Accurate and efficient gait data detection system
By combining deep learning algorithms and multi-source data fusion technology with distributed computing, efficient gait recognition in complex environments has been achieved, improving recognition accuracy and processing speed.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HUIZHOU KESHENG INFORMATION TECH CO LTD
- Filing Date
- 2025-03-12
- Publication Date
- 2026-05-01
AI Technical Summary
Existing gait recognition systems have low accuracy in complex environments and are difficult to adapt to changes in lighting, similar clothing among people, and items they are carrying.
The method employs convolutional neural networks and recurrent neural networks from deep learning algorithms, combined with multi-source data fusion technology, and utilizes high-definition cameras from multiple scenes and distributed computing to extract and recognize gait features.
It improves recognition accuracy in complex environments, reduces false positives and false negatives, and can quickly process large amounts of gait data to meet real-time detection needs.
Smart Images
Figure CN121963292A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of security detection technology, and in particular to a precise and efficient gait data detection system. Background Technology
[0002] Security monitoring refers to the use of various technologies and equipment to monitor and record specific areas or targets in real time in order to prevent and respond to various security risks. The core purpose of security monitoring is to protect life and property and prevent illegal intrusion, theft, vandalism, and other acts. Specialized security monitoring systems are required for conducting such monitoring.
[0003] Currently, common personnel identification technologies in the security detection field include fingerprint recognition and facial recognition. Fingerprint recognition requires contact with the target object to obtain fingerprint information, limiting its application scenarios; facial recognition is affected by factors such as lighting, occlusion, and posture changes. Gait recognition, as an emerging biometric identification technology, identifies individuals by analyzing the posture characteristics of human walking, and has the advantages of long-distance, non-contact data collection, thus gradually gaining attention.
[0004] Currently, some simple gait recognition systems mainly rely on a single camera to collect images and use traditional image feature extraction methods to analyze gait. However, the recognition accuracy is low and it is difficult to adapt to complex and ever-changing real-world scenarios. Therefore, we provide a precise and efficient gait data detection system. Summary of the Invention
[0005] To address the problem that existing systems in the aforementioned background technology face difficulties in gait feature extraction under complex environments, such as large changes in lighting, similar clothing worn by individuals, and items carried, leading to a significant decrease in recognition accuracy, this invention provides a precise and efficient gait data detection system.
[0006] This invention is achieved using the following technical solution: a precise and efficient gait data detection system, comprising:
[0007] Data statistics module: used to evaluate system performance based on metrics such as the number of data comparisons and the matching rate;
[0008] Gait comparison module: Used to generate data and provide data to the data statistics module;
[0009] Operation Log Module: Used to record user information and behaviors in detail during the operation of each module;
[0010] System Management Module: The various sub-modules under the System Management Module can be interconnected to jointly manage system permissions and interface display;
[0011] System settings module: Used to provide basic configurations for other modules.
[0012] As a further improvement to the above solution, the data statistics module performs the following process for data statistics:
[0013] Users upload videos, and the system automatically extracts gait features based on advanced algorithms and compares them with templates in the database to obtain verification results. The data statistics module then performs data statistical analysis accordingly.
[0014] As a further improvement to the above scheme, the gait feature recognition process is as follows:
[0015] The system uses convolutional neural networks (CNN) and recurrent neural networks (RNN) from deep learning algorithms to extract gait features from videos. Through the collaborative work of the two, the system can accurately extract unique gait features.
[0016] Convolutional neural networks (CNNs) are responsible for capturing static features of human walking posture, such as the proportions of different body parts and the position of joints.
[0017] Recurrent Neural Networks (RNNs) focus on analyzing the dynamic time-series features of gait, such as walking rhythm and stride length variations.
[0018] As a further improvement to the above solution, when no personnel data exists in the database, the system will automatically retrieve massive amounts of data from urban security monitoring scenarios for gradual comparison.
[0019] As a further improvement to the above solution, the system employs distributed computing and parallel processing technologies to quickly filter out personnel data that are similar to human gait characteristics.
[0020] As a further improvement to the above scheme, the comparison process combines multi-source data fusion technology to comprehensively analyze gait features with environmental information, time information, etc., in order to improve the accuracy of recognition.
[0021] As a further improvement to the above solution, the data is collected using high-definition cameras with multiple scenes and perspectives, in conjunction with advanced data acquisition equipment.
[0022] As a further improvement to the above scheme, the data is cleaned and preprocessed using data cleaning and preprocessing techniques to remove noise and interference, thereby improving data quality.
[0023] As a further improvement to the above solution, the system management module includes department management, user management, role management, and menu management.
[0024] As a further improvement to the above solution, the system management module will assign corresponding permissions based on the logged-in user's role.
[0025] Compared with the prior art, the beneficial effects of the present invention are as follows:
[0026] 1. This invention utilizes advanced deep learning algorithms and multi-source data fusion technology to enable the system to accurately extract and analyze gait features in complex environments, greatly improving the accuracy of personnel identification and effectively reducing the false positive and false negative rates.
[0027] 2. By employing parallel computing technology and optimized data processing flow, this invention enables the system to quickly process large amounts of gait data to meet the needs of real-time detection, facilitating timely and accurate identification and tracking of people even in scenarios where multiple people are present simultaneously. Attached Figure Description
[0028] Figure 1 This is a flowchart of the present invention;
[0029] Figure 2 This is a flowchart of the data processing process of the present invention. Detailed Implementation
[0030] The present invention will now be further described in conjunction with the accompanying drawings and specific embodiments. It should be noted that, without conflict, the various embodiments or technical features described below can be arbitrarily combined to form new embodiments.
[0031] Example 1:
[0032] Please combine Figure 1-2 This embodiment of a precise and efficient gait data detection system includes:
[0033] Data statistics module: used to evaluate system performance based on metrics such as the number of data comparisons and the matching rate;
[0034] The data statistics module performs the following statistical analysis:
[0035] Users upload videos, and the system automatically extracts gait features based on advanced algorithms and compares them with templates in the database to obtain verification results. The data statistics module then performs data statistical analysis accordingly.
[0036] The gait feature recognition process is as follows:
[0037] The system uses convolutional neural networks (CNN) and recurrent neural networks (RNN) from deep learning algorithms to extract gait features from videos. Through the collaborative work of the two, the system can accurately extract unique gait features.
[0038] Convolutional neural networks (CNNs) are responsible for capturing static features of human walking posture, such as the proportions of different body parts and the position of joints.
[0039] Recurrent Neural Networks (RNNs) focus on analyzing the dynamic time-series features of gait, such as walking rhythm and stride length variations.
[0040] Gait comparison module: Used to generate data and provide data to the data statistics module;
[0041] The system employs distributed computing and parallel processing technologies to quickly filter out data on individuals whose gait characteristics are similar to those of other individuals.
[0042] The comparison process combines multi-source data fusion technology to comprehensively analyze gait features with environmental and temporal information in order to improve the accuracy of identification.
[0043] The data is collected using high-definition cameras from multiple scenes and perspectives, along with advanced data acquisition equipment, to ensure that the collected gait data is comprehensive and accurate.
[0044] Data cleaning and preprocessing techniques are used to remove noise and interference from the data in order to improve data quality.
[0045] Operation Log Module: Used to record user information and behaviors in detail during the operation of each module;
[0046] System Management Module: The various sub-modules under the System Management Module can be interconnected to jointly manage system permissions and interface display;
[0047] The system management module includes department management, user management, role management, and menu management.
[0048] The system management module will assign corresponding permissions based on the logged-in user's role.
[0049] System settings module: Used to provide basic configurations for other modules.
[0050] The implementation principle of a precise and efficient gait data detection system in this application embodiment is as follows:
[0051] During data collection, users first log in to the system, which will assign corresponding permissions based on user roles. When performing gait comparison, users can quickly obtain verification results by uploading videos that meet the format requirements. System administrators can manage and adjust parameters in the system management module and system settings module. Ordinary users can view data statistics results to understand the comparison situation. All operations will be fully recorded by the operation log module.
[0052] In emergency situations, videos provided by on-site personnel may be unreliable for facial recognition due to various reasons, such as dim lighting or facial obstruction. This is where the advantages of this gait system become apparent. Staff can upload videos of individuals to the system, which quickly initiates the gait recognition process. First, it uses Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) from deep learning algorithms to extract gait features from the video. CNNs capture static features of human walking posture, such as the proportions of different body parts and joint positions; RNNs focus on analyzing dynamic time-series features of the gait, such as walking rhythm and stride length variations. Through the collaborative work of both, the system can accurately extract unique gait features.
[0053] If the personnel data is not in the established database of special personnel characteristics, the system will automatically retrieve massive amounts of data from urban security monitoring scenarios for step-by-step comparison. The system employs distributed computing and parallel processing technologies to quickly filter out personnel data similar to the individual's gait characteristics. During the comparison process, multi-source data fusion technology is combined to comprehensively analyze gait characteristics with environmental and temporal information, further improving the accuracy of identification. For example, considering the time and location of the person's appearance, data on personnel who appeared in the same time period and similar areas are selected for focused comparison. In this way, the system can efficiently and accurately confirm the identity of personnel from complex urban security monitoring data, providing crucial clues for case investigation and emergency response.
[0054] Example 2:
[0055] This embodiment is an improvement on embodiment 1, in that: when no personnel data exists in the database, the system will automatically retrieve massive amounts of data from urban security monitoring scenarios for step-by-step comparison, which facilitates efficient and accurate identification of personnel and provides key clues for case investigation and emergency response.
[0056] The above embodiments are merely preferred embodiments of the present invention and should not be construed as limiting the scope of protection of the present invention. Any non-substantial changes and substitutions made by those skilled in the art based on the present invention shall fall within the scope of protection claimed by the present invention.
Claims
1. A precise and efficient gait data detection system, characterized in that, include: Data statistics module: used to evaluate system performance based on metrics such as the number of data comparisons and the matching rate; Gait comparison module: Used to generate data and provide data to the data statistics module; Operation Log Module: Used to record user information and behaviors in detail during the operation of each module; System Management Module: The various sub-modules under the System Management Module can be interconnected to jointly manage system permissions and interface display; System settings module: Used to provide basic configurations for other modules.
2. The precise and efficient gait data detection system as described in claim 1, characterized in that, The data statistics module performs the following statistical analysis: Users upload videos, and the system automatically extracts gait features based on advanced algorithms and compares them with templates in the database to obtain verification results. The data statistics module then performs data statistical analysis accordingly.
3. The precise and efficient gait data detection system as described in claim 2, characterized in that, The gait feature recognition process is as follows: The system uses convolutional neural networks (CNN) and recurrent neural networks (RNN) from deep learning algorithms to extract gait features from videos. Through the collaborative work of the two, the system can accurately extract unique gait features. Convolutional Neural Networks (CNNs) are responsible for capturing static features of human walking posture, such as the proportions of different body parts and the position of joints. Recurrent Neural Networks (RNNs) focus on analyzing the dynamic time-series features of gait, such as walking rhythm and stride length variations.
4. The precise and efficient gait data detection system as described in claim 3, characterized in that, When no personnel data exists in the database, the system will automatically retrieve massive amounts of data from urban security monitoring scenarios for step-by-step comparison.
5. The precise and efficient gait data detection system as described in claim 4, characterized in that, The system employs distributed computing and parallel processing technologies to quickly filter out data on individuals whose gait characteristics are similar to those of other individuals.
6. The precise and efficient gait data detection system as described in claim 5, characterized in that, The comparison process combines multi-source data fusion technology to comprehensively analyze gait features with environmental information, time information, etc., in order to improve the accuracy of recognition.
7. The precise and efficient gait data detection system as described in claim 6, characterized in that, The data is collected using high-definition cameras with multiple scenes and perspectives, in conjunction with advanced data acquisition equipment.
8. The precise and efficient gait data detection system as described in claim 7, characterized in that, The data is cleaned and preprocessed using data cleaning and preprocessing techniques to remove noise and interference, thereby improving data quality.
9. The precise and efficient gait data detection system as described in claim 8, characterized in that, The system management module includes department management, user management, role management, and menu management.
10. The precise and efficient gait data detection system as described in claim 9, characterized in that, The system management module will assign corresponding permissions based on the logged-in user's role.