CN113642461A - Elevator door opening and closing state identification method and device based on deep learning - Google Patents
Elevator door opening and closing state identification method and device based on deep learning Download PDFInfo
- Publication number
- CN113642461A CN113642461A CN202110927257.1A CN202110927257A CN113642461A CN 113642461 A CN113642461 A CN 113642461A CN 202110927257 A CN202110927257 A CN 202110927257A CN 113642461 A CN113642461 A CN 113642461A
- Authority
- CN
- China
- Prior art keywords
- elevator door
- deep learning
- video data
- elevator
- state
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Indicating And Signalling Devices For Elevators (AREA)
- Image Analysis (AREA)
- Elevator Control (AREA)
Abstract
The invention discloses an elevator door opening and closing state identification method and device based on deep learning, electronic equipment and a computer storage medium. Wherein the method comprises the following steps: the method comprises the steps of obtaining first video data in an elevator, transmitting the first video data to a deep learning identification model when the pixel change condition of an elevator door area in the first video data is larger than a first threshold value, and outputting the state of a first elevator door by the deep learning identification model. The elevator door opening and closing state is monitored by adopting a computer vision technology, the cost is effectively reduced, and the training data of the deep learning recognition model comprises enough scenes, so that the elevator door opening and closing state recognition method has stronger generalization and high recognition accuracy.
Description
Technical Field
The invention relates to the technical field of image recognition, in particular to a method and a device for recognizing the opening and closing state of an elevator door based on deep learning, electronic equipment and a computer storage medium.
Background
With the continuous improvement of the computer industry level, the computer vision system and the artificial intelligence gradually become practical, and the computer vision develops rapidly due to the massive data brought by the rise of the internet and the wide application of the machine learning method, is widely applied to a plurality of fields related to computers, is more and more close to life, and is closely related to people. The computer vision takes the image as input, realizes the correct expression of the environment, researches the characteristic organization of the image, carries out target detection or scene recognition, and further provides scientific explanation for the event. The deep learning technology is the development direction of the current technology, and a large number of data sets and a complex convolutional neural network can realize strong expression capability, so that the model has high accuracy and robustness. However, the traditional computer vision algorithm process is separated, the accuracy rate is not high when encountering complex background, the deep learning algorithm can solve the problem and can realize end-to-end mapping, and at present, the combination of computer vision and deep learning has made remarkable progress.
The method for monitoring the opening and closing state of the elevator door based on computer vision is beneficial to saving cost, and the existing method for identifying the opening and closing state of the elevator door mainly uses a traditional optical flow algorithm, an algorithm based on deep learning target detection and the like, and has the problems of inaccuracy and long time consumption caused by motion blur.
Disclosure of Invention
The invention mainly aims to provide an elevator door opening and closing state identification method, device, electronic equipment and computer storage medium based on deep learning, and aims to solve part or all of the technical problems mentioned in the background technology.
In order to achieve the above object, a first aspect of the present invention provides a deep learning-based method for identifying an open/close door state of an elevator, the method comprising:
the method comprises the steps of obtaining first video data in an elevator, transmitting the first video data to a deep learning identification model when the pixel change condition of an elevator door area in the first video data is larger than a first threshold value, and outputting the state of a first elevator door by the deep learning identification model.
Preferably, calculating the pixel change condition of the elevator door area in the first video data comprises:
calculating all pixel change conditions of adjacent frames of an elevator door area, counting the number of changed pixels, and calculating the pixel change conditions based on the number of the pixels, wherein the pixel change conditions comprise the following steps:
taking the number of the changed pixels obtained by statistics as the pixel change condition; or the like, or, alternatively,
and taking the value obtained by converting the counted number of the changed pixels as the pixel change condition.
Preferably, the first video data is subjected to occupant identification counting, and a value of the first threshold is determined based on a counting result, wherein the value of the first threshold and the counting result have a positive correlation.
Preferably, the first elevator door state includes a fully opened door, a half opened door, and a closed door.
Preferably, before transmitting the first video data to the deep learning recognition model, the method further comprises a training step, including:
collecting second video data in the elevator, extracting a plurality of video frames containing the opening and closing states of the elevator door, carrying out manual marking on the video frames, and sending the video frames and corresponding marks into the deep learning identification model to obtain the trained deep learning identification model.
Preferably, the deep learning identification model comprises:
the convolution layer module is used for carrying out convolution and pooling processing on input video data for multiple times and extracting depth features with discriminability;
the three-branch convolutional layer module is used for learning characteristics output by the last layer of the convolutional layer module through different branches, wherein the convolutional layer weight of the three-branch convolutional layer module is randomly initialized in the training process;
a classifier layer module for classifying features output by the three-branch convolutional layer module
Preferably, after the deep learning identification model outputs that the first elevator door state is half-open door or closed door, the method further comprises:
analyzing a connected domain of a sub-image of an elevator door region based on the first video data, if the connected domain with the area larger than a second threshold value exists, calling third video data, inputting the third video data into the deep learning identification model, outputting a second elevator door state by the deep learning identification model, and if the second elevator door state is the same as the first elevator door state, outputting the first elevator door state as a final elevator door state; otherwise, performing connected domain analysis on the elevator door region subimage based on the third video data, and outputting the second elevator door state as the final elevator door state if the connected domain with the area larger than a second threshold value does not exist.
The invention provides an elevator door opening and closing state recognition device based on deep learning, which comprises an acquisition module, a transmission module, a deep learning recognition module and an output module, wherein the acquisition module is used for acquiring the state of an elevator door; wherein the content of the first and second substances,
the acquisition module is used for acquiring first video data in the elevator;
the transmission module is used for transmitting the first video data to the deep learning identification module when the pixel change condition of the elevator door area in the first video data is larger than a first threshold value,
the deep learning identification model is used for identifying the first video data and outputting the state of a first elevator door.
A third aspect of the invention provides an electronic device, the device comprising:
a memory storing executable program code;
a processor coupled with the memory;
the processor calls the executable program code stored in the memory to execute the method as described above.
A fourth aspect of the invention provides a computer storage medium having stored thereon computer instructions which, when invoked, perform the method as described above.
The technical scheme of the invention has the beneficial effects that:
1) the door opening and closing state of the elevator is monitored based on a computer vision technology, so that the deployment of sensors can be reduced, and the cost is effectively saved;
2) the elevator door state is rapidly recognized by utilizing the deep learning recognition model obtained through training, and as long as training data comprise enough scenes, the deep learning recognition model has stronger generalization and high recognition accuracy.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the structures shown in the drawings without creative efforts.
Fig. 1 is a schematic flowchart of a method for identifying an open/close door state of an elevator based on deep learning according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of an elevator door opening and closing state identification device based on deep learning according to a second embodiment of the present invention;
fig. 3 is a schematic structural diagram of an electronic device according to a third embodiment of the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
In order that the objects and advantages of the invention will be more clearly understood, the invention is further described below with reference to examples; it should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only for explaining the technical principle of the present invention, and do not limit the scope of the present invention.
It should be noted that in the description of the present invention, the terms of direction or positional relationship indicated by the terms "upper", "lower", "left", "right", "inner", "outer" are based on the directions or positional relationships shown in the drawings, which are merely for convenience of description, and do not indicate or imply that the device or element must have a specific orientation, be constructed in a specific orientation, and be operated, and thus, should not be construed as limiting the present invention.
Furthermore, it should be noted that, in the description of the present invention, unless otherwise explicitly specified and limited, the terms "mounted", "connected" and "connected" are to be construed broadly, e.g., as being capable of being fixedly connected, detachably connected, or integrally connected; can be a mechanical connection, but also an electrical connection; can be directly connected or indirectly connected through intervening media, and can communicate between the two elements. The specific meaning of the above terms in the present invention can be understood by those skilled in the art according to specific situations.
The first embodiment is as follows:
referring to fig. 1, fig. 1 is a schematic flowchart of an elevator door opening and closing state identification method based on deep learning according to an embodiment of the present application. As shown in fig. 1, an elevator door opening and closing state identification method based on deep learning according to an embodiment of the present application includes:
the method comprises the steps of obtaining first video data in an elevator, transmitting the first video data to a deep learning identification model when the pixel change condition of an elevator door area in the first video data is larger than a first threshold value, and outputting the state of a first elevator door by the deep learning identification model.
In the embodiment of the invention, the first video data in the elevator is sent to the deep learning identification model for classification and identification, so that the state of the elevator door can be quickly obtained. When the pixel change of the elevator door area is large, the switching of the open/close state of the elevator door is shown, or the switching of the open/close state is about to exist (for example, when the elevator is about to open the door, a passenger can be transposed to the vicinity of the elevator door, and the pixel change is large), and the detection and identification of the invalid state of the elevator door can be effectively reduced by identifying the elevator door at the moment. Of course, in this case, the first video data transmitted to the deep learning identification model may include video data before and/or after the "pixel change condition is greater than the first threshold" to implement the aforementioned condition of the occurring or impending transition of the gate state.
Preferably, calculating the pixel change condition of the elevator door area in the first video data comprises:
calculating all pixel change conditions of adjacent frames of an elevator door area, counting the number of changed pixels, and calculating the pixel change conditions based on the number of the pixels, wherein the pixel change conditions comprise the following steps:
taking the number of the changed pixels obtained by statistics as the pixel change condition; or the like, or, alternatively,
and taking the value obtained by converting the counted number of the changed pixels as the pixel change condition.
In the embodiment of the present invention, the number of pixels that have changed may be directly used as the value representing the change condition of the pixel, or the number may be subjected to conversion processing, for example, a ratio of the change amount to the total number of pixels is obtained, and the present invention is not limited in this respect.
Preferably, the first video data is subjected to occupant identification counting, and a value of the first threshold is determined based on a counting result, wherein the value of the first threshold and the counting result have a positive correlation.
In the embodiment of the invention, the more passengers in the elevator, the higher the possibility and the frequency of the occurrence of the pixel change, and in this case, in order to avoid the false recognition caused by the normal movement of the passengers, the magnitude of the first threshold value is adjusted in positive correlation based on the number of passengers, namely, the first threshold value is larger as the number of passengers is larger.
Preferably, the first elevator door state includes a fully opened door, a half opened door, and a closed door.
Preferably, before transmitting the first video data to the deep learning recognition model, the method further comprises a training step, including:
collecting second video data in the elevator, extracting a plurality of video frames containing the opening and closing states of the elevator door, carrying out manual marking on the video frames, and sending the video frames and corresponding marks into the deep learning identification model to obtain the trained deep learning identification model.
In the embodiment of the invention, video information in an elevator is collected firstly, the video containing the door opening and closing state of the elevator is processed in a frame division mode, the video is converted into pictures, samples are selected, the samples are labeled manually, the original pictures and labeled label results are put into a deep learning identification model for training, the characteristics of the learning samples are made, and therefore, a target model with the identification accuracy reaching the standard is obtained.
Preferably, the deep learning identification model comprises:
the convolution layer module is used for carrying out convolution and pooling processing on input video data for multiple times and extracting depth features with discriminability;
the three-branch convolutional layer module is used for learning characteristics output by the last layer of the convolutional layer module through different branches, wherein the convolutional layer weight of the three-branch convolutional layer module is randomly initialized in the training process;
a classifier layer module for classifying features output by the three-branch convolutional layer module
In the embodiment of the invention, the multi-branch convolutional layer module is provided with three branches, and the characteristics of the last layer of the convolutional layer module can be learned through different branches, so that more information in channels can be obtained, and the effect of fine tuning the weight of the highest layer in the neural network is achieved through randomly initializing the weight of the convolutional layer. In addition, in order to avoid the overfitting problem of the model, a classifier layer module can be arranged to select an L2 regularization term.
Preferably, after the deep learning identification model outputs that the first elevator door state is half-open door or closed door, the method further comprises:
analyzing a connected domain of a sub-image of an elevator door region based on the first video data, if the connected domain with the area larger than a second threshold value exists, calling third video data, inputting the third video data into the deep learning identification model, outputting a second elevator door state by the deep learning identification model, and if the second elevator door state is the same as the first elevator door state, outputting the first elevator door state as a final elevator door state; otherwise, performing connected domain analysis on the elevator door region subimage based on the third video data, and outputting the second elevator door state as the final elevator door state if the connected domain with the area larger than a second threshold value does not exist.
In the embodiment of the invention, because the situation of passengers in the elevator is complex, the accuracy of elevator door state identification cannot be ensured by only depending on the video image in the elevator. For example, passengers often enter the elevator with large objects (such as furniture with a certain height, decorative plates, etc.), and the objects may block the elevator door, and at this time, a wrong door state recognition result is likely to be obtained based on the first video data. For the problem, the connected domain analysis is firstly carried out on the first video data, when the connected domain (such as furniture with a certain height, a board material for decoration and the like) with the area larger than a second threshold value is found, the accuracy of the state of the first elevator door is difficult to guarantee, third video data (such as video data of a camera which can shoot the elevator door to the outer side outside the elevator corresponding to a floor) positioned outside the elevator is further taken for identification, and if the internal and external identification results are the same, the state of the first elevator door is accurate and is directly output; if the internal and external recognition results are different, whether a passenger carrying large articles (such as furniture with a certain height, a decoration plate and the like) outside the elevator door accurately enters the elevator is further judged, the connected domain analysis is also carried out at the moment, if the connected domain with the area larger than the second threshold value does not exist outside the elevator door, the state of the second elevator door is more credible, and the second elevator door is used as the final elevator door state to be output. In addition, when the outside also has a connected domain with the area larger than the second threshold, it indicates that the door state cannot be effectively identified based on the inside and outside data, and at this time, the elevator can be temporarily started and an alarm sound is given to wait for the passenger to move the carried article to a certain extent so as to clearly judge the elevator door state (for example, when the middle area of the elevator door is completely exposed). Of course, when the area of the connected domain is larger than the third threshold value, the article is too large, the elevator is not suitable for being carried, and the starting of the elevator can be suspended at the moment, and the passenger can be warned.
Example two
Referring to fig. 2, fig. 2 is a schematic structural diagram of an elevator door opening and closing state identification device based on deep learning according to an embodiment of the present application. As shown in fig. 2, an elevator door opening and closing state identification device based on deep learning according to an embodiment of the present application includes an acquisition module, a transmission module, a deep learning identification module, and an output module; wherein the content of the first and second substances,
the acquisition module is used for acquiring first video data in the elevator;
the transmission module is used for transmitting the first video data to the deep learning identification module when the pixel change condition of the elevator door area in the first video data is larger than a first threshold value,
the deep learning identification model is used for identifying the first video data and outputting the state of a first elevator door.
The specific functions of the elevator door opening and closing state identification device based on deep learning in this embodiment refer to the first embodiment, and since the system in this embodiment adopts all technical solutions of the first embodiment, at least all beneficial effects brought by the technical solutions of the first embodiment are achieved, and no further description is given here.
EXAMPLE III
Referring to fig. 3, fig. 3 is an electronic device disclosed in an embodiment of the present application, where the electronic device includes:
a memory storing executable program code;
a processor coupled with the memory;
the processor calls the executable program code stored in the memory to execute the elevator door opening and closing state identification method based on deep learning according to the embodiment one.
Example four
The embodiment of the application also discloses a computer storage medium, wherein a computer program is stored on the storage medium, and when the computer program is executed by a processor, the elevator door opening and closing state identification method based on deep learning is executed.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present application should be covered within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (10)
1. A method for identifying the door opening and closing state of an elevator based on deep learning is characterized by comprising the following steps: the method comprises the following steps:
the method comprises the steps of obtaining first video data in an elevator, transmitting the first video data to a deep learning identification model when the pixel change condition of an elevator door area in the first video data is larger than a first threshold value, and outputting the state of a first elevator door by the deep learning identification model.
2. The method of claim 1, wherein: calculating pixel change conditions of an elevator door area in the first video data, comprising:
calculating all pixel change conditions of adjacent frames of an elevator door area, counting the number of changed pixels, and calculating the pixel change conditions based on the number of the pixels, wherein the pixel change conditions comprise the following steps:
taking the number of the changed pixels obtained by statistics as the pixel change condition; or the like, or, alternatively,
and taking the value obtained by converting the counted number of the changed pixels as the pixel change condition.
3. The method of claim 2, wherein: and carrying out occupant identification counting on the first video data, and determining the value of the first threshold value based on a counting result, wherein the value of the first threshold value and the counting result have a positive correlation relationship.
4. A method according to any one of claims 1-3, characterized in that: the first elevator door state comprises a full-open door, a half-open door and a closed door.
5. The method of claim 1, wherein: before transmitting the first video data to a deep learning recognition model, the method further comprises a training step, including:
collecting second video data in the elevator, extracting a plurality of video frames containing the opening and closing states of the elevator door, carrying out manual marking on the video frames, and sending the video frames and corresponding marks into the deep learning identification model to obtain the trained deep learning identification model.
6. The method of claim 5, wherein: the deep learning identification model comprises:
the convolution layer module is used for carrying out convolution and pooling processing on input video data for multiple times and extracting depth features with discriminability;
the three-branch convolutional layer module is used for learning characteristics output by the last layer of the convolutional layer module through different branches, wherein the convolutional layer weight of the three-branch convolutional layer module is randomly initialized in the training process;
and the classifier layer module is used for classifying the characteristics output by the three-branch convolutional layer module.
7. The method of claim 4, wherein: after the deep learning identification model outputs that the state of the first elevator door is half-open or closed, the method further comprises the following steps:
analyzing a connected domain of a sub-image of an elevator door region based on the first video data, if the connected domain with the area larger than a second threshold value exists, calling third video data, inputting the third video data into the deep learning identification model, outputting a second elevator door state by the deep learning identification model, and if the second elevator door state is the same as the first elevator door state, outputting the first elevator door state as a final elevator door state; otherwise, performing connected domain analysis on the elevator door region subimage based on the third video data, and outputting the second elevator door state as the final elevator door state if the connected domain with the area larger than a second threshold value does not exist.
8. The utility model provides an elevator switch door state recognition device based on degree of depth study which characterized in that: the device comprises an acquisition module, a transmission module, a deep learning identification module and an output module; wherein the content of the first and second substances,
the acquisition module is used for acquiring first video data in the elevator;
the transmission module is used for transmitting the first video data to the deep learning identification module when the pixel change condition of the elevator door area in the first video data is larger than a first threshold value,
the deep learning identification model is used for identifying the first video data and outputting the state of a first elevator door.
9. An electronic device, characterized in that: the apparatus comprises:
a memory storing executable program code;
a processor coupled with the memory;
the processor calls the executable program code stored in the memory to perform the method of any of claims 1-7.
10. A computer storage medium, characterized in that: the storage medium stores computer instructions which, when invoked, perform the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110927257.1A CN113642461A (en) | 2021-08-13 | 2021-08-13 | Elevator door opening and closing state identification method and device based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110927257.1A CN113642461A (en) | 2021-08-13 | 2021-08-13 | Elevator door opening and closing state identification method and device based on deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113642461A true CN113642461A (en) | 2021-11-12 |
Family
ID=78421272
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110927257.1A Pending CN113642461A (en) | 2021-08-13 | 2021-08-13 | Elevator door opening and closing state identification method and device based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113642461A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116757253A (en) * | 2023-07-03 | 2023-09-15 | 浙江永贵博得交通设备有限公司 | Intelligent automatic learning door opening and closing algorithm for rail transit |
| CN117115740A (en) * | 2023-09-05 | 2023-11-24 | 北京智芯微电子科技有限公司 | Method, device and equipment for detecting elevator door opening and closing state based on deep learning |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2013100008A (en) * | 2011-11-08 | 2013-05-23 | Nippon Signal Co Ltd:The | Apparatus and method for detecting open and close of vehicle door |
| CN204175091U (en) * | 2014-09-21 | 2015-02-25 | 王俭俭 | Automatic door control system |
| CN104821025A (en) * | 2015-04-29 | 2015-08-05 | 广州运星科技有限公司 | Passenger flow detection method and detection system thereof |
| US20160348398A1 (en) * | 2015-06-01 | 2016-12-01 | Schlage Lock Company Llc | Door improvements and data mining via accelerometer and magnetometer electronic component |
| JP2017134685A (en) * | 2016-01-28 | 2017-08-03 | セコム株式会社 | Lift, control method of lift, and security system |
| JP2018095350A (en) * | 2016-12-09 | 2018-06-21 | 株式会社日立ビルシステム | Remote monitoring device |
| KR20190036443A (en) * | 2017-09-27 | 2019-04-04 | 최유강 | A door monitoring system and a process thereof |
| CN110002302A (en) * | 2018-08-09 | 2019-07-12 | 浙江新再灵科技股份有限公司 | A kind of elevator switch door detection system and method based on deep learning |
| CN110002314A (en) * | 2018-08-09 | 2019-07-12 | 浙江新再灵科技股份有限公司 | A kind of stranded number detection system of the elevator passenger based on deep learning |
| CN110127479A (en) * | 2019-04-17 | 2019-08-16 | 浙江工业大学 | A kind of elevator door switch method for detecting abnormality based on video analysis |
| CN110751091A (en) * | 2019-10-18 | 2020-02-04 | 江西理工大学 | Convolutional neural network model for static image behavior recognition |
| CN111807183A (en) * | 2020-07-20 | 2020-10-23 | 北京电通慧梯物联网科技有限公司 | Elevator door state intelligent detection method based on deep learning |
| CN112733668A (en) * | 2020-12-31 | 2021-04-30 | 青岛海纳云科技控股有限公司 | Video deep learning-based detection method for single elevator taking of pets in elevator |
-
2021
- 2021-08-13 CN CN202110927257.1A patent/CN113642461A/en active Pending
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2013100008A (en) * | 2011-11-08 | 2013-05-23 | Nippon Signal Co Ltd:The | Apparatus and method for detecting open and close of vehicle door |
| CN204175091U (en) * | 2014-09-21 | 2015-02-25 | 王俭俭 | Automatic door control system |
| CN104821025A (en) * | 2015-04-29 | 2015-08-05 | 广州运星科技有限公司 | Passenger flow detection method and detection system thereof |
| US20160348398A1 (en) * | 2015-06-01 | 2016-12-01 | Schlage Lock Company Llc | Door improvements and data mining via accelerometer and magnetometer electronic component |
| JP2017134685A (en) * | 2016-01-28 | 2017-08-03 | セコム株式会社 | Lift, control method of lift, and security system |
| JP2018095350A (en) * | 2016-12-09 | 2018-06-21 | 株式会社日立ビルシステム | Remote monitoring device |
| KR20190036443A (en) * | 2017-09-27 | 2019-04-04 | 최유강 | A door monitoring system and a process thereof |
| CN110002302A (en) * | 2018-08-09 | 2019-07-12 | 浙江新再灵科技股份有限公司 | A kind of elevator switch door detection system and method based on deep learning |
| CN110002314A (en) * | 2018-08-09 | 2019-07-12 | 浙江新再灵科技股份有限公司 | A kind of stranded number detection system of the elevator passenger based on deep learning |
| CN110127479A (en) * | 2019-04-17 | 2019-08-16 | 浙江工业大学 | A kind of elevator door switch method for detecting abnormality based on video analysis |
| CN110751091A (en) * | 2019-10-18 | 2020-02-04 | 江西理工大学 | Convolutional neural network model for static image behavior recognition |
| CN111807183A (en) * | 2020-07-20 | 2020-10-23 | 北京电通慧梯物联网科技有限公司 | Elevator door state intelligent detection method based on deep learning |
| CN112733668A (en) * | 2020-12-31 | 2021-04-30 | 青岛海纳云科技控股有限公司 | Video deep learning-based detection method for single elevator taking of pets in elevator |
Non-Patent Citations (2)
| Title |
|---|
| 李双全: "双目立体视觉在电梯安全检测中的应用", 《中国优秀硕士学位论文全文数据库信息科技辑》, no. 4, pages 31 - 32 * |
| 金晓磊 等: "机器人视觉的电梯轿厢门状态识别系统", 《单片机与嵌入式系统应用》, vol. 18, no. 4 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116757253A (en) * | 2023-07-03 | 2023-09-15 | 浙江永贵博得交通设备有限公司 | Intelligent automatic learning door opening and closing algorithm for rail transit |
| CN117115740A (en) * | 2023-09-05 | 2023-11-24 | 北京智芯微电子科技有限公司 | Method, device and equipment for detecting elevator door opening and closing state based on deep learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107123131B (en) | Moving target detection method based on deep learning | |
| CN110929593B (en) | Real-time significance pedestrian detection method based on detail discrimination | |
| CN107679471B (en) | Indoor personnel air post detection method based on video monitoring platform | |
| CN107915102B (en) | Elevator blocking door behavior detection system and detection method based on video analysis | |
| CN110298297B (en) | Flame identification method and device | |
| CN110880172A (en) | Video face tampering detection method and system based on cyclic convolution neural network | |
| CN107833221A (en) | A kind of water leakage monitoring method based on multi-channel feature fusion and machine learning | |
| CN108447078A (en) | The interference of view-based access control model conspicuousness perceives track algorithm | |
| CN113642461A (en) | Elevator door opening and closing state identification method and device based on deep learning | |
| CN105184818A (en) | Video monitoring abnormal behavior detection method and detections system thereof | |
| CN109657581A (en) | Urban track traffic gate passing control method based on binocular camera behavioral value | |
| EP2549759A1 (en) | Method and system for facilitating color balance synchronization between a plurality of video cameras as well as method and system for obtaining object tracking between two or more video cameras | |
| CN113536972B (en) | Self-supervision cross-domain crowd counting method based on target domain pseudo label | |
| KR102391853B1 (en) | System and Method for Processing Image Informaion | |
| CN111967433A (en) | Action identification method based on self-supervision learning network | |
| US20150262374A1 (en) | Method and apparatus for moving object detection using fisher's linear discriminant based radial basis function network | |
| CN113688761A (en) | Pedestrian behavior category detection method based on image sequence | |
| Dahirou et al. | Motion Detection and Object Detection: Yolo (You Only Look Once) | |
| CN111402185B (en) | Image detection method and device | |
| CN116443682A (en) | Intelligent elevator control system | |
| CN112581498B (en) | Road side shielding scene vehicle robust tracking method for intelligent vehicle road system | |
| CN115797770A (en) | Continuous image target detection method, system and terminal considering relative movement of target | |
| Delahoz et al. | A deep-learning-based floor detection system for the visually impaired | |
| CN114694090A (en) | Campus abnormal behavior detection method based on improved PBAS algorithm and YOLOv5 | |
| CN113496501A (en) | Method and system for detecting invader in dynamic scene based on video prediction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |