Video Surveillance Extraction via Camera Viewshed Intersection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current video surveillance methods require extensive manual effort to find and review videos from multiple cameras in a specified region, especially in densely populated areas, as they lack efficient ways to pinpoint relevant camera footage based on geographic information and time periods.

Innovation Solution

The method involves calculating the intersection between a target location and camera viewsheds to identify relevant cameras, extracting videos that match specific conditions, and reducing the time spent on artificial checks by using geographic information system (GIS) spatial calculations to determine the shooting time periods of these cameras.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If all lenses within a specified region are found and their videos are reviewed manually, then the relevant video footage can be identified, but the time and manpower costs increase significantly in areas with dense video lenses

Engineering Contradiction:
Improvevideo retrieval accuracyVSAvoidvideo review time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary spatial intersection calculations between camera viewsheds and target locations before video review, pre-identifying which cameras are relevant to the query. This preliminary filtering action eliminates the need for manual review of all cameras in the region, directly reducing video review time while maintaining retrieval accuracy

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent replaces the manual mechanical process of reviewing videos with an automated computer-based system that performs spatial intersection calculations. The system automatically determines which camera viewsheds intersect with the target location and retrieves only those videos, substituting human effort with computational automation

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If manual checking of long time period videos is performed to determine which lenses shot the target, then relevant footage can be found, but the energy and time consumption increases

Engineering Contradiction:
Improvevideo retrieval reliabilityVSAvoidvideo processing energy
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The system extracts only the necessary video segments by calculating intersection relations between camera viewsheds and target locations. Instead of processing entire long-time-period videos from all cameras, the system extracts and processes only the specific time periods and camera feeds that are relevant to the query, reducing energy consumption while maintaining retrieval reliability

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies partial action by performing spatial intersection calculations only for cameras that may have captured the target, rather than processing all cameras in the region. The system calculates viewshed intersections to identify a subset of relevant cameras, thereby reducing the overall computational energy required while ensuring reliable retrieval

Inventive Principle:
Principle #16Partial or excessive action

3Productivity

If videos from multiple cameras are extracted and reviewed, then the target footage can be found, but the complexity of the system increases

Engineering Contradiction:
Improvevideo extraction efficiencyVSAvoidsystem complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The system performs preliminary spatial intersection calculations to determine which camera viewsheds intersect with the target location before extracting videos. This preliminary step simplifies the overall system by pre-identifying relevant cameras, reducing the complexity of video extraction while improving productivity through targeted retrieval

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3091735B1Method and device for extracting surveillance record videos
Publication Date: 2019.05.01 HANGZHOU HIKVISION DIGITAL TECHNOLOGY CO LTD
  • EP3091735B1 patent drawingFigure 1
  • EP3091735B1 patent drawingFigure 2~3
  • EP3091735B1 patent drawingFigure 4

AI summary

The present invention relates to the field of video surveillance. Disclosed are a method and device for extracting surveillance record videos. In the present invention, the method for extracting surveillance record videos comprises the following steps: acquiring and storing lens viewsheds of cameras and an irradiation time period corresponding to each lens viewshed; extracting lens viewsheds corresponding to irradiation time periods that have intersection relationships with a query time period; calculating intersection relationships between the extracted lens viewsheds and a target location; obtaining a set of cameras corresponding to lens viewsheds that have intersection relationships with the target location; and extracting videos captured by the cameras according to irradiation time periods of the cameras in the camera set. Cameras relevant to the target can be found through intersection calculation performed on a designated target location in a designated time period of the user and selected camera viewsheds, so as to directly extract from the relevant cameras videos meeting conditions and useful for the practical application, thereby reducing the labor and time consumed in manual checking of video records.