Filtering Devices in Security Social Networks

The patent segments the large set of member devices into smaller subsets based on predetermined conditions such as device type, location, time of day, and event relevance. This segmentation allows the server to selectively request data from specific device subsets rather than all devices, thereby reducing the total data volume while maintaining comprehensive security coverage.

The patent implements local quality by applying different data collection criteria to different spatial and temporal contexts. For example, during a security event at a specific venue, the system adjusts which devices within the area of interest should transmit data based on their proximity to the event, device capabilities, and historical behavior patterns, optimizing data collection efficiency for each local context.

The patent applies preliminary action by requiring devices to undergo preliminary checks before image acquisition is authorized. Devices must first satisfy predetermined conditions such as being in an appropriate location, having proper orientation, and meeting quality thresholds before capturing images. This preliminary filtering prevents unwanted images from being captured in the first place, rather than filtering them afterward.

The patent implements feedback mechanisms where devices continuously report their status (location, orientation, motion state) to the server, which then provides real-time feedback on whether image acquisition should proceed. This closed-loop control ensures that only images meeting the predetermined criteria are captured, dynamically adjusting acquisition based on current device conditions.