WO2022011828A1 - 出入电梯物件检测系统及方法、物件检测系统、电梯光幕以及电梯设备 - Google Patents
出入电梯物件检测系统及方法、物件检测系统、电梯光幕以及电梯设备 Download PDFInfo
- Publication number
- WO2022011828A1 WO2022011828A1 PCT/CN2020/116371 CN2020116371W WO2022011828A1 WO 2022011828 A1 WO2022011828 A1 WO 2022011828A1 CN 2020116371 W CN2020116371 W CN 2020116371W WO 2022011828 A1 WO2022011828 A1 WO 2022011828A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- elevator
- image
- receiver
- outline
- module
- Prior art date
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 29
- 238000000034 method Methods 0.000 title claims abstract description 19
- 238000012549 training Methods 0.000 claims description 28
- 238000012545 processing Methods 0.000 claims description 14
- 238000013527 convolutional neural network Methods 0.000 claims description 11
- 238000013499 data model Methods 0.000 claims description 11
- 230000009471 action Effects 0.000 claims description 9
- 238000011176 pooling Methods 0.000 claims description 9
- 210000002569 neuron Anatomy 0.000 claims description 8
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 230000000717 retained effect Effects 0.000 claims description 5
- 238000007781 pre-processing Methods 0.000 claims description 3
- 230000000875 corresponding effect Effects 0.000 description 31
- 238000010586 diagram Methods 0.000 description 14
- 230000000903 blocking effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000004027 cell Anatomy 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000009877 rendering Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66B—ELEVATORS; ESCALATORS OR MOVING WALKWAYS
- B66B5/00—Applications of checking, fault-correcting, or safety devices in elevators
- B66B5/0006—Monitoring devices or performance analysers
- B66B5/0012—Devices monitoring the users of the elevator system
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66B—ELEVATORS; ESCALATORS OR MOVING WALKWAYS
- B66B5/00—Applications of checking, fault-correcting, or safety devices in elevators
- B66B5/02—Applications of checking, fault-correcting, or safety devices in elevators responsive to abnormal operating conditions
Definitions
- the invention belongs to the technical field of elevator equipment, and relates to elevator equipment, in particular to a system and method for detecting objects entering and leaving an elevator, and an elevator light curtain.
- Elevator is the most commonly used vertical transportation means in modern high-rise buildings. It saves people's time and physical strength and provides convenience for daily life. As a special equipment closely related to the life safety of the public, the safe operation of the elevator has attracted more and more attention from the society. However, due to the complex structure of the elevator, to ensure the safe and reliable operation of the elevator, and to detect its operating status and fault conditions have become an urgent need for elevator management, maintenance and safe operation.
- the safety light curtains use the oppositely arranged infrared transmitters and infrared receivers to send and receive signals to determine whether there are people or objects between the elevator doors.
- the existing elevator equipment recognizes objects entering and exiting the elevator, usually through manual viewing through the camera, and there is no solution that can automatically recognize the shape of the object, and the degree of intelligence of the elevator equipment needs to be further improved.
- the present invention provides a system and method for detecting objects entering and leaving an elevator, an object detection system, an elevator light curtain and elevator equipment, which can identify objects (including persons or/and objects) entering and exiting an elevator and improve the intelligence of the elevator equipment.
- a system for detecting objects entering and leaving an elevator comprising:
- the receiver signal acquisition module is used to acquire the signals received by each receiver of the elevator light curtain
- the outline drawing module is used to draw the outline image of the person or/and the object passing through the elevator light curtain according to the signal obtained by the signal acquisition module of the receiver; whether the sensor sensed by each receiver at the set time point in each time period is blocked or not signal, draw the outline image of the person or/and the object in the elevator light curtain at the set time point, and form the outline image of the person or/and the object passing through the elevator light curtain within the set time period;
- the object detection module is used for identifying and obtaining the attributes of the objects entering and leaving the elevator according to the outline of the object drawn by the outline drawing module.
- the system further includes: a data model establishment module, which is used to establish a data model related to objects and drawn images by using a convolutional neural network.
- the system further includes: a data training module for establishing a training set exceeding a set threshold, and the training set stores objects and corresponding drawn images.
- the data training module includes:
- the image width setting unit is used to obtain the original image of the input neural network, and set the specified width threshold range of the input image according to the width distribution of all original input images;
- the image processing unit is used to perform image processing on the original input image; for the original input image whose width is less than the minimum value of the specified width threshold range, the corresponding original input image is filled to meet the set width requirement; for the original input image If the width of the original input image is greater than the maximum value of the specified width threshold range, the corresponding original input image width is scaled to make the image width within the specified width threshold range.
- the data training module or/and the contour drawing module include:
- the repeated data merging unit is used to calculate the correlation coefficient of adjacent frames in the case of a large number of similar repeated data in the drawn image, if the correlation coefficients of adjacent frames exceeding the length of the first threshold B are all higher than the second threshold C , only the number of frames whose length is the first threshold B is retained.
- the outline drawing module includes:
- a frame image drawing unit used to draw the image of the corresponding frame according to the signal strength received by each receiver at each time point
- an outline drawing unit for sequentially splicing the frame images drawn by the frame image drawing unit in the order of time points to form the outline image of the corresponding object
- the frame image drawing unit is used to draw a frame of elevator light curtain data into an image that is highly correlated with the number of receivers; according to the signal strength received by each receiver, set the corresponding The gray value of the area, if the receiver is blocked, the gray value of the corresponding area is the first gray value; if the receiver is not blocked, the gray value of the corresponding area is the second gray value.
- the outline drawing module includes an image processing unit for performing image processing on the drawn image; for an image whose drawn image width is less than the minimum value of the specified width threshold range, the corresponding image is filled, Make it meet the set width requirement; for an image whose width is larger than the specified width threshold range, the corresponding image width will be scaled to make the image width within the specified width threshold range.
- the data model building module is used to build a shallow convolutional neural network
- the first layer is the input layer
- the input image size is 32*specified width*number of light curtain directions
- second The layer is a convolutional layer, the filter size is 3*3, and the number of filters is 32
- the third layer is a pooling layer, the filter size is 2*2, and the step size is 2
- the fourth layer is a convolutional layer, the filter size It is 3*3, and the number of filters is 64
- the fifth layer is a pooling layer, the filter size is 2*2, and the step size is 2
- the sixth layer is a fully connected layer, and the number of neurons is 128
- the seventh layer is a fully connected layer.
- the layer is the output layer, and the number of neurons is equal to the number of recognized object categories.
- the system further includes: a data preprocessing module for normalizing the images of the training set, and after one-hot encoding the labels, the labels are transmitted to the convolutional neural network for training.
- the system further includes: an early warning module for potential safety hazards, which is used to issue early warning information when it is detected that the set object enters the elevator.
- the contour drawing module is used to obtain the contours of objects that pass through the light curtain action time period completely;
- the complete light curtain action time period refers to: any one receiving unit is blocked from the beginning, until all receiving units are blocked. The cells are not blocked to end the corresponding time period.
- an object detection system the system includes:
- a plurality of receivers arranged on the second side of the setting area for receiving signals from the corresponding transmitters
- the contour drawing module is used to draw the contour images of people or/and objects passing through the set area according to the signals obtained by each receiver; draw the signals of whether the receivers are blocked or not sensed by the set time points in each time period.
- the object detection module is used for identifying and obtaining the attributes of the objects entering and leaving the set area according to the outline of the object drawn by the outline drawing module.
- an elevator equipment comprising the above-mentioned detection system for objects entering and leaving an elevator.
- a method for detecting objects entering and leaving an elevator comprising:
- the receiver signal acquisition step acquires the signals received by each receiver of the elevator light curtain;
- outline drawing step draw the outline image of the person or/and object passing through the elevator light curtain according to the signal acquired by the signal acquisition module of the receiver; set the time point in each time period to sense whether each receiver is blocked or not. , draw the outline image of the person or/and the object in the elevator light curtain at the set time point, and form the outline image of the person or/and the object passing through the elevator light curtain within the set time period;
- the object detection step according to the outline of the object drawn by the outline drawing module, the attributes of the objects entering and leaving the elevator are identified and obtained.
- the beneficial effects of the present invention are: the system and method for detecting objects entering and leaving the elevator, the object detection system, the elevator light curtain and the elevator equipment proposed by the present invention can identify the objects (including personnel or/and objects) entering and exiting the elevator, and improve the intelligence of the elevator equipment. . After identifying the objects entering and leaving the elevator, it is convenient for the elevator equipment to take further actions; for example, if an object that is not allowed enters the elevator, an alarm signal can be issued.
- FIG. 1 is a schematic diagram of the composition of a system for detecting objects entering and leaving an elevator in an embodiment of the present invention.
- FIG. 2 is a schematic diagram of the composition of a system for detecting objects entering and leaving an elevator in an embodiment of the present invention.
- FIG. 3 is a flowchart of a method for detecting objects entering and leaving an elevator in an embodiment of the present invention.
- FIG. 4-1 is a schematic diagram of an image of a person and a bicycle passing through a light curtain drawn in an embodiment of the present invention.
- 4-2 is a schematic diagram of an image of a person and a bicycle passing through a light curtain drawn in an embodiment of the present invention.
- FIG. 5-1 is a schematic diagram of an image of a person passing through a light curtain drawn in an embodiment of the present invention.
- FIG. 5-2 is a schematic diagram of an image of a person passing through a light curtain drawn in an embodiment of the present invention.
- 5-3 is a schematic diagram of an image of a person passing through a light curtain drawn in an embodiment of the present invention.
- 5-4 are schematic diagrams of images of people passing through a light curtain drawn in an embodiment of the present invention.
- FIG. 6-1 is a schematic diagram of an image of a person and an electric vehicle passing through a light curtain drawn in an embodiment of the present invention.
- FIG. 6-2 is a schematic diagram of an image of a person and an electric vehicle passing through a light curtain drawn in an embodiment of the present invention.
- FIG. 7-1 is a schematic diagram of images before and after optimizing repeated frames in an embodiment of the present invention.
- FIG. 7-2 is a schematic diagram of images before and after optimizing a repeated frame according to an embodiment of the present invention.
- connection in the specification includes both direct connection and indirect connection.
- object in the specification refers to a person or/and an object.
- FIG. 1 is a schematic diagram of the composition of a system for detecting objects in and out of an elevator in an embodiment of the present invention; please refer to Figure 1, the system includes: a receiver signal acquisition module 1, a contour drawing module 2 and object detection module 3.
- the receiver signal acquisition module 1 is used to acquire the signals received by each receiver of the elevator light curtain.
- the outline drawing module 2 is used to draw the outline image of the person or/and the object passing through the elevator light curtain according to the signal acquired by the signal acquisition module of the receiver; For the blocked signal, draw the outline image of the person or/and object in the elevator light curtain at the set time point, please refer to Figure 4-1, Figure 4-2, Figure 5-1, Figure 5-2, Figure 5- 3. As shown in Figure 5-4, Figure 6-1, and Figure 6-2; and thus form the outline image of the people or/and objects passing through the elevator light curtain within the set time period.
- the outline drawing module 2 is used to obtain the outline of each object passing through the light curtain action time period completely; the complete passing light curtain action time period means: any one receiving unit is blocked as a start, until all receiving units are blocked. Not being blocked is the end of the corresponding time period.
- the outline drawing module 2 includes: a frame image drawing unit and an outline drawing unit.
- the frame image drawing unit is used for drawing the image of the corresponding frame according to the signal strength received by each receiver at each time point.
- the outline drawing unit is used for sequentially splicing the frame images drawn by the frame image drawing unit in the order of time points to form the outline image of the corresponding object.
- described frame image drawing unit is in order to draw the elevator light curtain data of a frame into the image that is high and the number of receivers is associated with the image; According to the signal strength that each receiver receives, set the gray of the corresponding area. If the receiver is blocked, the gray value of the corresponding area is the first gray value; if the receiver is not blocked, the gray value of the corresponding area is the second gray value.
- the contour drawing module 2 may include a repeated data merging unit; the repeated data merging unit is used to calculate the correlation coefficient of adjacent frames when a large amount of similar repeated data appears in the drawn image, if it exceeds the first threshold value.
- the correlation coefficients of adjacent frames of length B are all higher than the second threshold C, then only the number of frames whose length is the first threshold B is retained. Refer to Figure 7-1 and Figure 7-2.
- the outline drawing module 2 may also include an image processing unit to perform image processing on the drawn image; for the image whose width after drawing is less than the minimum value of the specified width threshold range, the corresponding image is filled to make it meet the set requirements. If the width of the drawn image is larger than the maximum value of the specified width threshold range, the width of the corresponding image is scaled so that the width of the image is within the specified width threshold range.
- the object detection module 3 is used for identifying and obtaining the attributes of the objects entering and leaving the elevator according to the object outline drawn by the outline drawing module.
- the system further includes: a safety hazard warning module, which is used to issue a warning message when it is detected that the set object enters the elevator .
- FIG. 2 is a schematic diagram of the composition of a system for detecting objects in and out of an elevator in an embodiment of the present invention; please refer to FIG. 2 , in an embodiment of the present invention, the system further includes: a data model building module 4 for using convolutional neural The network builds data models about objects and rendering images.
- the data model building module 4 is used to build a shallow convolutional neural network, the first layer is the input layer, and the input image size is 32*specified width*light curtain direction direction number;
- the second layer is a convolutional layer, the filter size is 3*3, and the number of filters is 32;
- the third layer is a pooling layer, the filter size is 2*2, and the stride is 2;
- the fourth layer is a convolutional layer, filter The size is 3*3, and the number of filters is 64;
- the fifth layer is a pooling layer, the filter size is 2*2, and the step size is 2;
- the sixth layer is a fully connected layer, and the number of neurons is 128;
- the seventh layer is the output layer, and the number of neurons is equal to the number of recognized object categories.
- the system further includes: a data training module 5 for establishing a training set exceeding a set threshold, and the training set stores objects and corresponding drawn images.
- the data training module 5 includes: an image width setting unit and an image processing unit.
- the image width setting unit is used to obtain the original image input to the neural network, and set the specified width threshold range of the input image according to the width distribution of all original input images.
- the image processing unit is used to perform image processing on the original input image; for the original input image whose width is less than the minimum value of the specified width threshold range, the corresponding original input image is filled to meet the set width requirement; for the original input image If the width of the original input image is greater than the maximum value of the specified width threshold range, the width of the corresponding original input image is scaled to make the image width within the specified width threshold range.
- the data training module 5 further includes a repeated data merging unit; the repeated data merging unit is used to calculate the correlation coefficient of adjacent frames when a large amount of similar repeated data appears in the drawn image, If the correlation coefficients of adjacent frames exceeding the length of the first threshold B are all higher than the second threshold C, only the number of frames whose length is the first threshold B is retained.
- the system further includes: a data preprocessing module for normalizing the images of the training set, and after one-hot encoding the labels, the labels are transmitted to the convolutional neural network for training.
- the present invention discloses an object detection system.
- the object detection system of the present invention can be used not only in the field of elevators, but also in other fields.
- the object detection system comprises: several transmitters arranged on the first side of the setting area, several receivers arranged on the second side of the setting area, a contour drawing module and an object detection module.
- Each transmitter is used for transmitting the setting signal; each receiver is used for receiving the signal sent by the corresponding transmitter.
- the transmitter can transmit the signal horizontally, and the transmitted signal can also have a certain angle with the horizontal direction.
- one transmitter transmits signals to one receiver; in another embodiment, one transmitter may transmit signals to multiple receivers, and one receiver may receive signals transmitted by different transmitters.
- the contour drawing module is used to draw the contour image of the person or/and object passing through the set area according to the signal obtained by each receiver; according to the signal sensed by each receiver at the set time point in each time period, whether it is blocked or not, draw the device.
- a contour image of a person or/and an object between a set area at a fixed time point is used to form a contour image of a person or/and an object passing through the set area within a set time period.
- the object detection module is used for identifying and obtaining the attributes of the objects entering and leaving the set area according to the outline of the object drawn by the outline drawing module.
- the present invention discloses an elevator light curtain, which includes the above-mentioned detection system for objects entering and leaving the elevator.
- the present invention discloses an elevator equipment, including the above-mentioned detection system for objects entering and leaving the elevator.
- FIG. 3 is a flowchart of a method for detecting objects in and out of an elevator in an embodiment of the present invention; please refer to FIG. 3 , the method includes:
- Step S1 the receiver signal acquisition step; acquire the signal received by each receiver of the elevator light curtain;
- Step S2 outline drawing step; draw the outline image of the person or/and object passing through the elevator light curtain according to the signal acquired by the signal acquisition module of the receiver; The occluded signal, draw the outline image of the person or/and the object in the elevator light curtain at the set time point, and form the outline image of the person or/and the object passing through the elevator light curtain within the set time period;
- Step S3 an object detection step; according to the object outline drawn by the outline drawing module, the attributes of the objects entering and leaving the elevator are identified and obtained.
- the method of the present invention also includes:
- Data training step establish a training set that exceeds the set threshold, and the training set stores objects and corresponding drawn images.
- the training set stores objects and corresponding drawn images.
- the elevator light curtain is provided with 32 receivers.
- the method specifically includes:
- Step 1 When there is an object blocking the infrared transmitter and the corresponding receiver, the signal strength of the receiver will be significantly reduced.
- a frame of elevator light curtain data is drawn into a grayscale image with a height of 32 pixels and a width of 1 pixel. 32 pixels correspond to 32 receivers respectively. Judging according to the signal strength of the receivers, if the gray value of the corresponding pixel is blocked, set the If it is 0, if the gray value of the corresponding pixel is not blocked, the gray value is set to 255, and the gray image is drawn as the side view of the object between the light curtains at the corresponding time point.
- Step 2 Starting with any receiver being blocked, and ending when all receivers are not blocked, it can be considered as a complete movement of an object through the light curtain. Draw the side view of the occluder during this time, and label the category of the object as a data in the training set.
- Step 3 After establishing a sufficient number of training sets, it is necessary to perform certain analysis and processing on the data. Since the shape of the pictures input to the neural network should be consistent, observe the width distribution of all pictures, select a suitable width (with less impact on most data) as the width of the input picture, and the pictures whose original picture width is smaller than the specified width are randomly divided between two Pixels with a gray value of 0 with an indeterminate width are added to the side to fill the image width to 40. If the original image width is larger than the specified width, the width of the image needs to be scaled to make the image width equal to the specified width without changing the height of the image.
- X, Y are two adjacent frames of data
- Cov(X, Y) is the covariance of X and Y
- Var[X] is the variance of X
- Var[Y] is the variance of Y
- r(X, Y ) is the obtained correlation coefficient between X and Y.
- Step 4 Build a shallow convolutional neural network.
- the first layer is the input layer, and the input image size is 32*specified width*The number of light curtain directions; the second layer is the convolutional layer, the filter size is 3*3, and the filter size is 3*3.
- the number is 32; the third layer is a pooling layer, the filter size is 2*2, and the stride is 2; the fourth layer is a convolutional layer, the filter size is 3*3, and the number of filters is 64;
- the layer is a pooling layer with a filter size of 2*2 and a step size of 2;
- the fifth layer is a pooling layer with a filter size of 2*2 and a step size of 2;
- the sixth layer is a fully connected layer with the number of neurons is 128;
- the seventh layer is the output layer, and the number of neurons is equal to the number of recognized object categories.
- Step 5 Normalize the images of the training set, and pass the labels to the convolutional neural network for training after one-hot encoding.
- Step 6 Save the trained neural network model.
- an object passes through the light curtain, draw a side view of the obstacle covering the corresponding time period, adjust the size of the image and normalize it, and then load the neural network model to predict the type of the object.
- the category with the highest probability in the output of the neural network is the category of obstructions predicted by the neural network.
- the system and method for detecting objects entering and leaving the elevator can identify the objects (including persons or/and objects) entering and exiting the elevator, and improve the intelligence of the elevator equipment. After identifying the objects entering and leaving the elevator, it is convenient for the elevator equipment to take further actions; for example, if an object that is not allowed enters the elevator, an alarm signal can be issued.
Landscapes
- Image Analysis (AREA)
- Indicating And Signalling Devices For Elevators (AREA)
Abstract
一种出入电梯物件检测系统及方法、物件检测系统、电梯光幕以及电梯设备，出入电梯物件检测系统包括接收器信号获取模块（1）、轮廓绘制模块（2）及物件检测模块（3）；接收器信号获取模块（1）用以获取电梯各接收器接收的信号；轮廓绘制模块（2），用以根据所述接收器信号获取模块获取的信号绘制经过电梯光幕的人员或/和物体的轮廓图像；通过各时间段中设定时间点各接收器感应到的是否被遮挡的信号，绘制设定时间点处于光幕间的人员或/和物体的轮廓图像；物件检测模块（3）用以根据所述轮廓绘制模块绘制的物件轮廓，识别得到出入电梯的物件的属性。该系统及方法可识别出入电梯物件(包括人员或/和物体)，提高电梯设备的智能性。
Description
本发明属于电梯设备技术领域，涉及一种电梯设备，尤其涉及一种出入电梯物件检测系统及方法以及电梯光幕。
电梯是现代高层建筑中最常用的一种垂直运输交通工具，它节省了人们的时间和体力，为日常生活提供了方便。电梯作为一种与大众生命安全密切相关的特种设备，其安全运行越来越受到社会的关注。但由于电梯结构复杂，因此要保证电梯安全可靠的运行，检测其运行状态和故障情况成为电梯管理、维护和安全运行的迫切需要。
为提高电梯安全性，电梯均需要设置安全光幕，安全光幕利用相对设置的红外发送装置及红外接收装置收发信号，依此判断电梯门之间是否有人或物品。
同时，为了提高电梯的应急反应能力，电梯物联网的设想已经被逐步实现，物业、电梯运营公司、政府部门可以在远程实时监控电梯的状态，发现异常情况，可以及时获取相关信息。
此外，现有电梯设备对于进出电梯的物体识别，通常是通过摄像头人工查看，如今还没有能自动识别物体形状的方案，电梯设备的智能化程度还有待进一步改善。
有鉴于此，如今迫切需要设计一种新的电梯设备，以便克服现有电梯设备存在的上述至少部分缺陷。
发明内容
本发明提供一种出入电梯物件检测系统及方法、物件检测系统、电梯光幕以及电梯设备，可识别出入电梯物件(包括人员或/和物体)，提高电梯设备的智能性。
为解决上述技术问题，根据本发明的一个方面，采用如下技术方案：
一种出入电梯物件检测系统，所述系统包括：
接收器信号获取模块，用以获取电梯光幕各接收器接收的信号；
轮廓绘制模块，用以根据所述接收器信号获取模块获取的信号绘制经过电梯光幕的人员或/和物体的轮廓图像；通过各时间段中设定时间点各接收器感应到的是否被遮挡的信号，绘制设定时间点处于电梯光幕间的人员或/和物体的轮廓图像，并以此形成设定时间段内通过电 梯光幕间的人员或/和物体的轮廓图像；
物件检测模块，用以根据所述轮廓绘制模块绘制的物件轮廓，识别得到出入电梯的物件的属性。
作为本发明的一种实施方式，所述系统进一步包括：数据模型建立模块，用以利用卷积神经网络建立有关物件及绘制图像的数据模型。
作为本发明的一种实施方式，所述系统进一步包括：数据训练模块，用以建立一个超过设定阈值的训练集，训练集中存储有物件及对应的绘制图像。
作为本发明的一种实施方式，所述数据训练模块包括：
图像宽度设定单元，用以获取输入神经网络的原始图像，根据所有原始输入图像的宽度分布设定输入图像的指定宽度阈值范围；
图像处理单元，用以对原始输入图像进行图像处理；对于原始输入图像宽度小于指定宽度阈值范围最小值的原始输入图像，则填补对应原始输入图像，使其满足设定的宽度要求；对于原始输入图像宽度大于指定宽度阈值范围最大值的原始输入图像，则缩放对应原始输入图像的宽度，使图像宽度在指定宽度阈值范围内。
作为本发明的一种实施方式，所述数据训练模块或/和轮廓绘制模块包括：
重复数据合并单元，用以在绘制的图像出现大量类似的重复数据的情况下，计算相邻帧的相关系数，若超过第一阈值B长度的相邻帧的相关系数均高于第二阈值C，则只保留长度为第一阈值B的帧数。
作为本发明的一种实施方式，所述轮廓绘制模块包括：
帧图像绘制单元，用以根据各时间点各接收器接收的信号强度绘制对应帧的图像；
轮廓绘制单元，用以按时间点的次序将所述帧图像绘制单元绘制的帧图像依次拼接，形成对应物件的轮廓图像；
作为本发明的一种实施方式，所述帧图像绘制单元用以将一帧的电梯光幕数据绘制成高与接收器数量存在设定关联的图像；根据各接收器接收的信号强度设定对应区域的灰度值，若接收器被遮挡，则对应区域的灰度值为第一灰度值；若接收器没有被遮挡，则对应区域的灰度值为第二灰度值。
作为本发明的一种实施方式，所述轮廓绘制模块包括图像处理单元，用以对绘制后的图像进行图像处理；对于绘制后图像宽度小于指定宽度阈值范围最小值的图像，则填补对应图像，使其满足设定的宽度要求；对于绘制图像宽度大于指定宽度阈值范围最大值的图像，则缩放对应图像的宽度，使图像宽度在指定宽度阈值范围内。
作为本发明的一种实施方式，所述数据模型建立模块用以搭建一个浅层卷积神经网络，第一层为输入层，输入图像大小为32*指定宽*光幕方向方向数量；第二层为卷积层，filter大小为3*3，filter个数为32个；第三层为池化层，filter大小为2*2，步长为2；第四层为卷积层，filter大小为3*3，filter个数为64个；第五层为池化层，filter大小为2*2，步长为2；第六层为全连接层，神经元个数为128个；第七层为输出层，神经元个数等于识别物体类别数。
作为本发明的一种实施方式，所述系统进一步包括：数据预处理模块，用以将训练集图像归一化，并将标签进行one-hot编码后传入卷积神经网络进行训练。
作为本发明的一种实施方式，所述系统进一步包括：安全隐患预警模块，用以在检测出设定物件进入电梯时，发出预警信息。
作为本发明的一种实施方式，所述轮廓绘制模块用以获取各完整通过光幕动作时间段的物件轮廓；完整通过光幕动作时间段指：任意一个接收单元被遮挡为开始、至所有接收单元均不被遮挡为结束对应的时间段。
根据本发明的另一个方面，采用如下技术方案：一种物件检测系统，所述系统包括：
设置于设定区域第一侧的若干发射器，用以发射设定信号；
设置于设定区域第二侧的若干接收器，用以接收对应发射器发出的信号；
轮廓绘制模块，用以根据各接收器获取的信号绘制经过设定区域的人员或/和物体的轮廓图像；通过各时间段中设定时间点各接收器感应到的是否被遮挡的信号，绘制设定时间点处于设定区域间的人员或/和物体的轮廓图像，并以此形成设定时间段内通过设定区域的人员或/和物体的轮廓图像；以及
物件检测模块，用以根据所述轮廓绘制模块绘制的物件轮廓，识别得到出入设定区域的物件的属性。
根据本发明的又一个方面，采用如下技术方案：一种电梯设备，包括上述的出入电梯物件检测系统。
根据本发明的又一个方面，采用如下技术方案：一种出入电梯物件检测方法，所述方法包括：
接收器信号获取步骤；获取电梯光幕各接收器接收的信号；
轮廓绘制步骤；根据所述接收器信号获取模块获取的信号绘制经过电梯光幕的人员或/和物体的轮廓图像；通过各时间段中设定时间点各接收器感应到的是否被遮挡的信号，绘制设定时间点处于电梯光幕间的人员或/和物体的轮廓图像，并以此形成设定时间段内通过电梯光 幕间的人员或/和物体的轮廓图像；
物件检测步骤；根据所述轮廓绘制模块绘制的物件轮廓，识别得到出入电梯的物件的属性。
本发明的有益效果在于：本发明提出的出入电梯物件检测系统及方法、物件检测系统、电梯光幕以及电梯设备，可识别出入电梯物件(包括人员或/和物体)，提高电梯设备的智能性。在识别出出入电梯物件后，可以便于电梯设备做出进一步的动作；例如，如果有不被允许的物体进入电梯，则可以发出报警信号。
图1为本发明一实施例中出入电梯物件检测系统的组成示意图。
图2为本发明一实施例中出入电梯物件检测系统的组成示意图。
图3为本发明一实施例中出入电梯物件检测方法的流程图。
图4-1为本发明一实施例中绘制的人加自行车通过光幕的图像示意图。
图4-2为本发明一实施例中绘制的人加自行车通过光幕的图像示意图。
图5-1为本发明一实施例中绘制的人通过光幕的图像示意图。
图5-2为本发明一实施例中绘制的人通过光幕的图像示意图。
图5-3为本发明一实施例中绘制的人通过光幕的图像示意图。
图5-4为本发明一实施例中绘制的人通过光幕的图像示意图。
图6-1为本发明一实施例中绘制的人加电动车通过光幕的图像示意图。
图6-2为本发明一实施例中绘制的人加电动车通过光幕的图像示意图。
图7-1为本发明一实施例中优化重复帧前后的图像示意图。
图7-2为本发明一实施例中优化重复帧前后的图像示意图。
下面结合附图详细说明本发明的优选实施例。
为了进一步理解本发明，下面结合实施例对本发明优选实施方案进行描述，但是应当理解，这些描述只是为进一步说明本发明的特征和优点，而不是对本发明权利要求的限制。
该部分的描述只针对几个典型的实施例，本发明并不仅局限于实施例描述的范围。相同 或相近的现有技术手段与实施例中的一些技术特征进行相互替换也在本发明描述和保护的范围内。
说明书中的“连接”既包含直接连接，也包含间接连接。说明书中的“物件”，指人员或/和物体。
本发明揭示了一种出入电梯物件检测系统，图1为本发明一实施例中出入电梯物件检测系统的组成示意图；请参阅图1，所述系统包括：接收器信号获取模块1、轮廓绘制模块2及物件检测模块3。
所述接收器信号获取模块1用以获取电梯光幕各接收器接收的信号。
所述轮廓绘制模块2用以根据所述接收器信号获取模块获取的信号绘制经过电梯光幕的人员或/和物体的轮廓图像；通过各时间段中设定时间点各接收器感应到的是否被遮挡的信号，绘制设定时间点处于电梯光幕间的人员或/和物体的轮廓图像，可参阅图4-1、图4-2、图5-1、图5-2、图5-3、图5-4、图6-1、图6-2所示；并以此形成设定时间段内通过电梯光幕间的人员或/和物体的轮廓图像。
在一实施例中，所述轮廓绘制模块2用以获取各完整通过光幕动作时间段的物件轮廓；完整通过光幕动作时间段指：任意一个接收单元被遮挡为开始、至所有接收单元均不被遮挡为结束对应的时间段。
在本发明的一实施例中，所述轮廓绘制模块2包括：帧图像绘制单元、轮廓绘制单元。帧图像绘制单元用以根据各时间点各接收器接收的信号强度绘制对应帧的图像。轮廓绘制单元用以按时间点的次序将所述帧图像绘制单元绘制的帧图像依次拼接，形成对应物件的轮廓图像。在一实施例中，所述帧图像绘制单元用以将一帧的电梯光幕数据绘制成高与接收器数量存在设定关联的图像；根据各接收器接收的信号强度设定对应区域的灰度值，若接收器被遮挡，则对应区域的灰度值为第一灰度值；若接收器没有被遮挡，则对应区域的灰度值为第二灰度值。
在一实施例中，轮廓绘制模块2可以包括重复数据合并单元；重复数据合并单元用以在绘制的图像出现大量类似的重复数据的情况下，计算相邻帧的相关系数，若超过第一阈值B长度的相邻帧的相关系数均高于第二阈值C，则只保留长度为第一阈值B的帧数。可参阅图7-1、图7-2所示。
此外，所述轮廓绘制模块2还可以包括图像处理单元，用以对绘制后的图像进行图像处理；对于绘制后图像宽度小于指定宽度阈值范围最小值的图像，则填补对应图像，使其满足 设定的宽度要求；对于绘制图像宽度大于指定宽度阈值范围最大值的图像，则缩放对应图像的宽度，使图像宽度在指定宽度阈值范围内。
所述物件检测模块3用以根据所述轮廓绘制模块绘制的物件轮廓，识别得到出入电梯的物件的属性。
在识别进出电梯物件后，可以做出进一步的动作；如在本发明的一实施例中，所述系统进一步包括：安全隐患预警模块，用以在检测出设定物件进入电梯时，发出预警信息。
图2为本发明一实施例中出入电梯物件检测系统的组成示意图；请参阅图2，在本发明的一实施例中，所述系统进一步包括：数据模型建立模块4，用以利用卷积神经网络建立有关物件及绘制图像的数据模型。
在本发明的一实施例中，所述数据模型建立模块4用以搭建一个浅层卷积神经网络，第一层为输入层，输入图像大小为32*指定宽*光幕方向方向数量；第二层为卷积层，filter大小为3*3，filter个数为32个；第三层为池化层，filter大小为2*2，步长为2；第四层为卷积层，filter大小为3*3，filter个数为64个；第五层为池化层，filter大小为2*2，步长为2；第六层为全连接层，神经元个数为128个；第七层为输出层，神经元个数等于识别物体类别数。
请继续参阅图2，在本发明的一实施例中，所述系统进一步包括：数据训练模块5，用以建立一个超过设定阈值的训练集，训练集中存储有物件及对应的绘制图像。
在一实施例中，所述数据训练模块5包括：图像宽度设定单元、图像处理单元。图像宽度设定单元用以获取输入神经网络的原始图像，根据所有原始输入图像的宽度分布设定输入图像的指定宽度阈值范围。图像处理单元用以对原始输入图像进行图像处理；对于原始输入图像宽度小于指定宽度阈值范围最小值的原始输入图像，则填补对应原始输入图像，使其满足设定的宽度要求；对于原始输入图像宽度大于指定宽度阈值范围最大值的原始输入图像，则缩放对应原始输入图像的宽度，使图像宽度在指定宽度阈值范围内。
在本发明的一实施例中，所述数据训练模块5还包括重复数据合并单元；重复数据合并单元用以在绘制的图像出现大量类似的重复数据的情况下，计算相邻帧的相关系数，若超过第一阈值B长度的相邻帧的相关系数均高于第二阈值C，则只保留长度为第一阈值B的帧数。
在本发明的一实施例中，所述系统进一步包括：数据预处理模块，用以将训练集图像归一化，并将标签进行one-hot编码后传入卷积神经网络进行训练。
本发明揭示一种物件检测系统，在本发明的一实施例中，本发明物件检测系统不仅可以用在电梯领域，还可以用于其他领域。所述物件检测系统包括：设置于设定区域第一侧的若 干发射器、设置于设定区域第二侧的若干接收器、轮廓绘制模块及物件检测模块。
各发射器用以发射设定信号；各接收器用以接收对应发射器发出的信号。发射器可以水平发射信号，发射信号与水平方向也可以有一定角度。在一实施例中，一个发射器向一个接收器发射信号；在另一实施例中，一个发射器可以向多个接收器发射信号，一个接收器可以接收不同发射器发射的信号。
轮廓绘制模块用以根据各接收器获取的信号绘制经过设定区域的人员或/和物体的轮廓图像；通过各时间段中设定时间点各接收器感应到的是否被遮挡的信号，绘制设定时间点处于设定区域间的人员或/和物体的轮廓图像，并以此形成设定时间段内通过设定区域的人员或/和物体的轮廓图像。物件检测模块用以根据所述轮廓绘制模块绘制的物件轮廓，识别得到出入设定区域的物件的属性。轮廓绘制模块及物件检测模块的具体实现方式可以参考以上实施例的描述。
本发明揭示一种电梯光幕，包括上述的出入电梯物件检测系统。
本发明揭示一种电梯设备，包括上述的出入电梯物件检测系统。
本发明进一步揭示一种出入电梯物件检测方法，图3为本发明一实施例中出入电梯物件检测方法的流程图；请参阅图3，所述方法包括：
步骤S1、接收器信号获取步骤；获取电梯光幕各接收器接收的信号；
步骤S2、轮廓绘制步骤；根据所述接收器信号获取模块获取的信号绘制经过电梯光幕的人员或/和物体的轮廓图像；通过各时间段中设定时间点各接收器感应到的是否被遮挡的信号，绘制设定时间点处于电梯光幕间的人员或/和物体的轮廓图像，并以此形成设定时间段内通过电梯光幕间的人员或/和物体的轮廓图像；
步骤S3、物件检测步骤；根据所述轮廓绘制模块绘制的物件轮廓，识别得到出入电梯的物件的属性。
此外，本发明方法还包括：
数据模型建立步骤；利用卷积神经网络建立有关物件及绘制图像的数据模型；
数据训练步骤；建立一个超过设定阈值的训练集，训练集中存储有物件及对应的绘制图像。各步骤的具体实现过程可参阅以上有关系统的描述。
在本发明的一种使用场景中，电梯光幕设有32个接收器。所述方法具体包括：
步骤1、当红外发射器与对应的接收器之间有物体阻挡时，接收器信号值强度会明显降低，获取电梯光幕32个接收器的信号值强度，与预设的阈值A进行比较可以判断对应的光路上是否有遮挡物。将一帧的电梯光幕数据绘制成高32像素宽1像素的灰度图像，32个像素点分别 对应32个接收器，根据接收器信号强度判断，若被遮挡对应像素点的灰度值设为0，若没有被遮挡对应像素点的灰度值设为255，该灰度图像绘制的为对应时间点处于光幕间的物体的侧视图。结合考虑一段时间的电梯光幕数据可以绘制出随着人或物体进出电梯，接收器信号值随着被遮挡和失去遮挡信号值变化的情况，即处于光幕间的遮挡物侧视图的变化情况，由于图像完整记录了人或物体通过电梯光幕各个时刻的侧视图，可以明显观察到通过的人和物体的具体轮廓。
步骤2、以任意一个接收器被遮挡为开始，至所有接收器均不被遮挡为结束，可以认为是某物体的一次完整通过光幕的动作。绘制这段时间的遮挡物侧视图，并标注该物体的类别，作为训练集的一个数据。
步骤3、建立一个足够数量的训练集后，需要对数据进行一定的分析处理。由于输入神经网络的图片形状要一致，因此观察所有图片的宽度分布，选取一个适合的宽(对大部分数据的影响较小)作为输入图片的宽，原始图片宽度小于指定宽的图片随机在两侧补上不定宽度的灰度值为0的像素点将图片宽度填充至40，原始图片宽度大于指定宽的图片需要在不改变图片高度的前提下缩放图片的宽度使图片宽度等于指定宽。另外考虑到进出光幕时若停留一段时间，在绘制的图片上会出现大量的类似的重复数据，计算相邻帧的相关系数，若超过所设经验阈值B的相邻帧的相关系数均高于所设阈值C则只保留长度为B的帧数，其他的忽略不计，减少由于长时间停留导致的形状改变同时对于检测箱子等物体仍具有良好的适应性。
其中，X、Y为相邻的两帧数据，Cov(X,Y)为X与Y的协方差，Var[X]为X的方差，Var[Y]为Y的方差,r(X,Y)为求得的X与Y的相关系数。
步骤4、搭建一个浅层卷积神经网络，第一层为输入层，输入图片大小为32*指定宽*光幕方向方向数量；第二层为卷积层，filter大小为3*3，filter个数为32个；第三层为池化层，filter大小为2*2，步长为2；第四层为卷积层，filter大小为3*3，filter个数为64个；第四层为池化层，filter大小为2*2，步长为2；第五层为池化层，filter大小为2*2，步长为2；第六层为全连接层，神经元个数为128个；第七层为输出层，神经元个数等于识别物体类别数。
步骤5、将训练集图片归一化，并将标签进行one-hot编码后传入卷积神经网络进行训练。
步骤6、保存训练完的神经网络模型，当检测到有物体通过光幕时，绘制遮对应时间段的挡物侧视图，调整图片尺寸并归一化后加载神经网络模型预测该物体的类别，神经网络的输出中概率最大的那个类别即为神经网络预测的遮挡物类别，当检测到电瓶车等存在安全隐患的事物被带入电梯时发出预警。
综上所述，本发明提出的出入电梯物件检测系统及方法、物件检测系统、电梯光幕以及电梯设备，可识别出入电梯物件(包括人员或/和物体)，提高电梯设备的智能性。在识别出出入电梯物件后，可以便于电梯设备做出进一步的动作；例如，如果有不被允许的物体进入电梯，则可以发出报警信号。
以上所述实施例的各技术特征可以进行任意的组合，为使描述简洁，未对上述实施例中的各个技术特征所有可能的组合都进行描述，然而，只要这些技术特征的组合不存在矛盾，都应当认为是本说明书记载的范围。
这里本发明的描述和应用是说明性的，并非想将本发明的范围限制在上述实施例中。实施例中所涉及的效果或优点可因多种因素干扰而可能不能在实施例中体现，对于效果或优点的描述不用于对实施例进行限制。这里所披露的实施例的变形和改变是可能的，对于那些本领域的普通技术人员来说实施例的替换和等效的各种部件是公知的。本领域技术人员应该清楚的是，在不脱离本发明的精神或本质特征的情况下，本发明可以以其它形式、结构、布置、比例，以及用其它组件、材料和部件来实现。在不脱离本发明范围和精神的情况下，可以对这里所披露的实施例进行其它变形和改变。
Claims (16)
- 一种出入电梯物件检测系统，其特征在于，所述系统包括：接收器信号获取模块，用以获取电梯光幕各接收器接收的信号；轮廓绘制模块，用以根据所述接收器信号获取模块获取的信号绘制经过电梯光幕的人员或/和物体的轮廓图像；通过各时间段中设定时间点各接收器感应到的是否被遮挡的信号，绘制设定时间点处于电梯光幕间的人员或/和物体的轮廓图像，并以此形成设定时间段内通过电梯光幕间的人员或/和物体的轮廓图像；物件检测模块，用以根据所述轮廓绘制模块绘制的物件轮廓，识别得到出入电梯的物件的属性。
- 根据权利要求1所述的出入电梯物件检测系统，其特征在于：所述系统进一步包括：数据模型建立模块，用以利用卷积神经网络建立有关物件及绘制图像的数据模型。
- 根据权利要求2所述的出入电梯物件检测系统，其特征在于：所述系统进一步包括：数据训练模块，用以建立一个超过设定阈值的训练集，训练集中存储有物件及对应的绘制图像。
- 根据权利要求3所述的出入电梯物件检测系统，其特征在于：所述数据训练模块包括：图像宽度设定单元，用以获取输入神经网络的原始图像，根据所有原始输入图像的宽度分布设定输入图像的指定宽度阈值范围；图像处理单元，用以对原始输入图像进行图像处理；对于原始输入图像宽度小于指定宽度阈值范围最小值的原始输入图像，则填补对应原始输入图像，使其满足设定的宽度要求；对于原始输入图像宽度大于指定宽度阈值范围最大值的原始输入图像，则缩放对应原始输入图像的宽度，使图像宽度在指定宽度阈值范围内。
- 根据权利要求4所述的出入电梯物件检测系统，其特征在于：所述数据训练模块或/和轮廓绘制模块包括：重复数据合并单元，用以在绘制的图像出现大量类似的重复数据的情况下，计算相邻帧的相关系数，若超过第一阈值B长度的相邻帧的相关系数均高于第二阈值C，则只保留 长度为第一阈值B的帧数。
- 根据权利要求1所述的出入电梯物件检测系统，其特征在于：所述轮廓绘制模块包括：帧图像绘制单元，用以根据各时间点各接收器接收的信号强度绘制对应帧的图像；轮廓绘制单元，用以按时间点的次序将所述帧图像绘制单元绘制的帧图像依次拼接，形成对应物件的轮廓图像。
- 根据权利要求6所述的出入电梯物件检测系统，其特征在于：所述帧图像绘制单元用以将一帧的电梯光幕数据绘制成高与接收器数量存在设定关联的图像；根据各接收器接收的信号强度设定对应区域的灰度值，若接收器被遮挡，则对应区域的灰度值为第一灰度值；若接收器没有被遮挡，则对应区域的灰度值为第二灰度值。
- 根据权利要求1所述的出入电梯物件检测系统，其特征在于：所述轮廓绘制模块包括图像处理单元，用以对绘制后的图像进行图像处理；对于绘制后图像宽度小于指定宽度阈值范围最小值的图像，则填补对应图像，使其满足设定的宽度要求；对于绘制图像宽度大于指定宽度阈值范围最大值的图像，则缩放对应图像的宽度，使图像宽度在指定宽度阈值范围内。
- 根据权利要求2所述的出入电梯物件检测系统，其特征在于：所述数据模型建立模块用以搭建一个浅层卷积神经网络，第一层为输入层，输入图像大小为32*指定宽*光幕方向方向数量；第二层为卷积层，filter大小为3*3，filter个数为32个；第三层为池化层，filter大小为2*2，步长为2；第四层为卷积层，filter大小为3*3，filter个数为64个；第五层为池化层，filter大小为2*2，步长为2；第六层为全连接层，神经元个数为128个；第七层为输出层，神经元个数等于识别物体类别数。
- 根据权利要求3所述的出入电梯物件检测系统，其特征在于：所述系统进一步包括：数据预处理模块，用以将训练集图像归一化，并将标签进行one-hot编码后传入卷积神经网络进行训练。
- 根据权利要求1所述的出入电梯物件检测系统，其特征在于：所述系统进一步包括：安全隐患预警模块，用以在检测出设定物件进入电梯时，发出预警信息。
- 根据权利要求1所述的出入电梯物件检测系统，其特征在于：所述轮廓绘制模块用以获取各完整通过光幕动作时间段的物件轮廓；完整通过光幕动作时间段指：任意一个接收单元被遮挡为开始、至所有接收单元均不被遮挡为结束对应的时间段。
- 一种物件检测系统，其特征在于，所述系统包括：设置于设定区域第一侧的若干发射器，用以发射设定信号；设置于设定区域第二侧的若干接收器，用以接收对应发射器发出的信号；轮廓绘制模块，用以根据各接收器获取的信号绘制经过设定区域的人员或/和物体的轮廓图像；通过各时间段中设定时间点各接收器感应到的是否被遮挡的信号，绘制设定时间点处于设定区域间的人员或/和物体的轮廓图像，并以此形成设定时间段内通过设定区域的人员或/和物体的轮廓图像；以及物件检测模块，用以根据所述轮廓绘制模块绘制的物件轮廓，识别得到出入设定区域的物件的属性。
- 一种电梯光幕，其特征在于：包括权利要求1至12任一所述的出入电梯物件检测系统。
- 一种电梯设备，其特征在于：包括权利要求1至12任一所述的出入电梯物件检测系统。
- 一种出入电梯物件检测方法，其特征在于，所述方法包括：接收器信号获取步骤；获取电梯光幕各接收器接收的信号；轮廓绘制步骤；根据所述接收器信号获取模块获取的信号绘制经过电梯光幕的人员或/和物体的轮廓图像；通过各时间段中设定时间点各接收器感应到的是否被遮挡的信号， 绘制设定时间点处于电梯光幕间的人员或/和物体的轮廓图像，并以此形成设定时间段内通过电梯光幕间的人员或/和物体的轮廓图像；物件检测步骤；根据所述轮廓绘制模块绘制的物件轮廓，识别得到出入电梯的物件的属性。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010674638.9 | 2020-07-14 | ||
| CN202010674638.9A CN111762649B (zh) | 2020-07-14 | 2020-07-14 | 出入电梯物件检测系统及方法、物件检测系统、电梯光幕以及电梯设备 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2022011828A1 true WO2022011828A1 (zh) | 2022-01-20 |
Family
ID=72725344
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2020/116371 WO2022011828A1 (zh) | 2020-07-14 | 2020-09-21 | 出入电梯物件检测系统及方法、物件检测系统、电梯光幕以及电梯设备 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN111762649B (zh) |
| WO (1) | WO2022011828A1 (zh) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115849144A (zh) * | 2022-09-22 | 2023-03-28 | 杭州展特智能科技有限公司 | 用于货梯的货物识别与防撞控制方法 |
| CN117142281A (zh) * | 2023-09-21 | 2023-12-01 | 深圳市瀚强科技股份有限公司 | 一种电梯控制方法、装置和存储介质 |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112461155A (zh) * | 2020-11-02 | 2021-03-09 | 佳格科技(浙江)股份有限公司 | 光线跟踪识别物体的方法及电子设备 |
| CN112408170A (zh) * | 2020-11-16 | 2021-02-26 | 杭州奥立达电梯有限公司 | 一种多功能电梯光幕保护装置 |
| CN112433261A (zh) * | 2020-11-25 | 2021-03-02 | 猫岐智能科技（上海）有限公司 | 扫描成像监测系统及方法以及电动车入梯扫描成像监测系统 |
| CN112960505A (zh) * | 2021-02-20 | 2021-06-15 | 猫岐智能科技（上海）有限公司 | 一种识别电动车的方法 |
| CN113003340B (zh) * | 2021-02-22 | 2022-12-16 | 上海三菱电梯有限公司 | 电梯识别方法、装置与电梯 |
| CN113003339B (zh) * | 2021-02-22 | 2022-12-20 | 上海三菱电梯有限公司 | 电梯识别方法、识别系统与电梯 |
| CN113362406B (zh) * | 2021-04-26 | 2023-06-23 | 猫岐智能科技（上海）有限公司 | 光幕图像生成方法及系统以及识别系统 |
| CN113264440B (zh) * | 2021-06-04 | 2023-01-06 | 猫岐智能科技（上海）有限公司 | 光幕无光线区域获取系统及方法 |
| CN113256798B (zh) * | 2021-06-04 | 2023-05-05 | 猫岐智能科技（上海）有限公司 | 光幕遮挡区域生成系统及方法 |
| CN113776430B (zh) * | 2021-08-03 | 2023-11-28 | 邵阳先进制造技术研究院有限公司 | 一种基于光幕测量的尺寸数据处理方法 |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0344404A1 (de) * | 1988-06-03 | 1989-12-06 | Inventio Ag | Verfahren und Vorrichtung zur Steuerung der Türstellung einer automatischen Tür |
| CN1759613A (zh) * | 2003-03-20 | 2006-04-12 | 因温特奥股份公司 | 利用三维传感器对电梯范围内的空间监视 |
| EP1742182A2 (en) * | 2005-07-07 | 2007-01-10 | Electrolux Professional S.P.A. | Method for monitoring items passing through an entry/exit opening in a delimited space and apparatus to carry out such method |
| CN101723226A (zh) * | 2009-12-24 | 2010-06-09 | 杭州优迈科技有限公司 | 机器视觉三维探测电梯光幕的系统及方法 |
| CN105026300A (zh) * | 2013-03-18 | 2015-11-04 | 通力股份公司 | 电梯、用于监控楼层的移动门的开口和/或电梯轿厢的移动门的开口的光幕、以及用于在电梯中发出开门命令或关门命令的方法 |
| CN106006266A (zh) * | 2016-06-28 | 2016-10-12 | 西安特种设备检验检测院 | 一种应用于电梯安全监控的机器视觉建立方法 |
| CN108128678A (zh) * | 2017-12-25 | 2018-06-08 | 日立电梯（中国）有限公司 | 电梯运行控制方法、装置及系统 |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH1179632A (ja) * | 1997-09-03 | 1999-03-23 | Otis Elevator Co | エレベーターの乗客検出装置 |
| JPH11335046A (ja) * | 1998-05-28 | 1999-12-07 | Hitachi Building Systems Co Ltd | エレベータのかご閉じ込め防止装置 |
| GB2549761B (en) * | 2016-04-28 | 2018-04-25 | Ensota Guangzhou Tech Ltd | An automatic door installation and method of determining the presence of an obstacle |
-
2020
- 2020-07-14 CN CN202010674638.9A patent/CN111762649B/zh active Active
- 2020-09-21 WO PCT/CN2020/116371 patent/WO2022011828A1/zh active Application Filing
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0344404A1 (de) * | 1988-06-03 | 1989-12-06 | Inventio Ag | Verfahren und Vorrichtung zur Steuerung der Türstellung einer automatischen Tür |
| CN1759613A (zh) * | 2003-03-20 | 2006-04-12 | 因温特奥股份公司 | 利用三维传感器对电梯范围内的空间监视 |
| EP1742182A2 (en) * | 2005-07-07 | 2007-01-10 | Electrolux Professional S.P.A. | Method for monitoring items passing through an entry/exit opening in a delimited space and apparatus to carry out such method |
| CN101723226A (zh) * | 2009-12-24 | 2010-06-09 | 杭州优迈科技有限公司 | 机器视觉三维探测电梯光幕的系统及方法 |
| CN105026300A (zh) * | 2013-03-18 | 2015-11-04 | 通力股份公司 | 电梯、用于监控楼层的移动门的开口和/或电梯轿厢的移动门的开口的光幕、以及用于在电梯中发出开门命令或关门命令的方法 |
| CN106006266A (zh) * | 2016-06-28 | 2016-10-12 | 西安特种设备检验检测院 | 一种应用于电梯安全监控的机器视觉建立方法 |
| CN108128678A (zh) * | 2017-12-25 | 2018-06-08 | 日立电梯（中国）有限公司 | 电梯运行控制方法、装置及系统 |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115849144A (zh) * | 2022-09-22 | 2023-03-28 | 杭州展特智能科技有限公司 | 用于货梯的货物识别与防撞控制方法 |
| CN115849144B (zh) * | 2022-09-22 | 2024-04-26 | 杭州展特智能科技有限公司 | 用于货梯的货物识别与防撞控制方法 |
| CN117142281A (zh) * | 2023-09-21 | 2023-12-01 | 深圳市瀚强科技股份有限公司 | 一种电梯控制方法、装置和存储介质 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111762649A (zh) | 2020-10-13 |
| CN111762649B (zh) | 2021-07-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2022011828A1 (zh) | 出入电梯物件检测系统及方法、物件检测系统、电梯光幕以及电梯设备 | |
| CN107944359B (zh) | 基于视频的火焰检测方法 | |
| CN104691559B (zh) | 监测站台安全门与车门区间安全的辅助系统及其实现方法 | |
| KR101793636B1 (ko) | 통합 지능형 cctv 영상 처리 시스템 | |
| CN112804498A (zh) | 一种安全防护装具佩戴情况监控系统和方法 | |
| CN110021133B (zh) | 全天候消防火灾巡逻预警监控系统及火灾图像检测方法 | |
| CN103366506A (zh) | 一种驾驶员行车途中接打手机行为的自动监控装置及方法 | |
| CN109637068A (zh) | 智能烟火识别系统 | |
| CN107272637A (zh) | 一种视频监控系统故障自检自恢复控制系统及方法 | |
| CN110240036B (zh) | 一种检测试图进入电梯的电动自行车的装置和方法 | |
| CN109508667B (zh) | 一种电梯视频防夹方法及电梯视频监控装置 | |
| CN106815958A (zh) | 报警系统、报警分析/显示装置，报警分析/显示方法 | |
| CN104847211A (zh) | 一种站台安全门与车门区间安全的辅助系统及其实现方法 | |
| CN110759199A (zh) | 一种检测试图进入电梯的电动自行车的装置和方法 | |
| CN211236276U (zh) | 一种铁路站台端部入侵智能监测系统 | |
| CN111523386A (zh) | 一种基于机器视觉的高速铁路站台门监测防护方法及系统 | |
| CN110703350A (zh) | 列车门与屏蔽门空隙异物检测系统和方法 | |
| CN112158705B (zh) | 光幕成像方法及系统、识别方法及系统、光幕装置及电梯设备 | |
| CN101540891A (zh) | 基于监控视频的行李传送仓人体检测系统 | |
| CN109345788A (zh) | 一种基于视觉特征的监控预警系统 | |
| CN112723077A (zh) | 一种基于光流的电梯开关门检测方法 | |
| CN114436086B (zh) | 电梯轿厢昏暗的识别与告警方法及系统 | |
| CN110002305B (zh) | 一种基于计算机视觉技术的电梯门缝线检测方法与系统 | |
| CN113516691A (zh) | 一种基于机器视觉的高空抛物检测系统 | |
| CN113158888A (zh) | 一种电梯异常视频识别方法 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 20945389 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 20945389 Country of ref document: EP Kind code of ref document: A1 |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 20945389 Country of ref document: EP Kind code of ref document: A1 |
|
| 32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205 DATED 07.07.23) |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 20945389 Country of ref document: EP Kind code of ref document: A1 |