Anomalous Pixel Detection via Spatial Linearity Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing technologies are inefficient and unreliable in detecting anomalous pixels in imaging devices, which can lead to misleading representations of captured scenes due to hardware imperfections and manufacturing tolerances.
Innovation Solution
A system that includes a processing component to receive image frames, select kernels of pixels, determine linearity measurements, update spatial anomaly scores, and detect spatially anomalous pixels by comparing these scores to a threshold, effectively identifying and potentially correcting flickering or deviating pixels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If human detection is used to identify anomalous pixels, then detection accuracy can be maintained, but the process becomes impractical and cumbersome
Solution Approach 1:
The patent replaces the mechanical human visual inspection system with an automated computational image processing system. The system uses algorithms to calculate anomaly scores for each pixel by comparing it with neighboring pixels, automatically identifying defective pixels without human intervention. This substitution maintains detection accuracy while eliminating the operational burden of manual inspection.
2Extent of automation
If existing machine-based approaches are used for pixel anomaly detection, then automation is achieved, but detection reliability and efficiency remain unsatisfactory
Solution Approach 1:
The patent introduces a novel parameter-based detection approach by calculating anomaly scores for each pixel based on its deviation from neighboring pixels. The system computes a maximum anomaly score by comparing each pixel with its neighbors and identifies defective pixels when this score exceeds a threshold. This parameter-driven approach improves both automation efficiency and detection reliability compared to existing machine-based methods.
3Productivity
If simple pixel comparison methods are used, then processing speed is maintained, but detection precision deteriorates due to hardware imperfections and manufacturing tolerances
Solution Approach 1:
The patent applies local quality analysis by comparing each pixel with its immediate neighboring pixels rather than using global statistics. The anomaly score for each pixel is determined by its local deviation from neighbors, allowing the system to account for spatial variations and local patterns in the image. This local approach maintains processing speed while improving detection precision by considering the specific local context of each pixel.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Systems and methods are disclosed herein to detect pixels exhibiting anomalous behavior in captured image frames. In some examples, temporal anomalous behavior may be identified, such as flickering pixels exhibiting large magnitude changes in pixel values that vary rapidly from frame-to-frame. In some examples, spatial anomalous behavior may be identified, such as pixels exhibiting values that deviate from an expected linear response in comparison with other neighbor pixels.