Network Camera Metadata Extraction for Surveillance Reliability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Surveillance systems face issues with communication errors and low reliability due to massive video data transmission, and long search times for specific video channels, making it difficult to provide services to subscribers.
Innovation Solution
A surveillance system where network cameras capture and store video data, extract metadata, and transmit it to a server, allowing for efficient data management through a subscriber management server, event server, and search server, reducing the need for all cameras to transmit video data directly to the server, and enabling quick search and retrieval of event data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If compressed video data from multiple network cameras is simultaneously transmitted to the server, then complete surveillance data is available, but communication errors frequently occur and reliability decreases
Solution Approach 1:
The patent extracts only essential metadata (motion detection results, object attributes, event information) from the video data and transmits it to the server, while the full video data remains stored locally in the network camera. This selective extraction resolves the contradiction by transmitting minimal data for reliability while preserving complete surveillance capability locally.
Solution Approach 2:
The patent introduces a metadata extraction module as an intermediary between the video capture system and the server. This intermediary processes video data locally and transmits only processed metadata, acting as a buffer that reduces transmission volume while maintaining surveillance functionality through on-demand video retrieval.
2Loss of time
If a massive amount of video data is stored in the server, then all video channels are available, but search time for specific video increases
Solution Approach 1:
The patent performs preliminary processing of video data by extracting metadata (object detection, motion patterns, event markers) before storage. This metadata is stored in the server along with indexes to local video data, enabling rapid search and direct retrieval of relevant video segments without scanning entire video datasets.
Solution Approach 2:
The patent adds a metadata dimension to the traditional video storage model. Instead of storing only video data, the system creates a parallel metadata layer containing searchable attributes (object type, location, time, event type), allowing searches to operate in this new dimension and quickly locate relevant video data without scanning all stored video.
3Productivity
If all network cameras transmit video data to the server, then centralized monitoring is achieved, but communication network bandwidth is excessively consumed
Solution Approach 1:
The patent extracts only essential metadata (motion detection results, object attributes, event information) from the video data and transmits it to the server, while the full video data remains stored locally in the network camera. This selective extraction resolves the contradiction by transmitting minimal data for reliability while preserving complete surveillance capability locally.
Solution Approach 2:
The patent applies partial action by transmitting only the necessary portion of video data (metadata) to the server, rather than transmitting all video data. The server receives sufficient information for monitoring and search purposes, while full video quality is maintained locally for on-demand retrieval.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach reduces communication errors, improves surveillance function reliability, and allows for faster video search and retrieval, enabling various services to be provided to subscribers efficiently.
Implementation Method 1
The photoelectric converter converts light incident via the optical component to electric signals
Data Source
AI summary
Provided is a surveillance system including a plurality of network cameras; and at least one server connected to the network cameras via a communication network, wherein each of the network cameras captures and stores video data, extracts metadata, which is a basis of event data, from the video data, transmits the metadata to the server, and wherein, if requested by the server, each of the network cameras transmits the stored video data to the server.