Video Surveillance Parallel Processing Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Modern video surveillance systems face challenges in processing video analytics in real-time due to the resource-intensive nature of tasks like face detection and recognition, which can take several minutes or even hours, especially when dealing with complex data sequences, leading to delays in incident response.

Innovation Solution

Implementing parallel processing across multiple processors and cores to divide video streams into self-contained chunks, allowing simultaneous analysis by different application programs on various digital devices, leveraging deep-learning and neural networks for faster analytics results.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If sequential processing is used for video analytics, then processing accuracy can be maintained, but processing time increases linearly with data complexity and volume

Engineering Contradiction:
Improveprocessing timeVSAvoidanalytics throughput
Core Design Contradiction:
Loss of timeVSProductivity

Solution Approach 1:

The patent segments video data into discrete frames or chunks that can be processed independently. Each frame or chunk is divided into smaller units that can be distributed across multiple processing cores, enabling parallel analysis without compromising overall accuracy while significantly reducing total processing time.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transitions from sequential single-core processing to parallel multi-core processing by adding a temporal and spatial distribution dimension. Multiple application programs run simultaneously on different processing cores, transforming the processing architecture from one-dimensional sequence to multi-dimensional parallel execution.

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

2Measurement precision

If deep-learning analytics are applied to achieve reliable face recognition and person detection, then detection accuracy improves, but computational resource requirements and processing time increase

Engineering Contradiction:
Improvedetection accuracyVSAvoidcomputational resource consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent segments complex deep-learning analytics into separate application programs that can be distributed across multiple processing cores. Each core executes specialized deep-learning models for specific tasks (face detection, person detection, recognition), reducing the computational burden on any single core while maintaining high accuracy through distributed deep-learning processing.

Inventive Principle:
Principle #1Segmentation

3Productivity

If multiple application programs process different video streams simultaneously, then processing throughput increases, but system complexity increases

Engineering Contradiction:
Improveprocessing throughputVSAvoidsystem architecture complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent implements a universal processing framework where multiple application programs can be deployed on a common platform of processing cores. The system provides standardized interfaces and data formats that allow different deep-learning models to operate simultaneously without requiring custom integration for each program, reducing architectural complexity while enabling high throughput.

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

Data Source

PatentEP4391525A1Video surveillance system, computer-implemented video management process, computer program and storage medium
Publication Date: 2024.06.26 MILESTONE SYSTEMS
  • EP4391525A1 patent drawingFigure 1
  • EP4391525A1 patent drawingFigure 2~3
  • EP4391525A1 patent drawingFigure 4

AI summary

The present invention relates to a video surveillance system comprising a video management system, a plurality of computer-implemented digital devices and a plurality of application programs, the system being configured to run the programs across the plurality of computer-implemented digital devices, and configured to perform parallel processing of different groups of pictures from the said at least one video stream with the plurality of application programs across the plurality of computer-implemented digital devices.