Unlock instant, AI-driven research and patent intelligence for your innovation.
Pedestrian tracking method and device in escalator scene based on LSTM model, and medium
What is Al technical title?
Al technical title is built by PatSnap Al team. It summarizes the technical point description of the patent document.
A pedestrian tracking and escalator technology, applied in the field of pedestrian tracking, can solve the problems of limiting the early warning speed of the escalator early warning system and the efficiency bottleneck of the escalator scene, and achieve the effect of improving the accuracy, improving the tracking effect and ensuring the tracking efficiency.
Active Publication Date: 2020-11-24
HUAQIAO UNIVERSITY +1
View PDF7 Cites 5 Cited by
Summary
Abstract
Description
Claims
Application Information
AI Technical Summary
This helps you quickly interpret patents by identifying the three key elements:
Problems solved by technology
Method used
Benefits of technology
Problems solved by technology
[0004]
However, the current tracking methods used in the escalator warning system are purely using Kalman filter or particle filter. These methods cannot effectively combine the escalator scene, resulting in an efficiency bottleneck, which greatly limits the warning speed of the escalator warning system.
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more
Image
Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
Click on the blue label to locate the original text in one second.
Reading with bidirectional positioning of images and text.
Smart Image
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0070] This embodiment provides a pedestrian tracking method based on the LSTM model escalator scene, such as figure 1 shown, including:
[0071] Obtain and preprocess the video stream of pedestrians in the escalator scene in the surveillance video;
[0072] Create a target detection model (such as the YOLOv5 model), detect the preprocessed video stream, and obtain the tracking target;
[0073] Utilize Convolutional Neural Networks (Convolutional Neural Networks, CNN) to classify the tracking target according to the position of the pedestrian; And generate a target tracking track; when a tracking target is classified as a tracking target outside the scope of the escalator, end the tracking of the tracking target;
[0074] Use the neural network to train the trajectory data of pedestrians in the normal operation scene of the escalator, and obtain the routine movement trajectory of pedestrians;
[0075] Use the LSTM-based trajectory matching model to match the target tracking...
Embodiment 2
[0106] In this embodiment, a pedestrian tracking device based on the LSTM model escalator scene is provided, such as figure 2 As shown, including: preprocessing module, target detection module, classification tracking module, conventional motion trajectory module, matching module and key tracking module;
[0107] The preprocessing module is used to obtain and preprocess the video stream of pedestrians in the escalator scene in the monitoring video;
[0108] The target detection module is used to create a target detection model, detect the preprocessed video stream, and obtain a tracking target;
[0109] The classification tracking module is used to classify the tracking target according to the position of the pedestrian by using the convolutional neural network; when a tracking target is classified as a tracking target within the scope of the escalator, the tracking target is tracked using the inter-frame relationship , and generate a target tracking track; when a tracking t...
Embodiment 3
[0126] This embodiment provides a computer-readable storage medium, such as image 3 As shown, a computer program is stored thereon, and when the computer program is executed by a processor, any implementation manner in the first embodiment can be realized.
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More
PUM
Login to View More
Abstract
The invention discloses a pedestrian tracking method in an escalator scene based on an LSTM model. The pedestrian tracking method comprises the following steps: acquiring a video stream of pedestriansin the escalator scene in a monitoring video; creating a target detection model, and detecting the video stream to obtain a tracking target; classifying the tracking target according to the positionof the pedestrian, screening the tracking target, tracking the tracking target, and generating a target tracking trajectory; training the pedestrian trajectory data under the normal running conditionof the escalator by using a neural network to obtain a pedestrian conventional motion trajectory; matching the target tracking trajectory with the pedestrian conventional motion trajectory by using anLSTM-based trajectory matching model, and distinguishing a target with an abnormal trajectory; and carrying out key tracking on the abnormal tracking target. According to the pedestrian tracking method in the escalator scene based on the LSTM model, disclosed by the embodiment of the invention, the trajectory matching method of the LSTM model is fused with the escalator scene to classify and track and optimize multiple targets, so that the detection speed, the tracking efficiency and the tracking precision are improved.
Description
technical field [0001] The invention relates to the technical field of safety monitoring, in particular to a pedestrian tracking method, device and medium based on an LSTM model in an escalator scene. Background technique [0002] With the improvement of economic conditions, people's requirements for safety continue to increase. Surveillance cameras have covered almost every corner of public life. Traditional video surveillance is difficult to play the role of timely warning and alarm. To ensure real-time monitoring of abnormal behavior and take effective measures in a timely manner, a lot of manpower is required. In the face of multi-channel surveillance systems, the effect of human identification is limited, which has created the rise of intelligent video surveillance technology. [0003] Today, intelligent video surveillance technology has a wide range of application scenarios. Escalators are one of the most common devices in public places. Due to their potential safet...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More
Application Information
Patent Timeline
Application Date:The date an application was filed.
Publication Date:The date a patent or application was officially published.
First Publication Date:The earliest publication date of a patent with the same application number.
Issue Date:Publication date of the patent grant document.
PCT Entry Date:The Entry date of PCT National Phase.
Estimated Expiry Date:The statutory expiry date of a patent right according to the Patent Law, and it is the longest term of protection that the patent right can achieve without the termination of the patent right due to other reasons(Term extension factor has been taken into account ).
Invalid Date:Actual expiry date is based on effective date or publication date of legal transaction data of invalid patent.
Login to View More 
Login to View More