技术领域technical field
本发明涉及视频处理、图像处理等视频智能分析技术领域，特别是涉及一种基于自适应阈值的智能乘客流量自动统计方法。 The invention relates to the technical field of video intelligent analysis such as video processing and image processing, in particular to an automatic statistical method for intelligent passenger flow based on an adaptive threshold. the
背景技术Background technique
人流量自动统计是智能视频监测系统中一个重要的功能，它能有效地应用于商场、公交车、地铁出入口等公共场所。智能公交是未来公共交通发展的必然模式，对公交客车乘客流量的自动统计对实现公交的智能化有着重要的意义。 Automatic counting of people flow is an important function in the intelligent video monitoring system, which can be effectively applied to public places such as shopping malls, buses, subway entrances and exits. Intelligent public transportation is an inevitable mode for the development of public transportation in the future. The automatic statistics of the passenger flow of public transportation is of great significance to realize the intelligentization of public transportation. the
目前已有的基于计算机视觉技术的人流量统计方法主要包括以下几类： At present, the existing methods of people flow statistics based on computer vision technology mainly include the following categories:
(1)基于人头或头肩检测和跟踪的方法。该方法在于有效地检测视频中的人头或者头肩，并对其进行跟踪以达到统计人流量的目的。比如：申请号为201210208666.7，发明名称为“一种基于视频分析技术的人流量统计方法”，申请人为武汉烽火众智数字技术有限责任公司的专利申请，根据对人头特征和人体局部特征的检测和识别进行人头特征区域检测，并采用跟踪技术得到行人运动轨迹以判定行人的方向和流量；申请号为PCT/CN2010/070607，发明名称为“人流量统计的方法及系统”，申请人为杭州海康威视软件有限公司的专利申请，则采用并联的多类人头分类器检测各个人头，并分别对各个人头进行跟踪形成人头目标运动轨迹，最后根据该运动轨迹方向进行人流量计数；申请号为201210316862，发明名称为“基于智能视觉感知的电梯人流量统计方法及系统”，申请人为电子科技大学的专利申请，根据提前建立的头肩模型库检测实时图像中的目标，并进行跟踪以达到人流量统计的目的。该类方法需要提取人头或头肩的有效特征或者进行大量正反样本的训练以产生有效的分类器，以实现准确的人头或头肩的检测，但其容易产生较高的虚警率，而且需要跟踪技术获取目 标运动轨迹，这大大增加了算法的运算量。 (1) Methods based on human head or head and shoulders detection and tracking. The method is to effectively detect people's heads or head and shoulders in a video, and track them to achieve the purpose of counting people's flow. For example: the application number is 201210208666.7, and the title of the invention is "a method for counting people flow based on video analysis technology". The applicant is a patent application of Wuhan Fenghuo Zhongzhi Digital Technology Co., Ltd. Recognize and detect the characteristic area of the head, and use the tracking technology to obtain the pedestrian trajectory to determine the direction and flow of pedestrians; the application number is PCT/CN2010/070607, the invention name is "Method and System for People Flow Statistics", and the applicant is Hangzhou Haikang The patent application of Nuctech Software Co., Ltd. uses a parallel multi-type head classifier to detect each head, and tracks each head separately to form a head target movement trajectory, and finally counts the flow of people according to the direction of the movement trajectory; the application number is 201210316862 , the name of the invention is "Elevator People Flow Statistics Method and System Based on Intelligent Visual Perception". The applicant is a patent application of the University of Electronic Science and Technology of China. According to the head and shoulders model library established in advance, the target in the real-time image is detected and tracked to achieve the flow of people. Statistical Purposes. This type of method needs to extract effective features of the head or head and shoulders or conduct a large number of positive and negative sample training to generate an effective classifier to achieve accurate detection of the head or head and shoulders, but it is prone to high false alarm rates, and Tracking technology is required to obtain the target trajectory, which greatly increases the computational load of the algorithm. the
(2)基于人体分割的方法。该类方法对视频序列进行人体检测，需要人体的先验知识，比如人体形状，边缘信息等，以统计行人流量。比如：申请号为201110423349，发明名称为“一种基于多高斯计数模型的人流量统计方法”，申请人为重庆邮电大学的专利申请，利用带人数标记的训练视频序列图像样本建立多高斯计数模型，然后基于该模型分析未知运动目标区域中包含的行人个数，从而实现人流量统计；申请号为201110147358，发明名称为“基于启发信息的行人流量统计方法”，申请人为杭州电子科技大学的专利申请，则采用基于梯度方向直方图的方法进行行人检测，并通过若干检测结果与特定区域上的点的比值关系产生权重，最后采用稀疏光流法确定运动矢量的大小和方向，以达到行人流量统计的目的。该类方法不仅难以有效解决遮挡问题，而且对人体特征的检测运算量较大，难以实现实时检测。 (2) Methods based on human body segmentation. This type of method performs human detection on video sequences, which requires prior knowledge of the human body, such as human body shape, edge information, etc., to count pedestrian traffic. For example: the application number is 201110423349, and the title of the invention is "a method for counting people flow based on multi-Gaussian counting model". Then based on this model, analyze the number of pedestrians contained in the unknown moving target area, so as to realize the statistics of pedestrian flow; the application number is 201110147358, and the title of the invention is "Pedestrian Flow Statistics Method Based on Heuristic Information", and the applicant is a patent application of Hangzhou Dianzi University , the method based on the gradient direction histogram is used for pedestrian detection, and the weight is generated by the ratio relationship between several detection results and points on a specific area. Finally, the sparse optical flow method is used to determine the size and direction of the motion vector to achieve pedestrian flow statistics. the goal of. This type of method is not only difficult to effectively solve the occlusion problem, but also requires a large amount of calculation for the detection of human body features, making it difficult to achieve real-time detection. the
发明内容Contents of the invention
本发明的目的在于克服现有技术存在的缺点和不足，提供一种快速而又高效的基于自适应阈值的智能乘客流量自动统计方法，该方法采用非监督学习的方法产生自适应阈值来进行人数判定，以及采用基于几何学原理的方法进行乘客上下车的判定，并结合视频处理技术，达到准确、实时地对智能公交乘客流量进行统计，最后再将乘客流量统计结果通过有线或无线通讯设备实时反馈到公交调度中心和未到达的公交站台电子显示终端上，及时为管理者和乘客提供实时的公交车载客信息。 The purpose of the present invention is to overcome the shortcomings and deficiencies of the prior art, and provide a fast and efficient automatic statistical method for intelligent passenger flow based on an adaptive threshold. Judgment, and use the method based on geometric principles to judge passengers getting on and off, combined with video processing technology, to achieve accurate and real-time statistics of intelligent bus passenger flow, and finally pass the passenger flow statistical results through wired or wireless communication equipment in real time Feedback to the bus dispatching center and the electronic display terminal of the bus stop that has not arrived, and provide real-time bus passenger information for managers and passengers in time. the
为了实现本发明的上述目的，本发明提供了一种基于自适应阈值的智能乘客流量自动统计方法，该方法包括以下步骤： In order to achieve the above-mentioned purpose of the present invention, the present invention provides a kind of intelligent passenger flow automatic statistics method based on self-adaptive threshold, and this method comprises the following steps:
步骤1，通过监控设备实时采集一检测区域的视频序列，并对采集得到的视频序列提取前景运动目标； Step 1, collect a video sequence of a detection area in real time through the monitoring equipment, and extract the foreground moving target from the video sequence obtained;
步骤2，根据实际应用场景和摄像头到检测区域的距离，设置乘客流量检测框的位置和大小； Step 2, according to the actual application scenario and the distance from the camera to the detection area, set the position and size of the passenger flow detection frame;
步骤3，基于视频信息和所述乘客流量检测框，通过非监督学习产生自适应阈值； Step 3, based on the video information and the passenger flow detection frame, an adaptive threshold is generated through unsupervised learning;
步骤4，采用基于几何学原理的方法对于所述乘客流量检测框进行乘客上车或下车行为的判定； Step 4, using a method based on geometric principles to determine the behavior of passengers getting on or off the car for the passenger flow detection frame;
步骤5，根据所述步骤4的判定结果和所述自适应阈值，对于客流量进行判定与统计。 Step 5, according to the determination result of the step 4 and the self-adaptive threshold, determine and count the passenger flow. the
本发明的有益效果在于： The beneficial effects of the present invention are:
(1)通过设置乘客流量检测框，可以缩小检测的范围，提高算法的效率； (1) By setting the passenger flow detection frame, the detection range can be narrowed and the efficiency of the algorithm can be improved;
(2)仅需获取检测框中的前景像素个数作为处理对象，而不需要进行复杂的识别等操作，大大降低了算法的运算量，提高算法的实时性； (2) It is only necessary to obtain the number of foreground pixels in the detection frame as the processing object, without the need for complex identification and other operations, which greatly reduces the amount of calculation of the algorithm and improves the real-time performance of the algorithm;
(3)对检测到的检测框中的前景像素个数进行非监督学习，进而产生自适应阈值，用来区分一次经过检测框的人数，避免使用跟踪算法解决遮挡问题，进一步降低计算量； (3) Perform unsupervised learning on the number of foreground pixels in the detected detection frame, and then generate an adaptive threshold, which is used to distinguish the number of people passing through the detection frame at one time, avoiding the use of tracking algorithms to solve the occlusion problem, and further reducing the amount of calculation;
(4)仅采用对检测框中的前景像素个数和检测框之间的几何学原理进行分析，即可产生对乘客上车或下车事件的有效判定方法，提高了算法的速度和效率； (4) Only by analyzing the number of foreground pixels in the detection frame and the geometric principle between the detection frames, an effective judgment method for passengers getting on or off the car can be produced, which improves the speed and efficiency of the algorithm;
(5)通过无线传感器将流量统计结果发送到站台电子显示终端，增加了系统的实用性。 (5) The traffic statistics result is sent to the platform electronic display terminal through the wireless sensor, which increases the practicability of the system. the
本发明简单易行，性能稳定，速度快，效率高，可移植性强，具有较强的实时性，适用于智能公交客车的乘客流量统计。 The invention is simple and easy to operate, stable in performance, fast in speed, high in efficiency, strong in portability and strong in real-time, and is suitable for passenger flow statistics of intelligent public buses. the
附图说明Description of drawings
图1是本发明基于自适应阈值的智能乘客流量自动统计方法的流程图； Fig. 1 is the flow chart of the present invention's intelligent passenger flow automatic statistics method based on self-adaptive threshold;
图2根据本发明一实施例的乘客流量检测框设置示意图； Fig. 2 is set schematic diagram according to the passenger flow detection frame of an embodiment of the present invention;
图3根据本发明一实施例的乘客上车和下车判定的示意图； Fig. 3 is a schematic diagram of passengers boarding and getting off according to an embodiment of the present invention;
图4根据本发明中基于自适应阈值的智能乘客流量自动统计方法实验仿真图。 Fig. 4 is an experimental simulation diagram of the intelligent passenger flow automatic statistics method based on the adaptive threshold in the present invention. the
具体实施方式Detailed ways
为使本发明的目的、技术方案和优点更加清楚明白，以下结合具体实施例，并参照附图，对本发明进一步详细说明。 In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings. the
本发明提出一种基于自适应阈值的智能乘客流量自动统计方法，该方法可用于公共交通工具等乘客频繁出入的场所，为了方便起见，接下来以公交客车为例对于本发明进行进一步的说明。 The present invention proposes an automatic statistical method for intelligent passenger flow based on an adaptive threshold, which can be used in public transport and other places where passengers frequently come and go. For convenience, the present invention will be further described by taking a bus as an example. the
图1为本发明基于自适应阈值的智能乘客流量自动统计方法的流程图。如图1所示，所述方法包括如下步骤： Fig. 1 is a flow chart of the intelligent passenger flow automatic statistics method based on the self-adaptive threshold of the present invention. As shown in Figure 1, the method includes the following steps:
步骤1，通过监控设备实时采集一检测区域的视频序列，并对采集得到的视频序列提取前景运动目标； Step 1, collect a video sequence of a detection area in real time through the monitoring equipment, and extract the foreground moving target from the video sequence obtained;
在本发明一实施例中，利用背景差分法结合形态学处理对采集得到的视频序列提取前景运动目标，其中，背景差分法和形态学处理方法均为本领域技术人员熟知的图像处理技术，在此不作赘述。 In one embodiment of the present invention, the foreground moving target is extracted from the collected video sequence by using the background subtraction method combined with morphological processing, wherein both the background subtraction method and the morphological processing method are image processing techniques well known to those skilled in the art. This will not be repeated. the
在实际应用中，可选取公交客车的上车区域，比如前门区域作为乘客流量统计的位置，监控摄像头可静止安装在公交客车内部前门区域的上方，并倾斜向下(不必完全垂直向下)来提取客流信息，视频采集速度可设为25帧/秒。 In practical applications, the boarding area of the bus can be selected, such as the front door area, as the location for passenger flow statistics, and the monitoring camera can be installed statically above the front door area inside the bus, and tilted downward (not necessarily completely vertically downward). To extract passenger flow information, the video acquisition speed can be set to 25 frames per second. the
步骤2，根据实际应用场景和摄像头到检测区域的距离，设置乘客流量检测框的位置和大小，其中，所述乘客流量检测框的位置设定在乘客进出的必经区域且所述乘客流量检测框的长度大于乘客进出区域的宽度； Step 2, according to the actual application scene and the distance from the camera to the detection area, set the position and size of the passenger flow detection frame, wherein the position of the passenger flow detection frame is set in the necessary area for passengers to enter and exit and the passenger flow detection The length of the box is greater than the width of the passenger access area;
图2为根据本发明一实施例的乘客流量检测框设置示意图，在本发明一实施例中，在确保乘客流量统计效率的同时，尽可能的缩小检测范围，以提高检测速度，因此，乘客流量检测框的长和宽在设置时需要满足以下约束条件： Fig. 2 is a schematic diagram of the arrangement of the passenger flow detection frame according to an embodiment of the present invention. In an embodiment of the present invention, while ensuring the statistical efficiency of the passenger flow, the detection range is reduced as much as possible to improve the detection speed. Therefore, the passenger flow The length and width of the detection frame need to meet the following constraints when setting:
i.检测框的宽度W满足：2Wh≤W≤3Wh； i. The width W of the detection frame satisfies: 2Wh ≤ W ≤ 3Wh ;
ii.检测框的长度L满足：L≥BC。 ii. The length L of the detection frame satisfies: L≥BC. the
其中，Wh为视频帧图像中圆型人头直径的估值，BC为检测区域的宽度，图2中，阴影部分的矩形ABCD表示公交客车前门上车区域，BC为其宽度，位于中部的方框即为所设置的乘客流量检测框。 Among them, Wh is the estimate of the diameter of the round human head in the video frame image, BC is the width of the detection area, in Figure 2, the rectangle ABCD in the shaded part represents the boarding area at the front door of the bus, BC is its width, and the square in the middle The frame is the set passenger flow detection frame.
步骤3，基于视频信息和所述乘客流量检测框，通过非监督学习产生自适应阈值； Step 3, based on the video information and the passenger flow detection frame, an adaptive threshold is generated through unsupervised learning;
所述步骤3进一步包括以下步骤： Said step 3 further comprises the following steps:
步骤31，统计前N帧图像中有乘客进入所述检测框时产生的非零前景像素值，得到m(m<N)个前景像素值； Step 31, count the non-zero foreground pixel values generated when a passenger enters the detection frame in the previous N frames of images, and obtain m (m<N) foreground pixel values;
步骤32，采用K-means等聚类算法将前景像素值分为K类，取K类聚类中心值的均值作为所述自适应阈值。 Step 32, using a clustering algorithm such as K-means to divide the foreground pixel values into K categories, and taking the mean value of the cluster center values of the K categories as the adaptive threshold. the
对于公交客车，由于每次从前门上车的乘客最多为两个乘客，故K-means等聚类算法中可将聚类数量K取为2。随着N的增大，不同乘客以不同方式上车的情况越来越全面，数据容量m也越来越大，由聚类算法得到的阈值也就越来越适应乘客人数的判定。当乘客进入检测框中产生的非零前景像素值大于该阈值时，则判定此次上车的乘客人数为2人，否则为1人。 For a bus, since there are at most two passengers who get on the bus from the front door each time, the number of clusters K can be set to 2 in clustering algorithms such as K-means. With the increase of N, the situation of different passengers getting on the bus in different ways becomes more and more comprehensive, and the data capacity m becomes larger and larger, and the threshold value obtained by the clustering algorithm is more and more suitable for the judgment of the number of passengers. When the non-zero foreground pixel value generated by the passenger entering the detection frame is greater than the threshold, it is determined that the number of passengers on the bus is 2, otherwise it is 1. the
步骤4，采用基于几何学原理的方法对于所述乘客流量检测框进行乘客上车或下车行为的判定； Step 4, using a method based on geometric principles to determine the behavior of passengers getting on or off the car for the passenger flow detection frame;
考虑到实际公交客车的运营情况，在拥挤的情况下，客车前门也可能会存在乘客下车的行为，所以为了精确统计公交客车乘客流量情况，需要对乘客上下车行为均进行判定。 Considering the actual operating conditions of the bus, passengers may also get off at the front door of the bus in a crowded situation. Therefore, in order to accurately count the passenger flow of the bus, it is necessary to determine the behavior of passengers getting on and off the bus. the
所述步骤4进一步包括以下步骤： Said step 4 further comprises the following steps:
步骤41，将所述乘客流量检测框分成多个子区域； Step 41, dividing the passenger flow detection frame into multiple sub-regions;
图3为根据本发明一实施例的乘客上车和下车判定示意图，如图3所示，在本发明一实施例中，可使用线段EH和线段FG将矩形ABCD分成三个小的矩形，其中，矩形ABCD为之前设置的乘客流量检测框。 Fig. 3 is a schematic diagram of passenger boarding and alighting determination according to an embodiment of the present invention. As shown in Fig. 3, in an embodiment of the present invention, the rectangle ABCD can be divided into three small rectangles by using the line segment EH and the line segment FG, Among them, the rectangle ABCD is the passenger flow detection frame set before. the
步骤42，计算所述子区域中乘客产生的前景目标像素面积； Step 42, calculating the foreground target pixel area generated by passengers in the sub-region;
假设图3中椭圆代表由乘客产生前景目标，则图3中黑色部分分别表示为前景目标进入到矩形AEHD和矩形FBCG产生的前景像素面积，分别命名为S1、S2或者S3、S4，如图3A和图3B所示。 Assuming that the ellipse in Figure 3 represents the foreground object generated by the passenger, the black parts in Figure 3 represent the foreground pixel area generated by the foreground object entering the rectangle AEHD and the rectangle FBCG respectively, which are named S1, S2 or S3, S4, respectively, as shown in Figure 3A and shown in Figure 3B. the
步骤43，根据同一子区域中乘客产生的前景目标像素面积的变化来判断所述乘客流量检测框中乘客的上车或下车行为。 Step 43 , judging the boarding or disembarking behavior of the passengers in the passenger flow detection frame according to the change of the foreground target pixel area generated by the passengers in the same sub-region. the
在本发明一实施例中，可根据位于所述乘客流量检测框上端部和下端部的两个子区域中，乘客产生的前景目标像素面积的变化来判断乘客的上车或下车行为，即，假设乘客通过乘客流量检测框的过程是由图3A所示的状态进入图3B所示的状态时，则通过下式来判断乘客的上车或下车行为： In an embodiment of the present invention, the passenger's boarding or disembarking behavior can be judged according to the change in the foreground target pixel area generated by the passenger in the two sub-regions located at the upper end and the lower end of the passenger flow detection frame, that is, Assuming that the process of passengers passing through the passenger flow detection frame is from the state shown in Figure 3A to the state shown in Figure 3B, the passenger's boarding or getting off behavior is judged by the following formula:
步骤5，根据所述步骤4的判定结果和所述自适应阈值，对于客流量进行判定与统计。 Step 5, according to the determination result of the step 4 and the self-adaptive threshold, determine and count the passenger flow. the
所述步骤5进一步包括以下步骤： Said step 5 further comprises the following steps:
步骤51，将参数的取值初始化为0，所述参数至少包括：参数flag＝false，参数up＝true，乘客流量检测框中前景的最小像素值low_pixel，乘客流量检测框中前景的最大像素值high_pixel，视频当前帧检测框中的前景像素值current_Fgpixel和视频前一帧检测框中的前景像素值last_pixel； Step 51, the value of the parameter is initialized to 0, and the parameter at least includes: parameter flag=false, parameter up=true, minimum pixel value low_pixel of the foreground in the passenger flow detection frame, maximum pixel value of the foreground in the passenger flow detection frame high_pixel, the foreground pixel value current_Fgpixel in the detection frame of the current frame of the video and the foreground pixel value last_pixel in the detection frame of the previous frame of the video;
步骤52，判断当前图像帧是否为视频序列的第一帧，若是，则令low_pixel＝high_pixel＝current_Fgpixel；否则，如果current_Fgpixel-last_Fgpixel>＝0，则令up＝true；如果current_Fgpixel-low_pixel>Thresh，flag＝false，up＝true三个条件均满足，其中，Thresh表示一经验阈值，一般取high_pixel的十分之一即可，此时判断为乘客进入乘客流量检测框，令flag＝true，同时计算步骤4中前景目标进入乘客流量检测框上部子区域和下部子区域中的前景像素面积，比如前景像素面积S1和S2；当flag＝true时，时刻更新high_pixel的值，使其取high_pixel、current_Fgpixel二者中的最大值；如果high_pixel-current_Fgpixel>Thresh，flag＝true两个条件同时满足，此时判断为乘客走出乘客流量检测框，则令flag＝false，up＝false，同时计算步骤4中前景目标进入乘客流量检测框上部子区域和下部子区域中的前景像素面积，比如前景像素面积S3和S4，然后根据得到的前景像素面积S1、S2、S3和S4并结合步骤4中的乘客上下车事件的判定，得出该前景经过乘客流量检测框时是属于上车事件还是下车事件； Step 52, judge whether the current image frame is the first frame of the video sequence, if so, then make low_pixel=high_pixel=current_Fgpixel; otherwise, if current_Fgpixel-last_Fgpixel>=0, then make up=true; if current_Fgpixel-low_pixel>Thresh, flag ＝false, up＝true three conditions are all satisfied, among them, Thresh represents an experience threshold, generally take one-tenth of high_pixel, at this time it is judged that the passenger has entered the passenger flow detection frame, set flag=true, and calculate the steps at the same time In 4, the foreground object enters the foreground pixel area in the upper sub-area and the lower sub-area of the passenger flow detection frame, such as the foreground pixel area S1 and S2; when flag=true, the value of high_pixel is updated at all times to make it take both high_pixel and current_Fgpixel The maximum value in; if high_pixel-current_Fgpixel>Thresh, flag=true two conditions are satisfied at the same time, at this time it is judged that the passenger has walked out of the passenger flow detection frame, then set flag=false, up=false, and calculate the foreground target in step 4 to enter The foreground pixel area in the upper sub-region and the lower sub-region of the passenger flow detection frame, such as the foreground pixel areas S3 and S4, and then according to the obtained foreground pixel areas S1, S2, S3 and S4 combined with the passenger getting on and off events in step 4 Determine whether the prospect is a boarding event or a disembarking event when passing through the passenger flow detection frame;
步骤53，比较high_pixel与所述步骤3得到的自适应阈值，如果high_pixel值较大，则对上述判定出的上车事件或者下车事件的流量计数值加2；否则，其流量计数值加1，如此即统计得到客流量。 Step 53, comparing high_pixel with the adaptive threshold obtained in step 3, if the high_pixel value is larger, then add 2 to the traffic count value of the above-mentioned determined boarding event or getting off event; otherwise, add 1 to the traffic count value , so that the passenger flow is obtained by statistics. the
需要说明的是，在本发明实施例中，步骤4中对于乘客上下车事件的定义，可根据实际情况准确定义一个方向为上车事件另一个方向为下车事件。 It should be noted that, in the embodiment of the present invention, for the definition of passenger getting on and off events in step 4, one direction can be accurately defined as the boarding event and the other direction as the getting off event according to the actual situation. the
图4为本发明基于自适应阈值的智能乘客流量自动统计方法的应用实例图，图4中的左图为检测框的设置示意图，如黑色方框所示，并定义上行是上车事件，下行是下车事件，图4中的中图为检测框前景检测结果图，图4中的右图为行人统计的结果图。 Fig. 4 is the application example diagram of the intelligent passenger flow automatic counting method based on adaptive threshold value of the present invention, and the left figure among Fig. 4 is the setting schematic diagram of detection frame, as shown in black box, and definition uplink is the event of getting on the bus, downlink is the getting off event. The middle picture in Figure 4 is the result of the foreground detection of the detection frame, and the right picture in Figure 4 is the result of pedestrian statistics. the
根据多次仿真实验结果得出，本发明提出的行人流量自动统计方法能够自动准确地统计乘客流量。 According to the results of multiple simulation experiments, the automatic counting method of pedestrian flow proposed by the present invention can automatically and accurately count passenger flow. the
以上所述的具体实施例，对本发明的目的、技术方案和有益效果进行了进一步详细说明，所应理解的是，以上所述仅为本发明的具体实施例而已，并不用于限制本发明，凡在本发明的精神和原则之内，所做的任何修改、等同替换、改进等，均应包含在本发明的保护范围之内。 The specific embodiments described above have further described the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above descriptions are only specific embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention. the
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410363764.7ACN104239908A (en) | 2014-07-28 | 2014-07-28 | Intelligent ridership automatic statistical method based on self-adaptive threshold value |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410363764.7ACN104239908A (en) | 2014-07-28 | 2014-07-28 | Intelligent ridership automatic statistical method based on self-adaptive threshold value |
| Publication Number | Publication Date |
|---|---|
| CN104239908Atrue CN104239908A (en) | 2014-12-24 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410363764.7APendingCN104239908A (en) | 2014-07-28 | 2014-07-28 | Intelligent ridership automatic statistical method based on self-adaptive threshold value |
| Country | Link |
|---|---|
| CN (1) | CN104239908A (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105551266A (en)* | 2015-12-08 | 2016-05-04 | 合肥寰景信息技术有限公司 | Method of calculating pedestrian flow threshold of traffic signal controller |
| CN110781748A (en)* | 2019-09-24 | 2020-02-11 | 上海商米科技集团股份有限公司 | IPC-based image processing method and camera |
| CN112333431A (en)* | 2020-10-30 | 2021-02-05 | 深圳市商汤科技有限公司 | Scene monitoring method and device, electronic equipment and storage medium |
| CN113822223A (en)* | 2021-10-12 | 2021-12-21 | 精英数智科技股份有限公司 | Method and device for detecting shielding movement of camera |
| CN113870604A (en)* | 2021-09-29 | 2021-12-31 | 湖南省交通规划勘察设计院有限公司 | Method and system for rational allocation and coordination of traffic hub passenger flow based on mobile phone signaling |
| CN116563287A (en)* | 2023-07-10 | 2023-08-08 | 长沙海信智能系统研究院有限公司 | Passenger flow volume detection method of bus and electronic equipment |
| WO2023159371A1 (en)* | 2022-02-23 | 2023-08-31 | 京东方科技集团股份有限公司 | Traffic statistical method and apparatus |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060227862A1 (en)* | 2005-04-06 | 2006-10-12 | March Networks Corporation | Method and system for counting moving objects in a digital video stream |
| CN101231755A (en)* | 2007-01-25 | 2008-07-30 | 上海遥薇实业有限公司 | Moving target tracking and quantity statistics method |
| CN101383005A (en)* | 2007-09-06 | 2009-03-11 | 上海遥薇实业有限公司 | Method for separating passenger target image and background by auxiliary regular veins |
| US7787656B2 (en)* | 2007-03-01 | 2010-08-31 | Huper Laboratories Co., Ltd. | Method for counting people passing through a gate |
| CN102622578A (en)* | 2012-02-06 | 2012-08-01 | 中山大学 | Passenger counting system and passenger counting method |
| US8295545B2 (en)* | 2008-11-17 | 2012-10-23 | International Business Machines Corporation | System and method for model based people counting |
| CN103021059A (en)* | 2012-12-12 | 2013-04-03 | 天津大学 | Video-monitoring-based public transport passenger flow counting method |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060227862A1 (en)* | 2005-04-06 | 2006-10-12 | March Networks Corporation | Method and system for counting moving objects in a digital video stream |
| CN101231755A (en)* | 2007-01-25 | 2008-07-30 | 上海遥薇实业有限公司 | Moving target tracking and quantity statistics method |
| US7787656B2 (en)* | 2007-03-01 | 2010-08-31 | Huper Laboratories Co., Ltd. | Method for counting people passing through a gate |
| CN101383005A (en)* | 2007-09-06 | 2009-03-11 | 上海遥薇实业有限公司 | Method for separating passenger target image and background by auxiliary regular veins |
| US8295545B2 (en)* | 2008-11-17 | 2012-10-23 | International Business Machines Corporation | System and method for model based people counting |
| CN102622578A (en)* | 2012-02-06 | 2012-08-01 | 中山大学 | Passenger counting system and passenger counting method |
| CN103021059A (en)* | 2012-12-12 | 2013-04-03 | 天津大学 | Video-monitoring-based public transport passenger flow counting method |
| Title |
|---|
| 郭荣庆等: "公交车人流量检测系统设计", 《长安大学学报（自然科学版）》* |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105551266A (en)* | 2015-12-08 | 2016-05-04 | 合肥寰景信息技术有限公司 | Method of calculating pedestrian flow threshold of traffic signal controller |
| CN110781748A (en)* | 2019-09-24 | 2020-02-11 | 上海商米科技集团股份有限公司 | IPC-based image processing method and camera |
| CN112333431A (en)* | 2020-10-30 | 2021-02-05 | 深圳市商汤科技有限公司 | Scene monitoring method and device, electronic equipment and storage medium |
| CN113870604A (en)* | 2021-09-29 | 2021-12-31 | 湖南省交通规划勘察设计院有限公司 | Method and system for rational allocation and coordination of traffic hub passenger flow based on mobile phone signaling |
| CN113822223A (en)* | 2021-10-12 | 2021-12-21 | 精英数智科技股份有限公司 | Method and device for detecting shielding movement of camera |
| WO2023159371A1 (en)* | 2022-02-23 | 2023-08-31 | 京东方科技集团股份有限公司 | Traffic statistical method and apparatus |
| CN116563287A (en)* | 2023-07-10 | 2023-08-08 | 长沙海信智能系统研究院有限公司 | Passenger flow volume detection method of bus and electronic equipment |
| Publication | Publication Date | Title |
|---|---|---|
| CN105844234B (en) | A method and device for people counting based on head and shoulders detection | |
| CN113011367B (en) | An Abnormal Behavior Analysis Method Based on Target Trajectory | |
| CN104239908A (en) | Intelligent ridership automatic statistical method based on self-adaptive threshold value | |
| Xia et al. | Towards improving quality of video-based vehicle counting method for traffic flow estimation | |
| CN103077423B (en) | To run condition detection method based on crowd's quantity survey of video flowing, local crowd massing situation and crowd | |
| WO2015131734A1 (en) | Method, device, and storage medium for pedestrian counting in forward looking surveillance scenario | |
| CN110472467A (en) | The detection method for transport hub critical object based on YOLO v3 | |
| CN103986910A (en) | A method and system for counting passenger flow based on intelligent analysis camera | |
| WO2019179024A1 (en) | Method for intelligent monitoring of airport runway, application server and computer storage medium | |
| CN105844229B (en) | Calculation method and system for passenger congestion degree | |
| CN106203513B (en) | Statistical method based on pedestrian head and shoulder multi-target detection and tracking | |
| CN103310444B (en) | A kind of method of the monitoring people counting based on overhead camera head | |
| CN108052859A (en) | A kind of anomaly detection method, system and device based on cluster Optical-flow Feature | |
| JP2007128513A (en) | Scene analysis | |
| WO2017156772A1 (en) | Method of computing passenger crowdedness and system applying same | |
| CN104680557A (en) | Intelligent detection method for abnormal behavior in video sequence image | |
| CN111369596B (en) | Escalator passenger flow volume statistical method based on video monitoring | |
| CN102156983A (en) | Pattern recognition and target tracking based method for detecting abnormal pedestrian positions | |
| Zheng et al. | Cross-line pedestrian counting based on spatially-consistent two-stage local crowd density estimation and accumulation | |
| CN101321269A (en) | Method and system for passenger flow detection based on computer vision | |
| CN105631418A (en) | People counting method and device | |
| CN103646253A (en) | Bus passenger flow statistics method based on multi-motion passenger behavior analysis | |
| CN102063613A (en) | People counting method and device based on head recognition | |
| CN114612860A (en) | Identification and prediction method of passenger flow in rail transit station based on computer vision | |
| CN107038411A (en) | A kind of Roadside Parking behavior precise recognition method based on vehicle movement track in video |
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | Application publication date:20141224 | |
| WD01 | Invention patent application deemed withdrawn after publication |