法. 算法首先提出一种自适应前景、背景阈值的双阈值选择方法, 用于像素分类. 该方法用双阈值能克服用单阈值分类存在的不足, 阈值的选择能自适应进行, 且能适应不同的场景. 在此基础上, 本文提出了基于概率的背景更新模型, 按照像素的概率来更新背景, 并利用帧间差分背景模型和KDE分类结果, 来解决背景更新中的死锁问题, 同时检测背景的突然变化. 实验证明了所提出方法的适应性和可靠性.-Adaptive kernel density estimation method of motion detection of an adaptive kernel density (kernel density estimation, KDE) estimated motion detection algorithm. Algorithm first proposed an adaptive prospects, the background of the dual-threshold threshold selection method for pixel classification. The method used to overcome the dual-threshold single classification threshold shortcomings, the choice of threshold can be adaptive, and able to adapt to different scenes. On this basis, this paper, based on the probability of the background update model, in accordance with the probability of pixels to update the background, and take advantage of interframe differential background model and the classification results of KDE, to resolve the deadlock in the background update, while detection of abrupt changes in the background. Experiments show the proposed method of adaptability and reliability. Platform: |
Size: 1480704 |
Author:maolei |
Hits:
Description: Chesnokov Yuriy写的人脸检测库，内容涵盖肤色检测、运动估计、SVM分类、PCA/LDA/ICA特征提取以及神经网络分类器等。-The article demonstrates face detection SSE optimized C++ library for color and gray scale data with skin detection, motion estimation for faster processing, small sized SVM and NN rough face prefiltering, PCA/LDA/ICA/any dimensionality reduction/projection and final NN classification Platform: |
Size: 463872 |
Author:刘子木 |
Hits:
Description: This method is used for tracking wavelet/optical flow-based
detection for automatic target recognition in the following paper:
Dessauer, M. and Dua S. “Wavelet-based optical flow object detection, motion estimation, and tracking on moving vehicles” Platform: |
Size: 1355776 |
Author:MemoSergey |
Hits:
Description: 《DragonEye Fast Object Tracking and
Camera Motion Estimation》DragonEye是一个基于目标直接浏览的视频播放系统。比Dimp和Dragon来说具有实时性-Recently, multiple research groups proposed the concept of direct manipulation,
which makes the task of navigation in videos directly related to content of the
video. However, the developed systems are still too slow for a broad application of
the technique. To increase the accessibility of direct manipulation a system has to
fulfill certain requirements:
• Computation times have to be so small that the system can work in situation
with time constraints.
• The system has to accommodate for the user’s greater comprehension of a
scene. This includes camera motion, fore- and background, and occlusions.
Based on the requirements, we developed and implemented the DRAGONEYE system.
The tracker employs point tracking, with points provided by SIFT [Lowe,
2004], and color tracking with the CAMShift algorithm [Comaniciu et al., 2003]. A
model of the tracked object is built and continuously updated to fit the current conditions.
This model allows the detection and recover Platform: |
Size: 38917120 |
Author:童癸 |
Hits:
Description: 本篇論文提出的整合式解交錯(Integrated De-interlacing)的演算法，可以有效提昇移
動區域的畫面，但是當移動估計不正確時，反而會使移動補償後的畫面變得很差，為了
改善這種情況，因此結合移動可適性解交錯的優點，並將空間圖場內插(Spatial
Interpolation)的方式改成ELA(Edge Line Average)來設計，經過電腦模擬的結果發現，不僅在視覺上提高畫面的解析度，在某些影像峰值訊號雜訊比(Peak Signal Noise Ratio ,
PSNR)也比線平均解交 錯(Line Average De-interlacing)多出好幾分貝的畫質增益。
此外，在整合式解交錯演算法中也增加影片偵測(Film Detection)和影像加強(Image
Enhancement)的演算法設計，在這樣演算法的架構下，透過影片偵測的演算法，我們可
真實地還原3:2 Pull Down 的影片格式，而不會有鋸齒狀(Saw-Toothed)的畫面出現，而影
像加強的演算法，則可以在解交錯後，經過影像的調整，使輸出畫面呈現不同的效果，
達到消費者的需求。-The main theme of this thesis is an integrated de-interlacing system, which incorporates
several known and improved techniques in a nice manner to produce good de-interlaced image
quality. We first develop an accurate motion detector that classifies image regions into
stationary, low-motion, and high-motion categories. The simple field merging method is
applied to the stationary regions. The edge line average interpolation method is applied to the
slow-motion regions. Finally, the motion-compensated interpolation is applied to the
high-motion regions. In addition, hierarchical motion estimation and motion vector smoothing
techniques are employed to enhance the quality of estimated motion vectors. Our computer
simulation shows that the subjective image quality is improved by using the proposed scheme.
Also, its PSNR measures are better than the conventional spatial or temporal interpolation
schemes. Platform: |
Size: 1171456 |
Author:robin |
Hits:
Description: 摘要
车辆牌照识别(License Plate Recognition System, LPR)作为目标自动识别的一
种重要形式，可用于电子收费、出入控制、车流监控等众多场合，从而提高交通
管理自动化的程度，它的相关技术的研究正逐渐受到人们的重视。
本文主要介绍基于Run Length原理和Tamura纹理的车牌定位系统，该系统是基
于视频流进行开发的，主要包括车辆运动区域检测、车牌图像处理、车牌定位。
其中车辆运动区域检测利用多帧求平均的背景估计方法实现；车牌图像处理包括
灰度变换、中值滤波、边缘检测和二值化；车牌定位包括车牌水平定位、车牌垂
直定位、颜色判断和细定位。
该定位系统是在C++Builder环境下实现的，针对不同大小车牌、不同光照条件
的1400张图片进行测试，其定位率可以达到92 以上；同时也将该算法封装成DLL
库，应用在雷达测试仪中，对各种光照条件下的视频流进行测试。测试结果证明
本文算法简单、快速、准确率高，满足车牌识别系统实时性、通用性、和准确性
的要求，具有一定的应用价值。
关键词：车牌定位Run Length Tamura纹理-Abstract
As an important application form of target automatic recognition, vehicle license
plate recognition( LPR) technology is used in some fields, such as electron charging,
pass controlling and automobile stream supervising etc.
The paper introduces license plate location system based on the principle of Run
Length and Tamura texture feature. The system is developed based on video streams,
mainly includes vehicle motion region detection, license plate image processing, vehicle
license plate location. The detection of vehicle motion region uses the background
estimation methods based on the average of multi-frame license plate image processing
includes gray-scale transformation, median filtering, edge detection and binarization
the license plate location includes license plate horizontal location, license plate vertical
position, color judgments and accurate location.
The system is achieved in the C++ Builder environment, and it uses 1400 images
that are of different Platform: |
Size: 2509824 |
Author:王述 |
Hits:
Description: 自适应核密度估计运动检测方法 提出一种自适应的核密度(kernel density estimation, KDE)估计运动检测算法. 算法首先提出一种自适应前景、背景阈值的双阈值选择方法, 用于像素素分类. 该方法用双阈值能克服用单阈值分类存在的不足, 阈值的选择能自适应进行, 且能适应不同的场景. 在此基础上, 本文提出了基于概率的背景更新模型, 按照像素的概率来更新背景, 并利用帧间差分背景模
-Adaptive kernel density estimation motion detection method, an adaptive kernel density (kernel density estimation, KDE) estimated that the motion detection algorithm, the algorithm first raised the prospect of an adaptive dual-threshold of the background threshold selection method for the classification of the pixels prime The method can overcome the dual-threshold with the shortcomings of a single threshold classification, the choice of threshold can be adaptive, and can adapt to different scenarios. On this basis, this paper presents a probability-based background update model, according to the probability of the pixel to update the background, and frame difference and background mode Platform: |
Size: 1480704 |
Author:弯曲 |
Hits:
Description: 基于图像帧间差分检测的图像运动估计，时间检测图像帧间差分，判断目标对象的位置。-Time detecting an image interframe difference, and determine the position of the target object based on the image interframe difference detecting image motion estimation. Platform: |
Size: 624640 |
Author:chenghuachao |
Hits:
Description: 针对目前大多数形状特征描述的全局性以及对旋转 缩放等变化的敏感性， 采用一种基于目标近似多边形的形状特征描述， 这种描述方式具有局部性和紧凑性， 同时结合运动参数预测及递归估计的方法实现二维目标的检测和定位-Most shape for the global characterization of and sensitivity to changes in the rotation scaling, using a polygon shape based on characteristics of the target approximate description, this description has localized and compact manner, combined with motion parameters and recursive prediction estimation method to achieve the goal of a two-dimensional detection and localization... Platform: |
Size: 432128 |
Author:xuxin |
Hits:
Description: 基于特征的视觉里程计系统主要由特征检测与跟踪模块以及位姿计算模块两部分组成.为分析车载视觉里程计系统中引入车辆运动学约束的位姿计算算法性能,根据摄像机成像及视觉几何学原理,采用Matlab结合车辆动力学仿真软件CarSim建立车载视觉里程计仿真平台.该仿真平台由车辆运动仿真模块、成像仿真模块、数据显示与分析模块组成,仿真平台的测试对象为视觉里程计的位姿估计算法模块.该仿真平台充分考虑车载视觉定位系统的运动特性,为研究车辆运动学约束在视觉里程计系统中的应用提供新的思路和工具.对提出的一种全新的基于车辆运动学约束的位姿估计内层算法,在此仿真平台上进行性能验证.仿真结果表明,该算法在计算精度与效率上都能够满足实时车载视觉定位的要求.-Mainly by the feature detection and tracking module and pose calculation module composed of two parts based on visual odometry system features. In order to analyze the visual odometry system onboard the vehicle kinematic constraints introduced pose calculation algorithm performance, according to the camera image and visual geometry principle, combined with the use of vehicle dynamics simulation software Matlab CarSim establish visual odometry vehicle simulation platform. The simulation platform by the vehicle motion simulation module, imaging simulation module, the data shows the composition and analysis module, the test object simulation platform for visual odometry bit Pose estimation module. The simulation platform fully into account the motion characteristics of automotive vision positioning system, vehicle kinematic constraints apply to study in visual odometry system to provide new ideas and tools for a new vehicle based on the kinematics proposed inner constraints pose estimatio Platform: |
Size: 1504256 |
Author:zzz |
Hits:
Description: 针对成像平台运动情况下的运动目标检测问题，提出了一种从特征点稀疏运动场估计到运动分类的目标检测算法。
首先通过快速特征点检测与跟踪恢复出图像稀疏运动场；然后依据特征点之间运动一致性关系实现属于同一运动模式的特征
点分类，根据分类得到的各组特征点计算场景图像重建误差，剔除重建误差最小的特征点组，实现对前景目标的检测。仿真实
验对该算法在复杂场景中检测运动目标的有效性进行了验证。-】In order to detect target in the background motion video，an algorithm with 8parse motion field estimation，motion classification is prorMl．
Firstly，8pa_rl*motion field is recovered by fast local feature detection and吮king．The local features that belong to the same motion pattern are classified
according to their motion consistency．Then，the resulting local feature groups a_re used to reconstructed scene image，and the foreground targets aI_e identifled
by getting rid of the group with the least reconstruction en Dr．The proposed method is tested on the dataset of real complex scenarios and its effectiveness
is demonstrated in the results． Platform: |
Size: 1252352 |
Author:哈哈 |
Hits:
Description: 复杂环境下基于人群密度估计和人群运动检测的异常事件检测。-Complex environment based on population and population density estimation motion detection of abnormal events detected. Platform: |
Size: 4348928 |
Author:耿月 |
Hits:
Description: 详细讲解了25个MATLAB图像与视频处理实用案例（含可运行程序），涉及雾霾去噪、答题卡自动阅卷、肺部图像分割、小波数字水印、图像检索、人脸二维码识别、车牌定位及识别、霍夫曼图像压缩、手写数字识别、英文字符文本识别、眼前节组织提取、全景图像拼接、小波图像融合、基于语音识别的音频信号模拟灯控、路面裂缝检测识别、视频运动估计追踪、Simulink图像处理等多项重要技术(Explain in detail the 25 MATLAB image and video processing utility case (including operating procedures), involving haze denoising, sheet automatic marking, lung image segmentation, wavelet digital watermarking, image retrieval, face recognition, license plate location and two-dimensional code recognition, image compression, Hoffman handwritten digit recognition, character recognition, text English anterior segment tissue extraction, panoramic image mosaic, image fusion, light control, analog audio signal of speech recognition pavement crack detection and identification, video tracking, motion estimation of Simulink image processing technology based on a number of important) Platform: |
Size: 1324032 |
Author:糖糖的小糖
|
Hits:
Search - motion detection and estimation