Intelligent Video Recording with Adaptive Resolution Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video surveillance systems face challenges in optimizing hard disk utilization and bandwidth usage due to the storage of high-resolution video data streams, particularly when most of the recorded data does not add value to users, as they save entire streams without differentiating between areas of interest and non-interest.

Innovation Solution

The system intelligently records video data streams by saving areas of interest in high resolution and areas of non-interest in low resolution, allowing for independent recording and transmission of multiple streams from a single camera with different resolutions, and includes features like motion detection, QR code association, and camera adjustments for improved image clarity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If high resolution video data streams are saved for all areas, then video quality is improved, but hard disk space consumption increases

Engineering Contradiction:
Improvevideo qualityVSAvoidhard disk space consumption
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent applies local quality by differentiating resolution requirements across different spatial regions. Areas of interest (AOI) are identified and recorded at high resolution, while non-interest areas are recorded at low resolution. This is achieved through frame analysis that detects motion and identifies AOI, then selectively applying different encoding resolutions to different regions of the video frame, thereby optimizing storage space while maintaining quality where needed.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The video stream is segmented into multiple regions based on area of interest detection. The system divides each video frame into AOI and non-AOI regions, processes them separately with different resolution settings, and combines them into a unified output stream. This segmentation allows independent resolution control for different parts of the video, resolving the contradiction between overall quality and storage efficiency.

Inventive Principle:
Principle #1Segmentation

2Loss of information

If multiple video data streams are transmitted from a single camera, then user value is improved, but bandwidth usage increases

Engineering Contradiction:
Improveuser valueVSAvoidbandwidth usage
Core Design Contradiction:
Loss of informationVSLoss of energy

Solution Approach 1:

The system dynamically adjusts the number and resolution of video streams based on real-time area of interest detection. Instead of transmitting multiple full-resolution streams continuously, the system creates adaptive streams where only AOI regions are transmitted at high resolution, and non-AOI regions are transmitted at low resolution or omitted entirely. This dynamic approach maintains user value by providing relevant high-quality data while reducing bandwidth consumption through selective transmission.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP3506227B1Systems and methods for intelligently recording video data streams
Publication Date: 2020.11.04 ADEMCO INC
  • EP3506227B1 patent drawingFigure 1

AI summary

Systems and methods for intelligently recording video data streams are provided. Some methods can include determining whether an area of interest in a first high resolution video data stream of a field of view of a camera includes motion. When the area of interest includes the motion, the area of interest can be cropped from the first high resolution video data stream to form a second high resolution video data stream of the area of interest, and the first high resolution video data stream of an area of non-interest in the field of view can be converted into a first low resolution video data stream of the area of non-interest. When the area of interest fails to include the motion, the first high resolution video data stream of the field of view can be converted into a second low resolution video data stream of the field of view.