Adaptive Video Processing Using Thermal Data for Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video surveillance methods face challenges in detecting objects at large distances while minimizing false detections, as objects far away appear small and are often mistaken for noise, leading to a trade-off between detection range and false detection reduction.
Innovation Solution
Modifying video processing algorithm parameters based on thermal information from a thermal camera to enhance sensitivity in areas of interest, such as human temperature ranges, while reducing sensitivity in other areas, allowing for longer detection ranges with reduced false detections.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If video processing algorithms use fixed parameters for object detection, then false detections are reduced through noise filtering, but detection sensitivity at large distances deteriorates
Solution Approach 1:
The patent applies dynamics by making the video processing parameters adaptive rather than fixed. The system dynamically adjusts detection parameters based on thermal information from a thermal camera, allowing the parameters to change in response to scene conditions. This enables the system to maintain high detection sensitivity at large distances when thermal signatures are present while still filtering noise effectively in other conditions.
Solution Approach 2:
The patent implements parameter changes by modifying video processing algorithm parameters based on thermal data. Specifically, the system changes detection thresholds and sensitivity parameters according to the thermal signatures detected in the scene, allowing optimal detection performance across varying conditions without being constrained by fixed parameter settings.
2Length of stationary object
If video processing parameters are made sensitive to detect small objects at large distances, then detection range is improved, but false detections increase due to noise
Solution Approach 1:
The patent uses thermal information as an intermediary to resolve the contradiction between detection range and false detection rate. The thermal camera provides intermediate data that helps distinguish actual objects from noise, allowing the system to extend detection range to small objects at large distances while maintaining reliability by using thermal signatures as a filtering mechanism.
Solution Approach 2:
The patent applies local quality by making detection sensitivity location-dependent based on thermal signatures. Rather than uniformly increasing sensitivity across the entire field of view, the system selectively enhances sensitivity in regions where thermal signatures indicate the presence of objects, thereby extending detection range locally without increasing false detections in other areas.
3Measurement precision
If a second camera is controlled to follow specific target objects with high precision, then object identification accuracy is improved, but the system can only detect one object at a time reducing productivity
Solution Approach 1:
The patent applies segmentation by dividing the scene into multiple regions of interest based on thermal signatures. Instead of controlling a second camera to follow one object at a time, the system segments the field of view into multiple areas where objects are detected, allowing parallel processing and monitoring of multiple targets simultaneously, thereby improving detection throughput while maintaining identification accuracy.
Solution Approach 2:
The patent implements universality by making the video processing system capable of handling multiple object types and scenarios through adaptive parameter adjustment. The system can simultaneously detect and track multiple objects with different thermal signatures, making it versatile for various surveillance scenarios rather than being limited to single-object detection.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables effective detection of objects and activity at large ranges with reduced false detections by tailoring video processing algorithm parameters to thermal content, improving detection accuracy and range without increasing false positives.
Implementation Method 1
thermal information in a first video sequence captured by a thermal camera
Data Source
Figure 1
Figure 2
Figure 3
AI summary
There is provided a method, a device and a system for modifying at least one parameter used by a video processing algorithm, such as a motion detection algorithm, an object detection algorithm, or an object tracking algorithm, for monitoring of a scene (102). The method comprises: receiving a first (1 05a) and a second (1 05b) video sequence of the scene (102), wherein the first video sequence (105a) is captured using a thermal camera (104a) such that the first video sequence (1 05a) comprises thermal information being indicative of temperatures in the scene (102), and applying the video processing algorithm to the second video sequence (1 05b), wherein at least one parameter used by the video processing algorithm is modified based on the thermal information comprised in the first video sequence (105a), such that said at least one parameter (309, 409) of the video processing algorithm is made temperature dependent, wherein the thermal information is used for determining in what regions of the second video sequence said at least one parameter is modified and how much it is modified in said regions.