Tracking System with Adaptive Resolution Regions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing tracking systems face delays and decreased accuracy due to the high processing load of high-resolution image analysis, especially when tracking fast-moving targets, as they require generating tracking image data and specifying target positions in series, leading to increased processing steps and potential delays.
Innovation Solution
A tracking system that generates tracking image data with high and low-resolution regions based on target size or speed, allowing the tracking image generation and calculation processes to be performed in parallel, thereby reducing processing time and maintaining accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high-resolution image data is used for tracking target position specification, then tracking accuracy is improved, but processing load increases causing delays and reduced frame rate
Solution Approach 1:
The patent applies local quality by generating tracking image data with different resolution characteristics in different regions. Specifically, the tracking target region maintains high resolution for accurate position specification, while other regions have reduced resolution to decrease overall processing load. This allows the system to achieve both high tracking accuracy and improved processing efficiency by optimizing resolution distribution according to actual needs.
2Reliability
If tracking image data generation is performed before target position specification, then proper image processing is ensured, but the number of process steps increases causing delays
Solution Approach 1:
The patent implements preliminary action by generating the tracking image data with optimized resolution distribution before the target position specification process begins. By pre-processing the image data to highlight the tracking target region with appropriate resolution characteristics, the system ensures that subsequent position specification can proceed efficiently without requiring additional processing steps, thereby reducing overall processing delay while maintaining reliability.
Data Source
AI summary
In a preferred aspect of the present invention, at least one of a camera-side controller or a terminal-side controller performs a tracking image generation process (P1) of generating tracking image data from captured image data. Further, at least one of the camera-side controller or the terminal-side controller performs a tracking calculation process (P2) of acquiring target information on the basis of the tracking image data. The tracking image data includes a high-resolution region with relatively higher resolution and a low-resolution region with relatively lower resolution. In the tracking image generation process (P1), basic tracking information including information on at least one of a size or a speed of the tracking target is acquired, and a high-resolution region and a low-resolution region are determined on the basis of the basic tracking information. The tracking image generation process (P1) and the tracking calculation process (P2) are performed in parallel.