Dynamic High Sampling Rate for Region of Interest in Imaging Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing imaging systems face challenges in balancing computation resources and time while processing high-resolution image frames, particularly when only a selected region of interest (ROI) is processed, which may miss important details and increase computational load.
Innovation Solution
A system and method that enhance the sampling rate for a selected region of interest (SROI) by defining SROI in image frames, acquiring partial datasets during residual time within image frame handling cycles, and using a tracking module to detect rails and define safety zones, allowing for increased resolution without altering frame rate or computational load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high resolution computation is applied to the entire image frame, then measurement precision is improved, but computation resources increase and computation time increases
Solution Approach 1:
The image frame is segmented into multiple regions of interest (ROIs) based on detected features such as rails, obstacles, or areas with motion. Only these segmented ROIs are processed at high resolution, while the rest of the frame is processed at lower resolution or skipped, thereby maintaining detection precision for critical areas while reducing overall computation resources and time.
Solution Approach 2:
Different regions of the image frame are assigned different processing qualities. High-resolution computation is applied locally to ROIs that contain important features or potential obstacles, while other regions receive lower-resolution processing. This local quality differentiation maintains measurement precision where needed while improving overall productivity.
2Productivity
If high resolution computation is applied to a selected region of interest, then computation time is reduced, but important image details may be missed
Solution Approach 1:
A low-resolution pre-processing stage is performed on the entire image frame before selecting ROIs for high-resolution computation. This preliminary action identifies potential regions of interest containing important features or obstacles, ensuring that no critical data is missed before applying the computationally intensive high-resolution processing only to those identified areas.
Solution Approach 2:
The system uses feedback from initial low-resolution analysis and ongoing detection results to dynamically adjust and expand ROIs. If important features or obstacles are detected at the boundaries of current ROIs, the ROI selection is updated to include these areas, ensuring that no significant information is lost while maintaining computational efficiency.
3Reliability
If the frame rate is increased to improve obstacle detection, then probability of detection is improved, but computation resources increase
Solution Approach 1:
Instead of processing the entire frame at high frame rates, the system segments the frame into ROIs and applies high-frame-rate processing only to these segmented regions. This maintains a high probability of detecting obstacles in critical areas while reducing the overall computation load by excluding processed regions from further high-frequency analysis.
Data Source
Figure 1A
Figure 1B
Figure 2
AI summary
Generally, a system including an imaging device and a processing unit is disclosed. The imaging device may be configured to acquire a plurality of datasets of corresponding plurality of image frames by performing corresponding plurality of image frame handling cycles. The processing unit may be configured to define a special region of interest (SROI) in each of at least some of the plurality of the image frames acquired by the imaging device, based on the datasets of the respective image frames. The imaging device may be further configured to acquire at least one partial dataset of the SROI, during each of at least some of the plurality of image frame handling cycles and within a residual time between an end of an image frame acquiring time and an end of the respective image frame handling cycle.