Oversampled Imaging Sensor Pixel Correlation for Dim Target Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current imaging sensors face limitations in detecting dim targets at longer distances without creating false positives or false alarms, particularly due to undersampling and the challenges of photon arrival uncertainty, which leads to aliasing, noise, and computational complexity in super-resolution reconstruction techniques.
Innovation Solution
The implementation of an oversampled imaging system with small pixels that correlate spatial and temporal data, using superpixels to improve sensitivity and reduce noise, allowing for real-time detection of dim targets while suppressing false alarms through spatial temporal filtering and pixel correlation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If undersampled imaging sensors are used, then device complexity is reduced, but measurement precision deteriorates due to aliasing and noise
Solution Approach 1:
The imaging sensor divides the detection area into multiple pixels with different sensitivity characteristics. Each pixel segment responds differently to photon arrivals, allowing the system to capture spatial-temporal patterns that improve target detection precision while maintaining manageable device complexity through modular pixel design.
Solution Approach 2:
The patent introduces temporal dimension by detecting the time of photon arrival at each pixel. This transforms a 2D spatial sampling problem into a 3D space-time problem, where the additional temporal information allows super-resolution reconstruction and improved target detection precision without requiring higher spatial sampling density.
2Measurement precision
If super-resolution reconstruction techniques are used, then measurement precision is improved, but device complexity increases due to computational requirements
Solution Approach 1:
The imaging sensor performs preliminary spatial-temporal sampling during photon detection, capturing time-stamped arrival data at multiple pixel locations. This preliminary action during the physical detection phase reduces the computational burden of subsequent super-resolution reconstruction by providing pre-processed spatial-temporal patterns that are closer to the final high-resolution result.
Solution Approach 2:
The pixel array itself performs correlation operations by comparing photon arrival patterns across multiple pixels and time samples. This self-service approach distributes the computational workload across the sensor hardware rather than relying entirely on external processing systems, reducing overall device complexity while maintaining measurement precision.
3Measurement precision
If pixel size is reduced to increase sampling density, then measurement precision is improved, but manufacturing precision becomes more difficult
Solution Approach 1:
The patent uses pixels that are larger than the theoretical minimum required for critical sampling of the optical blur. This excessive spatial sampling, combined with temporal sampling of photon arrivals, achieves super-resolution without requiring extremely small pixel dimensions, thereby avoiding the manufacturing precision challenges associated with sub-micron pixel fabrication.
Solution Approach 2:
The system changes the detection parameter from purely spatial to include temporal dimension. By detecting when photons arrive at each pixel rather than only where they land, the system achieves improved measurement precision with larger, more manufacturable pixels. This parameter change from spatial-only to space-time sampling relaxes the manufacturing precision requirements.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enhances the detection of dim targets at longer ranges with improved resolution and acuity, reducing false alarms and computational complexity, and enabling better performance in challenging conditions like motion and atmospheric disturbances.
Implementation Method 1
Each pixel converts the incoming photons focused onto a focal plane into electrons and holes, known as photocharge
Data Source
AI summary
An over sampled image sensor in which the pixel size is small enough to provide spatial oversampling of the minimum blur size of the sensor optics is disclosed. Image processing to detect targets below the typical limit of 6× the temporal noise floor is also disclosed. The apparatus is useful in detecting dimmer targets and targets at a longer range from the sensors. The inventions exploits signal processing, which allows spatial temporal filtering of the superpixel image in such manner that the Noise Equivalent Power is reduced by a means of Superpixel Filtering and Pooling, which increases the sensitivity far beyond a non-oversampled imager. Overall visual acuity is improved due to the ability to detect dimmer targets, provide better resolution of low intensity targets, and improvements in false alarm rejection.