Pedestrian Detection Algorithm Research Articles

Precise Segmentation and Estimation of Pedestrian Trajectory Using On-Board Monocular Cameras

This paper presents a general algorithm for pedestrian detection by on-board monocular camera which can be applied to cameras of various view ranges. Under the assumption that motion of background can be nearly approximated as a linear function, the Spatio-Temporal MRF (S-T MRF) model segments foreground objects. The foreground objects contain both pedestrian and non-pedestrian urban objects, verification was conducted by a cascaded classifier. However, the segmentation results based on motion were not exactly fit into pedestrian on the image so that shrunk or inflated pedestrian were generated. This causes errors on extracting pedestrian trajectory. For precise positioning, we implemented two types of feedback algorithm for ROI correction using the Kalman filter and the voting result of Motion-classifier and HOG-classifier. We confirmed that those ROI Corrections successfully extract precise area of pedestrian and decrease the false negative rate. Elaborately extracted pedestrian trajectory could be used as a useful source for predicting collision to pedestrian.

Fast and Robust Pedestrian Detection Algorithm with Multi-Scale Orientation Features

PDF HTML阅读 XML下载 导出引用 引用提醒 基于多尺度方向特征的快速鲁棒人体检测算法 DOI: 10.3724/SP.J.1001.2011.03987 作者: 作者单位: 作者简介: 通讯作者: 中图分类号: 基金项目: 国家自然科学基金(61039003, 60872143); 国家重点基础研究发展计划(973)(2011CB706900, 2010CB731800) Fast and Robust Pedestrian Detection Algorithm with Multi-Scale Orientation Features Author: Affiliation: Fund Project: 摘要 | 图/表 | 访问统计 | 参考文献 | 相似文献 | 引证文献 | 资源附件 | 文章评论 摘要:提出一种多尺度方向(multi-scale orientation,简称MSO)特征描述子用于静态图片中的人体目标检测.MSO 特征由随机采样的图像方块组成,包含了粗特征集合与精特征集合.其中,粗特征是图像块的方向,而精特征由Gabor 小波幅值响应竞争获得.对于两种特征,分别采用贪心算法进行选择,并使用级联Adaboost 算法及SVM 训练检测模型.基于粗特征的Adaboost 分类器能够保证高的检测速度,而基于精特征的SVM 分类器则保证了检测精度.另外,通过MSO 特征块的平移,使得所提算法能够检 Abstract:The multi-scale orientation (MSO) features for pedestrian detection in still images are put forwarded in this paper. Extracted on randomly sampled square image blocks (units), MSO features are made up of coarse and fine features, which are calculated with a unit gradient and the Gabor wavelet magnitudes respectively. Greedy methods are employed respectively to select the features. Furthermore, the selected features are inputted into a cascade classifier with Adaboost and SVM for classification. In addition, the spatial location of MSO units can be shifted, are used to the handle multi-view problem and assembled; therefore, the occluded features are completed with average features of training positives, given an occlusion model, which enable the proposed approach to work in crowd scenes. Experimental results on INRIA testset and SDL multi-view testset report the state-of-arts results on INRIA include it is 12.4 times the faster than SVM+HOG method. 参考文献 相似文献 引证文献

Viewpoint-Independent Object Detection Based on Two-Dimensional Contours and Three-Dimensional Sizes

We propose a viewpoint-independent object-detection algorithm that detects objects in videos based on their 2-D and 3-D information. Object-specific quasi-3-D templates are proposed and applied to match objects' 2-D contours and to calculate their 3-D sizes. A quasi-3-D template is the contour and the 3-D bounding cube of an object viewed from a certain panning and tilting angle. Pedestrian templates amounting to 2660 and 1995 vehicle templates encompassing 19 tilting and 35 panning angles are used in this study. To detect objects, we first match the 2-D contours of object candidates with known objects' contours, and some object templates with large 2-D contour-matching scores are identified. In this step, we exploit some prior knowledge on the viewpoint on which the object is viewed to speed up the template matching, and the viewpoint likelihood for each contour-matched template is also assigned. Then, we calculate the 3-D widths, heights, and lengths of the contour-matched candidates, as well as the corresponding 3-D-size-matching scores. The overall matching score is obtained by combining the aforementioned likelihood and scores. The major contributions of this paper are to explore the joint use of 2-D and 3-D features in object detection. It shows that, by considering 2-D contours and 3-D sizes, one can achieve promising object detection rates. The proposed algorithms were evaluated on both pedestrian and vehicle sequences. It yielded significantly better detection results than the best results reported in PETS 2009, showing that our algorithm outperformed the state-of-the-art pedestrian-detection algorithms.

Selective integration of local‐feature detector by boosting for pedestrian detection

AbstractAn example‐based classification algorithm for pedestrian detection is presented. The classifier integrates component‐based classifiers according to the AdaBoost algorithm. A probability estimate by a kernel‐SVM is used for the outputs of base learners, which are independently trained for local features. The base learners are determined by selecting the optimal local feature according to sample weights determined by the boosting algorithm with cross‐validation. Our method was applied to the MIT CBCL pedestrian image database, and 54 subregions were extracted from each image as local features. The experimental results showed a good classification ratio for unlearned samples. © 2011 Wiley Periodicals, Inc. Electr Eng Jpn, 177(4): 12–22, 2011; Published online in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/eej.21195

On-board Pedestrian Detection by the Motion and the Cascaded Classifiers

This paper presents a general algorithm for pedestrian detection by on-board monocular camera, which can be applied to cameras of various viewing angle. The foreground objects which have discriminative motion-differences compared to that of background buildings are extracted as Regions of Interest (ROIs). Then, those ROIs are tracked by Spatio-Temporal MRF(S-T MRF) model as a possible pedestrian. Before classification process, some ROIs are rejected quickly by geometric constraints, calculated from coordinates between position of ROI and heights in real-world. Finally, ROIs are verified by HOG/Fisher cascade. The geometrical pre-processing saves the overall computational costs as well as decrease false positive rate. Also, algorithm compensates the detection rate due to temporal classification error in tracking cues, which improves the stability of tracking.

Pedestrian detection algorithms based on an analysis of the autocorrelation function of a seismic signal

A pedestrian detection algorithm based on an analysis of the autocorrelation function of a seismic signal is proposed. The efficiency of this method in terms of detection and false alarm probabilities is shown. Approaches to reducing the complexity of the proposed algorithm are considered. A comparative study is made of the effects of the point sampling frequency and the method of calculating the autocorrelation function on the performance and complexity of the pedestrian detection algorithm.

A shape-based voting algorithm for pedestrian detection and tracking

This paper presents the MOUGH (mixture of uniform and Gaussian Hough) Transform for shape-based object detection and tracking. We show that the edgels of a rigid object at a given orientation are approximately distributed according to a Gaussian mixture model (GMMs). A variant of the generalized Hough transform is proposed, voting using GMMs and optimized via Expectation-Maximization, that is capable of searching images for a mildly-deformable shape, based on a training dataset of (possibly noisy) images with only crude estimates of scale and centroid of the object in each image. Further modifications are proposed to optimize the algorithm for tracking. The method is able to locate and track objects reliably even against complex backgrounds such as dense moving foliage, and with a moving camera. Experimental results indicate that the algorithm is superior to previously published variants of the Hough transform and to active shape models in tracking pedestrians from a side view.

Pedestrian Detection and Tracking in an Urban Environment Using a Multilayer Laser Scanner

Pedestrians are the most vulnerable participants in urban traffic. The first step toward protecting pedestrians is to reliably detect them in a real-time framework. In this paper, a new approach is presented for pedestrian detection in urban traffic conditions using a multilayer laser sensor mounted onboard a vehicle. This sensor, which is placed on the front of a vehicle, collects information about the distance distributed according to four planes. Like a vehicle, a pedestrian constitutes, in the vehicle environment, an obstacle that must be detected, and located and then identified and tracked if necessary. To improve the robustness of pedestrian detection using a single laser sensor, a detection system based on the fusion of information located in the four laser planes is proposed. The method uses a nonparametric kernel-density-based estimation of pedestrian position of each laser plane. The resulting pedestrian estimations are then sent to a decentralized fusion according to the four planes. Temporal filtering of each object is finally achieved within a stochastic recursive Bayesian framework (particle filter), allowing a closer observation of pedestrian random movement dynamics. Many experimental results are given and validate the relevance of our pedestrian-detection algorithm with regard to a method using only a single-row laser-range scanner.

Scene Modeling and Statistical Learning Based Robust Pedestrian Detection Algorithm

A scene model and statistic learning based method for pedestrian detection in complicated real-world scenes is proposed. A unique characteristic of the algorithm is its ability to train a special cascade classifier dynamically for each individual scene. The benefit is that the classifier only focuses on the differences between the positive samples and the limited negative samples of each individual scene, thus greatly reduces the complexity of classification, and achieves robust detection result even with a few classifiers. A highly efficient weak classifier selection method and a novel boosting architecture are presented to speed up feature selection and classifier training. To evaluate the proposed algorithm, we captured pedestrian videos under different weathers, seasons and camera motions, and labeled 4 300 positive samples. Moreover, a real-time pedestrian detection system named as background modeling and Adaboost training (BMAT) was developed, which produced fast and robust detection results as demonstrated by extensive experiments performed using video sequences under different environments.

Selective Integration of Local-Feature Detector by Boosting for Pedestrian Detection

An example-based classification algorithm for pedestrian Detection is presented. The classifier integrates component-based classifiers according to the AdaBoost algorithm. A probability estimate by a kernel-SVM is used for the outputs of base learners, which are independently trained for local features. The base learners are determined by selecting the optimal local feature according to sample weights determined by the boosting algorithm with cross-validation. Our method was applied to the MIT CBCL pedestrian image database, and 54 sub-regions were extracted from each image as local features. The experimental results showed a good classification ratio for unlearned samples.

Pedestrian detection using a single-monochrome camera

A car-mounted single monochrome camera-based pedestrian detection algorithm is discussed. The detection range is divided into two sub-regions, that is, the near distance range and the far distance range. Two different detection algorithms are applied in the two regions. For the near distance range, where the direction of the motion of the detected obstacle is important, a motion segmentation approach using interest points is utilised. For the far distance range where the motion of the detected obstacle is not as important, a robust and computationally efficient modified inverse perspective mapping-based obstacle detection is utilised. Finally, a low-level pedestrian-oriented segmentation algorithm, which is aimed at the depth information of the detected pedestrian candidate, is also presented.

Development of Bicycle and Pedestrian Detection and Classification Algorithm for Active-Infrared Overhead Vehicle Imaging Sensors

Existing algorithms used with active-infrared overhead vehicle-imaging sensors consider vehicle size and speed attributes as basic parameters to detect and classify 11 categories of motorized vehicles. These algorithms could not detect and classify bicycles and pedestrians. This research focused on developing and evaluating algorithms for active-infrared overhead vehicle-imaging sensor technology to detect and classify non-motorized users. Development of the theory and algorithm used to automate the simultaneous detection and classification of bicycles and pedestrians along with the field investigations to evaluate its effectiveness are described. The new algorithm used the concept of message sequencing to incorporate existing active-infrared technology theory. Bicycles and pedestrians intersected the infrared scan patterns in different sequences that, along with the infrared images, provided unique detection and classification identification. The algorithm was integrated within the existing active-infrared technology, and a field evaluation was conducted on bicycle and pedestrian trails. The algorithm created an intelligent technology to detect and classify bicycles and pedestrians. Nearly 100% of bicycles and pedestrians were detected, and about 92% of them were successfully classified. Automated data collection technology can be useful in obtaining more comprehensive travel data and in forecasting demand for design and policy making related to nonmotorized transportation.

Development of Bicycle and Pedestrian Detection and Classification Algorithm for Active-Infrared Overhead Vehicle Imaging Sensors

Existing algorithms used with active-infrared overhead vehicle-imaging sensors consider vehicle size and speed attributes as basic parameters to detect and classify 11 categories of motorized vehicles. These algorithms could not detect and classify bicycles and pedestrians. This research focused on developing and evaluating algorithms for active-infrared overhead vehicle-imaging sensor technology to detect and classify nonmotorized users. Development of the theory and algorithm used to automate the simultaneous detection and classification of bicycles and pedestrians along with the field investigations to evaluate its effectiveness are described. The new algorithm used the concept of message sequencing to incorporate existing active-infrared technology theory. Bicycles and pedestrians intersected the infrared scan patterns in different sequences that, along with the infrared images, provided unique detection and classification identification. The algorithm was integrated within the existing active-infrared technology, and a field evaluation was conducted on bicycle and pedestrian trails. The algorithm created an intelligent technology to detect and classify bicycles and pedestrians. Nearly 100% of bicycles and pedestrians were detected, and about 92% of them were successfully classified. Automated data collection technology can be useful in obtaining more comprehensive travel data and in forecasting demand for design and policy making related to nonmotorized transportation.

Real-Time Vision-based Detection of Waiting Pedestrians

This paper presents a real-time digital signal processing system to detect the presence of pedestrians waiting to cross an intersection. A pedestrian detection algorithm is developed and implemented on a Texas Instruments TMS320C40 digital signal processor (DSP) chip such that videosignals captured and transmitted from an actual intersection are processed in real-time. The detection algorithm consists of a background update and background subtraction components applied to image blocks. The background update is done in order to cope with brightness changes that occur in actual outdoor scenes. In addition, waiting and moving pedestrians are distinguished by using the temporal changes of image block intensities. The results obtained indicate that this DSP implementation provides an acceptable detection rate for this traffic management problem.