Semi-supervised Elastic net for pedestrian counting

Pedestrian counting plays an important role in public safety and intelligent transportation. Most pedestrian counting algorithms based on supervised learning require much labeling work and rarely exploit the topological information of unlabelled data in a video. In this paper, we propose a Semi-Supervised Elastic Net (SSEN) regression method by utilizing sequential information between unlabelled samples and their temporally neighboring samples as a regularization term. Compared with a state-of-the-art algorithm, extensive experiments indicate that our algorithm can not only select sparse representative features from the original feature space without losing their interpretability, but also attain superior prediction performance with only very few labelled frames.     

基于计算机视觉的人流量双向统计

提出了一种采用视频监控系统对人行通道口进行双向人流量计数的方法。首先建立发色模型与头部形状模型,采用形态学运算提取人的头部目标,然后跟踪目标建立人头目标移动链,依据目标链位置信息判别行人的进出方向,最后设置感兴趣的检测区域,并对通过该检测区域的行人计数。实验结果表明,该方法能实时有效地统计通道口处双向人流量。     

Directional guidance from audible pedestrian signals for street crossing

Typical audible pedestrian signals indicate when the pedestrian walk interval is in effect but provide little, or even misleading information for directional alignment. In three experiments, blind and blindfolded sighted adults crossed a simulated crossing with recorded traffic noise to approximate street sounds. This was done to investigate how characteristics of signal presentation affected usefulness of the auditory signal for guiding crossing behaviour. Crossing was more accurate when signals came only from the far end of the crossing rather than the typical practice of presenting signals simultaneously from both ends. Alternating the signal between ends of the crossing was not helpful. Also, the customary practice of signalling two parallel crossings at the same time drew participants somewhat toward the opposite crossing. Providing a locator tone at the end of the crossing during the pedestrian clearance interval improved crossing accuracy. These findings provide a basis for designing audible pedestrian signals to enhance directional guidance. The principal findings were the same for blind and sighted participants and applied across a range of specific signals (e.g. chirps, clicks, voices).     

Directional guidance from audible pedestrian signals for street crossing

Typical audible pedestrian signals indicate when the pedestrian walk interval is in effect but provide little, or even misleading information for directional alignment. In three experiments, blind and blindfolded sighted adults crossed a simulated crossing with recorded traffic noise to approximate street sounds. This was done to investigate how characteristics of signal presentation affected usefulness of the auditory signal for guiding crossing behaviour. Crossing was more accurate when signals came only from the far end of the crossing rather than the typical practice of presenting signals simultaneously from both ends. Alternating the signal between ends of the crossing was not helpful. Also, the customary practice of signalling two parallel crossings at the same time drew participants somewhat toward the opposite crossing. Providing a locator tone at the end of the crossing during the pedestrian clearance interval improved crossing accuracy. These findings provide a basis for designing audible pedestrian signals to enhance directional guidance. The principal findings were the same for blind and sighted participants and applied across a range of specific signals (e.g. chirps, clicks, voices).     

Courbage神经元映射模型的随机共振研究

采用神经元二维映射模型,通过数字仿真研究了高斯白噪声对神经元非线性动力学特性的影响.研究发现,噪声可以诱导具有次阈值输入信号的神经元产生动作电位和随机共振.随机共振现象的产生与否和噪声强度的大小以及输入信号的频率具有密切的关系.另外,还研究了系统的控制参数对随机共振现象的影响.     

可满足问题中的模型计数

模型计数问题是指计算给定问题的解的个数,这是一类比决策更困难的问题,也是人工智能领域研究的一个热点问题.对模型计数问题的研究不仅可以提高算法的求解效率,更能促进对问题困难本质的了解.以可满足问题（命题可满足（SAT）和约束可满足问题（CSP））为例,从精确算法和近似求解两方面综述了模型计数问题的研究现状,重点介绍了相关概念以及各个算法之间的优缺点,并提出了有待解决的开放性问题,对模型计数问题的研究予以了总结和展望.     

基于光流法和行人面积特征的人数统计方法研究

为了提高视频监控领域中出入口的人数统计的准确率,提出构建一种近似 俯视运动行人面积作为人数统计的特征。使用垂直悬挂的单目摄像头对出入口进行拍摄,设 置一条虚拟计数线,利用Lucas-Kanade 光流算法计算计数线上像素的速度矢量,并构建近 似俯视运动行人面积作为特征对过线人数进行估计并计数。实验结果表明,使用该方法可以 降低光照原因产生的阴影以及图像预处理对人数统计的干扰,比较准确地对出入口进行人数 统计。     

行人运动建模技术综述*

随着计算机数值模拟技术的迅速发展,行人运动仿真技术开始成为人群疏散仿真、城市规划和计算机疏散软件开发方面的研究热点。介绍了对行人建模仿真的初步研究,重点介绍了基于行人行为的微观行人运动建模方法,并对各个模型进行了比较分析和研究展望。     

高斯白噪声对神经元映射模型动力学的影响*

研究了噪声引发的神经元相干共振现象。基于神经元二维映射模型,用高斯白噪声模拟生物神经系统中的环境噪声,通过数字仿真研究了噪声对神经元动力学特性的影响。研究发现,噪声可以提高静态神经元的可兴奋性,诱导其产生动作电位,当噪声达到某一适中强度时,动作电位的有序性可以达到最佳,产生相干共振现象。研究了系统参数的选取对神经元共振特性的影响,发现控制参数越接近分岔点,相干共振性越好。结果表明合适强度的噪声可以促进神经元的信息传递。     

Epidemiological modeling with a population density map-based cellular automata simulation system

We present a new numerical, two-dimensional cellular automata framework for simulation of the spread of an infectious disease in a region with non-homogenous spatial population distribution. For the simulation the real map of the population density in Poland is used, where the sources of the infection are located. Presented model is a combination of SEIR and IBM models complemented with additional factors like variable population density, death, birth and some stochastic parameters to reflect the more realistic population dynamics. In proposed model the states of individuals are tracked through time like in IBM model and the evolution of the whole system is described by SEIR transition function which determines how cells interact with their neighbours, influencing global behavior of the system. Presented model requires less complicated input than IBM models and is less expensive computationally. We explore influenza as the contagious disease in our map-based simulation. The results of the simulation show the spreading-rate of the disease and can be used to describe possible actions for preventing pandemic.     

Simulation of bi-directional pedestrian flow by using a cell transmission model

An extended cell transmission model is proposed to simulate bi-directional pedestrian flow in the corridor. In the model, the walking space is discretized into regular hexagonal cells. Three walking preferences of pedestrians are taken into account, including walking on the right-hand side, following front people in the same direction, and avoiding conflicts with ones in the opposite direction. An implementation of the model with periodic boundary condition is then presented. Furthermore, by simulation experiments, we show the effects of the model parameters on flow distributions and fundamental diagrams. The model is also calibrated through comparing the flow-density relationships from empirical data and model simulations. In addition, the model can successfully reproduce typical self-organization phenomena in bi-directional pedestrian flow, e.g., two-lane formation and multi-lane formation, although it is not a microscopic model.     

A single-shot model for traffic-related pedestrian detection

Pattern Analysis and Applications - Traffic-related pedestrian detection is important for advanced driving-assistant systems and autonomous driving. In addition to pedestrian detection,...     

右转车流及对向行人影响下的行人过街延误模型

通过分析车头时距随交通流量而变化的统计特性,提出了基于车头时距分布的自适应的行人过街冲突延误计算模型。根据低交通流量情况下车头时距服从负指数分布时的行人过街冲突延误,分别推导出中等交通流量及高交通流量情况下的行人过街冲突延误计算模型;结合无干扰条件下的相位延误和对向行人流阻滞延误得到干扰条件下的行人过街总延误计算模型;通过Vissim仿真实验对模型进行了准确性验证,其偏差小于3%。此外,将提出的行人过街延误模型与以机动车到达间距服从负指数分布为假设前提的行人过街延误模型进行误差比较,结果表明所提出的延误模型误差更小。     

Probabilistic symbolic model checking with PRISM: a hybrid approach

In this paper we present efficient symbolic techniques for probabilistic model checking. These have been implemented in PRISM, a tool for the analysis of probabilistic models such as discrete-time Markov chains, continuous-time Markov chains and Markov decision processes using specifications in the probabilistic temporal logics PCTL and CSL. Motivated by the success of model checkers such as SMV which use BDDs (binary decision diagrams), we have developed an implementation of PCTL and CSL model checking based on MTBDDs (multi-terminal BDDs) and BDDs. Existing work in this direction has been hindered by the generally poor performance of MTBDD-based numerical computation, which is often substantially slower than explicit methods using sparse matrices. The focus of this paper is a novel hybrid technique which combines aspects of symbolic and explicit approaches to overcome these performance problems. For typical examples, we achieve a dramatic improvement over the purely symbolic approach. In addition, thanks to the compact model representation using MTBDDs, we can verify systems an order of magnitude larger than with sparse matrices, while almost matching or even beating them for speed.     

An intelligence-based route choice model for pedestrian flow in a transportation station

This study proposes a method that uses an artificial neural network (ANN) to mimic human decision-making about route choice in a crowded transportation station. Although ANN models have been developed rapidly and widely adopted in various fields in the last three decades, their application to predict human decision-making in pedestrian flows is limited, because the video clip technology used to collect pedestrian movement data in crowded conditions is still primitive. Data collection must be carried out manually or semi-manually, which requires extensive resources and is time consuming. This study adopts a semi-manual approach to extract data from video clips to capture the route choice behaviour of travellers, and then applies an ANN to mimic such decision-making. A prediction accuracy of 86% (ANN model with ensemble approach) is achieved, which demonstrates the feasibility of applying the ANN approach to decision-making in pedestrian flows.     

改进R-FCN模型的小尺度行人检测

目的 为了有效解决传统行人检测算法在分辨率低、行人尺寸较小等情境下检测精度低的问题,将基于区域全卷积网络（region-based fully convolutional networks,R-FCN）的目标检测算法引入到行人检测中,提出一种改进R-FCN模型的小尺度行人检测算法。方法 为了使特征提取更加准确,在ResNet-101的conv5阶段中嵌入可变形卷积层,扩大特征图的感受野;为提高小尺寸行人检测精度,在ResNet-101中增加另一条检测路径,对不同尺寸大小的特征图进行感兴趣区域池化;为解决小尺寸行人检测中的误检问题,利用自举策略的非极大值抑制算法代替传统的非极大值抑制算法。结果 在基准数据集Caltech上进行评估,实验表明,改进的R-FCN算法与具有代表性的单阶段检测器（single shot multiBox detector,SSD）算法和两阶段检测器中的Faster R-CNN（region convolutional neural network）算法相比,检测精度分别提高了3.29%和2.78%;在相同ResNet-101基础网络下,检测精度比原始R-FCN算法提高了12.10%。结论 本文提出的改进R-FCN模型,使小尺寸行人检测精度更加准确。相比原始模型,改进的R-FCN模型对行人检测的精确率和召回率有更好的平衡能力,在保证精确率的同时,具有更大的召回率。     

基于注意力机制的行人轨迹预测生成模型

针对长短期记忆网络（LSTM）在行人轨迹预测问题中孤立考虑单个行人,且无法进行多种可能性预测的问题,提出基于注意力机制的行人轨迹预测生成模型（AttenGAN）,来对行人交互模式进行建模和概率性地对多种合理可能性进行预测。AttenGAN包括一个生成器和一个判别器,生成器根据行人过去的轨迹概率性地对未来进行多种可能性预测,判别器用来判断一个轨迹是真实的还是由生成器伪造生成的,进而促进生成器生成符合社会规范的预测轨迹。生成器由一个编码器和一个解码器组成,在每一个时刻,编码器的LSTM综合注意力机制给出的其他行人的状态,将当前行人个体的信息编码为隐含状态。预测时,首先用编码器LSTM的隐含状态和一个高斯噪声连接来对解码器LSTM的隐含状态初始化,解码器LSTM将其解码为对未来的轨迹预测。在ETH和UCY数据集上的实验结果表明,AttenGAN模型不仅能够给出符合社会规范的多种合理的轨迹预测,并且在预测精度上相比传统的线性模型（Linear）、LSTM模型、社会长短期记忆网络模型（S-LSTM）和社会对抗网络（S-GAN）模型有所提高,尤其在行人交互密集的场景下具有较高的精度性能。对生成器多次采样得到的预测轨迹的可视化结果表明,所提模型具有综合行人交互模式,对未来进行联合性、多种可能性预测的能力。     

视觉注意机制下结合语义特征的行人检测

目的 为研究多场景下的行人检测,提出一种视觉注意机制下基于语义特征的行人检测方法。方法 首先,在初级视觉特征基础上,结合行人肤色的语义特征,通过将自下而上的数据驱动型视觉注意与自上而下的任务驱动型视觉注意有机结合,建立空域静态视觉注意模型;然后,结合运动信息的语义特征,采用运动矢量熵值计算运动显著性,建立时域动态视觉注意模型;在此基础上,以特征权重融合的方式,构建时空域融合的视觉注意模型,由此得到视觉显著图,并通过视觉注意焦点的选择完成行人检测。结果 选用标准库和实拍视频,在Matlab R2012a平台上,进行实验验证。与其他视觉注意模型进行对比仿真,本文方法具有良好的行人检测效果,在实验视频上的行人检测正确率达93%。结论 本文方法在不同的场景下具有良好的鲁棒性能,能够用于提高现有视频监控系统的智能化性能。     

Effective pedestrian detection using deformable part model based on human model

Recently, pedestrian detection systems have become an important technology in the development of the advanced driver assistance system (ADAS) for the autonomous car. The histogram of oriented gradients (HOG) is currently the most basic algorithm for detecting pedestrians, but it treats the entire body of the pedestrian as one single feature. In other words, if the entire body of the pedestrian is not visible, the detection rate under HOG decreases markedly. To solve this problem, we propose a detection system using a deformable part model (DPM) that divides the pedestrian data into two parts using a latent support vector machine (SVM)-based machine-learning technique. Experimental results show that our approach achieves better performance in a detection system than the existing method. In practice, there are many occlusions in the environment in front of the vehicle. For example, the surrounding transport facilities, such as a car or another obstacle, can occlude a pedestrian. These occlusions can increase the false detection rate and cause difficulties during the detection process. Our proposed method uses a different approach and can easily be applied in real-world scenarios, regardless of occlusions.

     

Modified diffusion model with multiple products using a hybrid GA approach

As technology advances, the speed in which new products are developed also increases. Due to such increases, product forecasting has become much more vital for a company. The Bass diffusion model is a demand-forecast model that explores the phases of a product’s life cycle that have been successful in the diffusion of forecasting innovation in new products. Recognizing the need for an efficient parameter estimation method for multi-product forecasting, we have conducted research using the hybrid genetic algorithm (HGA). The research conducted will provide an alternate approach to explore the forecasting capability of the diffusion models without having as many limitations as the original method. We used both published data and LCD-monitor global sales data to test and verify our method. Results show that the proposed model using a hybrid GA approach can improve the forecasting efficiency.