Driving DVR Event Metadata Retrieval

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing digital video recorder (DVR) systems for vehicular fleet management rely on frame-based timelines, making it difficult to retrieve video segments based on events such as driver identity, vehicle location, speed, or weather conditions, which are not efficiently searchable or manageable in terms of storage and processing.

Innovation Solution

A driving-oriented DVR system that captures video and generates metadata tags corresponding to specific driving events, allowing for the identification and retrieval of video segments based on event types and locations, using event objects and metadata tags to efficiently store and query video data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If video is stored and retrieved based on frame-based timelines, then video playback functionality is maintained, but retrieval efficiency for event-based queries deteriorates

Engineering Contradiction:
Improvevideo retrieval efficiencyVSAvoiddata structure complexity
Core Design Contradiction:
Ease of operationVSDevice complexity

Solution Approach 1:

The patent introduces metadata as an intermediary layer between the video frames and query operations. This metadata contains event information (speed, location, driver identity, weather conditions) that enables efficient event-based retrieval without requiring complex searches through the actual video frame data. The metadata acts as a mediator that translates event-based queries into efficient data access operations.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent segments video data into discrete events with associated metadata. Each event (such as speeding, location-based events, driver actions) is treated as a separate searchable unit with its own metadata record. This segmentation allows the system to retrieve specific event segments independently rather than processing entire video timelines, dramatically improving retrieval efficiency for targeted queries.

Inventive Principle:
Principle #1Segmentation

2Reliability

If all video frames are stored in memory, then complete video data is preserved, but storage space requirements increase

Engineering Contradiction:
Improvevideo data completenessVSAvoidstorage space
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The patent extracts only the essential event information from the video data and stores it as metadata, while the full video frames are maintained for completeness. This extraction approach preserves the complete video data for reliability while reducing the amount of data that needs to be actively managed and queried, as metadata provides a compact representation of key events.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent creates metadata copies of event information that reference the actual video frames. Instead of duplicating the entire video data, the system creates lightweight metadata records that contain event details and references to the corresponding video segments. This copying strategy maintains data completeness while significantly reducing storage requirements for searchable content.

Inventive Principle:
Principle #26Copying

3Productivity

If video retrieval is based on timeline position only, then simple sequential access is achieved, but event-based search capability deteriorates

Engineering Contradiction:
Improvequery processing speedVSAvoidsearch criterion flexibility
Core Design Contradiction:
ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent creates a universal metadata structure that serves multiple functions: it enables timeline-based retrieval, event-based retrieval, and various filtered searches (by speed, location, driver, weather conditions). This multi-functional metadata system allows the same data structure to support diverse query types without requiring separate processing mechanisms, thereby improving both productivity and adaptability.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent adds a new dimension to video retrieval by introducing event-based metadata attributes beyond simple timeline positioning. Instead of only temporal sequencing, the system incorporates spatial (location), contextual (weather conditions), and behavioral (driver actions, vehicle speed) dimensions. This dimensional expansion enables versatile event-based searching while maintaining efficient query processing through structured metadata.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Data Source

PatentEP3707910B1Driving oriented digital video recorder system
Publication Date: 2023.09.20 BENDIX COMMERCIAL VEHICLE SYSTEMS LLC
  • EP3707910B1 patent drawingFigure 1
  • EP3707910B1 patent drawingFigure 2
  • EP3707910B1 patent drawingFigure 3~5

AI summary

A driving oriented digital video recorder (DVR) system may comprise a camera; a memory; and a control circuit. The occurrence of a predefined driving event type may be determined based on received data. An event object that includes details of the determined event may be generated, including the type of the determined event and a time association with a segment of captured driving video. Metadata tags may be generated, which contain the type of the determined event and the location of the event object in the memory of the DVR. The DVR may retrieve segments of the driving video by identifying the metadata tag as having the same event type as a query parameter and, in turn, identify the event object using the location of the event object included in the metadata tag. The driving video segment may then be located via the association with the event object.