Temporally Adaptive Image Processing for Target Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image analysis systems for target detection and tracking in dynamic scenes use fixed frame data, which limits their ability to adapt to varying target velocities and clutter, leading to suboptimal detection performance.
Innovation Solution
A method that computes background relative velocity and hypothesizes target relative velocity to calculate an optimal revisit time, adjusting the frame capture or processing rate based on the difference between the two velocities, using spatial-temporal matched filters to enhance target detection by optimizing frame separation times and clutter suppression.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If fixed frame data is used for target detection, then system complexity is reduced, but target detection performance deteriorates due to inability to adapt to varying target velocities and clutter
Solution Approach 1:
The patent applies dynamics by making the frame capture rate and processing rate adjustable rather than fixed. The system dynamically adapts the frame rate based on computed background relative velocity and hypothesized target relative velocity to optimize detection performance for varying target velocities and clutter conditions
Solution Approach 2:
The patent changes the parameter of frame capture rate based on velocity differences. By computing background relative velocity and hypothesized target relative velocity, the system adjusts the frame capture rate parameter to optimize the revisit time for detecting targets with different velocities
2Reliability
If frame capture rate is increased to improve detection probability, then detection performance improves, but data processing load and false alarms increase
Solution Approach 1:
The patent uses feedback by computing background relative velocity from image data and using this information to adjust the frame capture rate. This feedback loop allows the system to adapt to actual scene conditions, increasing detection probability while managing false alarms through intelligent rate adjustment
Solution Approach 2:
The patent applies periodic action through the concept of revisit time, which is computed based on velocity differences. The system periodically captures frames at optimized intervals, allowing sufficient time for target-background separation while avoiding excessive sampling that would increase false alarms
3Reliability
If frame processing rate is adjusted based on target revisit time, then detection performance improves, but system complexity increases
Solution Approach 1:
The patent applies preliminary action by computing background relative velocity and target revisit time before adjusting the frame processing rate. This advance computation allows the system to optimize processing parameters based on scene characteristics, improving detection performance while managing complexity through pre-planned adjustments
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method of image processing for finding a target in a scene includes receiving a series of images from a sensor and computing a background relative velocity in the series of images. The method further includes estimating a hypothesized target relative velocity of the target in the series of images. The method further includes computing a target revisit time based on a difference of the target relative velocity and the background relative velocity. The method further includes adjusting one or more of a frame capture rate or a frame processing rate based on the target revisit time.