Human tracking device and human tracking program
A technology for tracking devices and people, which is applied in the field of people tracking programs, and can solve problems such as overlapped outlines and inapplicability
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment approach 1
[0072] figure 1 It is a configuration diagram showing a person tracking device according to Embodiment 1 of the present invention.
[0073] exist figure 1 Among them, a plurality of cameras 1 constituting the imaging unit are installed at different positions in the upper part of the elevator car as the monitoring target area, and simultaneously take pictures of the inside of the car from different angles.
[0074] However, the type of camera 1 is not particularly limited, and it may be a general surveillance camera, a visible camera, a high-sensitivity camera capable of photographing near-infrared regions, a far-infrared camera capable of photographing heat sources, and the like. In addition, an infrared distance sensor capable of measuring a distance, a laser range finder, or the like may be used instead.
[0075] The image acquisition unit 2 is an image input interface for acquiring images of the inside of the elevator car captured by the plurality of cameras 1 , and per...
Embodiment approach 2
[0599] In Embodiment 1 above, a plurality of three-dimensional movement trajectory graphs are searched, candidates for three-dimensional movement paths satisfying the entry and exit conditions are calculated, candidates for three-dimensional movement paths from entry to exit are listed, and the cost function is calculated by using MCMC or the like Probability is maximized to find the best combination of candidates for three-dimensional movement trajectories. However, in the case of complex structures of three-dimensional movement trajectory diagrams, sometimes the number of candidates for three-dimensional movement trajectories that meet the entry and exit conditions becomes an astronomical number, which cannot be achieved in practice. processing within the time.
[0600] Therefore, in the second embodiment, labels are added to the vertices of the three-dimensional trajectory graph (each three-dimensional trajectory constituting the graph), and the cost function considering ent...
Embodiment approach 3
[0683] In the above-mentioned second embodiment, labels are added to vertices of the three-dimensional movement trajectory graph (each three-dimensional movement trajectory constituting the graph), and the cost function considering entry and exit conditions is probabilistically maximized, thereby estimating the three-dimensional movement in real time. The best combination of trajectories. However, when the number of people captured in the image increases and the structure of the two-dimensional trajectory map is complicated, the number of candidates for the three-dimensional trajectory fragments obtained as a result of stereoscopic vision becomes astronomical. method, and it is impossible to end the processing within the actual time.
[0684] Therefore, in Embodiment 3, labels are probabilistically added to the vertices of the two-dimensional movement trajectory graph (each two-dimensional movement trajectory constituting the graph), and stereoscopic vision of the three-dimens...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com