Autofocus Engine Sub-Image Edge Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autofocus systems face challenges in air-to-ground applications, particularly in focusing entire images quickly and in low light conditions, as the range to the subject and background can be similar, making it difficult to distinguish and focus on the subject while maintaining a clear background, and existing methods struggle with edge detection in high contrast areas.
Innovation Solution
The method involves dividing edge detected images into sub-images, calculating a normalized edge detection strength by forming a ratio of maximum and minimum pixel values, and comparing this to a threshold to determine the presence of edges, thereby calculating an autofocus metric that helps reposition the focus cell for optimal image focus.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional edge detection methods are used in air-to-ground imaging, then the focus can be determined for ground-based subjects, but the entire image cannot be focused uniformly because the background is at a different range than the subject
Solution Approach 1:
The patent applies local quality by creating a focus mask that differentiates between subject regions and background regions. The mask assigns different focus weights to different areas of the image, allowing the autofocus algorithm to optimize focus for the subject area while treating the background area differently. This resolves the contradiction by making the focus determination locally adaptive rather than uniformly applied across the entire image.
2Measurement precision
If autofocus systems focus on the entire image uniformly, then all areas are in focus, but in air-to-ground applications this prevents selective focusing on specific subjects against the background
Solution Approach 1:
The focus mask creates local quality differences by assigning high focus weights to subject regions and low or zero weights to background regions. This allows the system to achieve selective subject focus while maintaining a controlled approach to background rendering, resolving the contradiction between uniform focus and selective focus capabilities.
3Measurement precision
If focus sweep is performed across the entire image, then comprehensive focus data is collected, but the processing time increases significantly
Solution Approach 1:
The patent segments the image into subject regions and background regions using the focus mask. During focus sweep processing, the algorithm processes only the masked subject regions with high priority, reducing the computational burden compared to processing the entire image. This segmentation approach maintains focus metric accuracy for the subject while significantly reducing processing time.
4Measurement precision
If traditional autofocus metrics are used, then focus can be determined in well-lit conditions, but edge detection fails in high contrast areas and low light conditions
Solution Approach 1:
The patent changes the parameters used for edge detection by incorporating gradient-based metrics and adaptive thresholding that work effectively in low light and high contrast conditions. The focus mask also changes the spatial distribution of measurement parameters, concentrating measurements on subject regions where edges are most relevant. This resolves the contradiction by making edge detection robust across varying illumination conditions.
Data Source
Figure 1
Figure 2A
Figure 2B
AI summary
An autofocus metric approach for focusing video images, automatically, based on images taken during a focus sweep in which a focus cell is repositioned for each of the images is provided. The approach includes, given an edge detected image from the focus sweep and an associated focus cell position in the focus sweep, an autofocus engine dividing the edge detected image into sub-images. For each sub-image, the autofocus engine calculates a normalized edge detection strength and compares it to a threshold. Based on the comparison, the autofocus engine determines whether an edge is present in the sub-image. Based on the determinations of edges in the sub-images, the autofocus engine calculates an autofocus metric associated with the given focus cell position. The autofocus engine provides the autofocus metric together with autofocus metrics associated with other focus cell positions to focus the video images.