PTZ Camera Control via Dynamic Object Detection and Zoom Adjustment

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing PTZ camera control methods require frequent calibration of observation positions to adapt to changes in the camera's surroundings, leading to increased maintenance costs and often result in either excessive detail or loss of overall view.

Innovation Solution

A method for dynamically establishing observation positions and adjusting zoom levels using a learning phase that detects objects of interest, optimizing fields of view, and storing settings for automated monitoring, allowing for adaptive camera control without manual calibration.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If predetermined observation positions are used for PTZ camera control, then the camera can systematically survey the scene, but regular calibration is required to adapt to changes in surroundings, increasing maintenance costs

Engineering Contradiction:
Improvesurveillance efficiencyVSAvoidmaintenance requirement
Core Design Contradiction:
ProductivityVSEase of operation

Solution Approach 1:

The system performs automatic calibration by detecting objects of interest and autonomously determining optimal observation positions, eliminating the need for manual intervention and regular maintenance calibration

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system performs a learning phase before normal operation to pre-establish observation positions based on detected objects of interest, so that during normal operation no further calibration is needed

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If high zoom level is used to capture details, then object detection precision improves, but the overall view of the scene is lost

Engineering Contradiction:
Improveobject detection precisionVSAvoidoverall scene view
Core Design Contradiction:
Measurement precisionVSLoss of information

Solution Approach 1:

The zoom level is dynamically adjusted based on the detected object's characteristics and distance, allowing the system to automatically select between detailed views and overall scene views as needed

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

Different parts of the scene are captured at different zoom levels - high zoom for specific objects of interest and low zoom for overall scene context, with both views available in the surveillance system

Inventive Principle:
Principle #3Local quality

3Adaptability or versatility

If manual calibration of observation positions is performed, then the camera adapts to changes in surroundings, but maintenance costs and time consumption increase

Engineering Contradiction:
Improveadaptation to surroundingsVSAvoidcalibration time
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The camera system automatically adapts to changes in surroundings by detecting objects of interest and autonomously recalibrating observation positions without requiring manual intervention

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system continuously monitors the scene for objects of interest and uses this feedback to automatically adjust and recalibrate observation positions, enabling real-time adaptation to environmental changes

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3367353B1Control method of a ptz camera, associated computer program product and control device
Publication Date: 2023.03.22 THALES SA
  • EP3367353B1 patent drawingFigure 1~2
  • EP3367353B1 patent drawingFigure 3

AI summary

The present invention relates to a method for controlling a PTZ camera comprising a camera capture device adjustable to a plurality of observation positions, each observation position defining a field of view of the camera. The method comprises a monitoring phase (PS) including the steps of adjusting the camera capture device to one of the observation positions and capturing images (180) from the field of view corresponding to the observation position according to which the camera capture device is adjusted. The method is characterized in that it further comprises a learning phase (PA) implemented at least once before the monitoring phase (PS) and comprising the dynamic establishment of a list of settings based on objects of interest present in different fields of view of the camera.