Predicted ROI Framing for Low-Load Moving Object Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The ROI function for tracking a subject in image processing has a high processing load due to the need to detect the subject in each frame image.
Innovation Solution
A signal processing device that includes a position detection unit to detect the target object's position, a position estimation unit to predict its position in subsequent frames based on movement speed, and a frame setting unit to set a cut-out frame accordingly, reducing the need for extensive image processing during tracking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If image processing is performed to detect the target object in each frame image for tracking, then the target object can be tracked accurately, but the processing load becomes large
Solution Approach 1:
The patent applies preliminary action by detecting the target object's position and movement speed in advance during initial frames, then using this pre-acquired information to predict the object's position in subsequent frames. This allows the system to switch from intensive image processing to lightweight prediction-based tracking, reducing processing load while maintaining accuracy.
Solution Approach 2:
The patent replaces the mechanical image processing system with a prediction-based system. Instead of continuously processing images to detect the object, the system uses mathematical prediction models (position estimation unit) to forecast object location, substituting computational imaging with computational prediction to reduce processing requirements.
2Measurement precision
If image processing is performed continuously to track the target object, then the object position can be updated accurately, but power consumption increases
Solution Approach 1:
The patent applies periodic action by performing intensive image processing only at specific intervals (initial detection phase) and using prediction models during continuous tracking phases. This periodic switching between processing modes maintains position accuracy while significantly reducing overall power consumption compared to continuous processing.
3Productivity
If the cut-out frame is set based on predicted position, then processing load is reduced, but the complexity of position estimation increases
Solution Approach 1:
The patent applies parameter changes by using the target object's movement speed as a key parameter to simplify position estimation. Instead of complex image analysis, the system changes the estimation approach to rely on motion parameters (position and speed) derived from initial frames, which reduces computational complexity while enabling efficient frame setting.
Data Source
AI summary
A signal processing device includes: a position detection unit that, on the basis of a captured image by an imaging unit that captures a moving target object, detects a position of the target object in the captured image as an object position; a position estimation unit that estimates a position of the target object in the captured image of next and subsequent frames as a predicted position on the basis of the detected object position and a movement speed of the target object; and a frame setting unit that sets a cut-out frame in the captured image of the next and subsequent frames according to the predicted position.