Conveyance Video Analytics Metadata Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conveyance systems, such as elevator systems, face challenges with bandwidth consumption and human observer limitations in detecting subtle changes, especially in structures with multiple systems operating in parallel, where video feeds require significant resources and human observers struggle to notice anomalies.
Innovation Solution
A method involving a video camera that captures image data and initiates analytics to determine various conditions within the conveyance system, summarizing the status as metadata for a support system to initiate corrective actions, including luminescence levels, component damage, operational status, occupancy, door operations, and potential vandalism, utilizing machine learning to adapt to variations in lighting and system arrangements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If video feeds from multiple conveyance systems are transmitted to a centralized surveillance station, then monitoring capability is improved, but bandwidth consumption increases significantly
Solution Approach 1:
The patent extracts only the essential information from video feeds by implementing analytics that generate metadata outputs representing specific conditions (luminescence levels, component damage, operational status, occupancy, door operations, entrapment, vandalism). This selective extraction of critical information rather than transmitting complete video feeds significantly reduces bandwidth consumption while maintaining monitoring capability.
Solution Approach 2:
The patent introduces an intermediary analytics processing layer between the video cameras and the centralized surveillance station. This intermediary processes video data locally to generate condensed metadata outputs, which then are transmitted to the surveillance station. This intermediary layer acts as a mediator that transforms high-volume video data into low-volume informative metadata, resolving the bandwidth contradiction.
2Reliability
If multiple dedicated video links are used for parallel conveyance systems, then monitoring reliability is improved, but system complexity increases
Solution Approach 1:
The patent implements a universal analytics platform that processes video data from multiple conveyance systems through a single integrated system. The analytics engine performs multiple functions (detecting luminescence, damage, occupancy, door operations, entrapment, vandalism) across all systems, eliminating the need for multiple dedicated video links and reducing overall system complexity while maintaining monitoring reliability.
Solution Approach 2:
The patent merges the monitoring of multiple conveyance systems into a unified analytics framework. Instead of maintaining separate dedicated video links for each system, the analytics platform consolidates processing of video feeds from multiple sources, generating unified metadata outputs that represent the state of all monitored systems. This merging approach reduces complexity while preserving comprehensive monitoring capability.
3Device complexity
If human observers monitor video feeds to detect abnormal conditions, then system simplicity is maintained, but detection precision of subtle changes deteriorates
Solution Approach 1:
The patent replaces the mechanical human observation system with an automated analytics system that processes video data. The analytics engine automatically detects subtle changes in conditions (luminescence levels, component damage, operational status, occupancy, door operations, entrapment, vandalism) with precision beyond human capability. This substitution maintains relative system simplicity while dramatically improving detection precision of subtle anomalies.
Solution Approach 2:
The patent transforms video data into different parameter representations through analytics processing. The system changes the parameter space from raw pixel data to meaningful metadata parameters (luminescence levels, damage indicators, occupancy counts, door cycle counts, entrapment detection, vandalism indicators). This parameter transformation enables precise detection of subtle changes that would be imperceptible to human observers while maintaining an automated yet conceptually simple system architecture.
Data Source
AI summary
According to an aspect, a method includes capturing image data from a video camera at a conveyance system. Analytics of the image data can be initiated to determine a plurality of conditions of the conveyance system. A status of the conditions can be summarized as a metadata output. The metadata output can be transmitted to a support system operable to initiate a corrective action responsive to the status of the conditions.