Digital Focal Plane Array High Dynamic Range Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional high dynamic range imaging techniques face challenges such as long image acquisition periods, information loss due to counter rollover, and complex processing, which limit the ability to capture scenes with wide dynamic ranges and achieve full-rate video imaging.
Innovation Solution
A digital focal plane array system that uses an m-bit counter and a processor to detect photons during multiple integration periods with different gains, generating a digital representation of the scene by estimating photon flux and compensating for counter rollover through modulo operations, thereby extending the dynamic range beyond the intrinsic counter capacity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple images are acquired at different exposures to achieve high dynamic range, then dynamic range is improved, but image acquisition time increases and scene motion susceptibility increases
Solution Approach 1:
The imaging system divides the dynamic range into multiple segments by using multiple exposures with different integration times. Each exposure captures a specific dynamic range segment, and these segments are stitched together to form the complete high dynamic range image. This segmentation allows the system to capture both dim and bright objects in the same scene without requiring a single extremely long exposure time.
Solution Approach 2:
The system employs periodic switching between different integration times (exposure durations) to capture images at different dynamic range levels. By alternating between short and long integration times in a periodic manner, the system can efficiently acquire multiple dynamic range segments without requiring continuous long exposure, thus reducing total acquisition time while maintaining high dynamic range capability.
2Measurement precision
If long integration periods are used to capture dim objects, then signal quality is improved, but counter rollover information loss increases
Solution Approach 1:
The system uses feedback from the counter rollover detection mechanism to adjust the integration time selection. When counter rollover is detected or anticipated, the system switches to shorter integration times to prevent information loss. This feedback loop ensures that the system maintains optimal signal quality while avoiding counter rollover information loss by continuously monitoring and adapting the exposure parameters.
Solution Approach 2:
The system dynamically changes the integration time parameter based on the signal strength and counter status. For dim objects, longer integration times are used to improve signal quality. When the counter approaches its rollover limit or when bright objects are detected, the system switches to shorter integration times to prevent information loss. This parameter adaptation allows the system to optimize between signal quality and information preservation.
3Adaptability or versatility
If multiple images are stitched together to achieve high dynamic range, then dynamic range is improved, but processing complexity increases
Solution Approach 1:
The system merges multiple low dynamic range images into a single high dynamic range image by combining their information. The processing algorithm integrates the dynamic range segments from different exposures, aligning them spatially and temporally, and stitching them together to form the final HDR image. This merging process consolidates the complexity into a unified output, making the system manageable while achieving extended dynamic range.
4Device complexity
If fixed well depth is used in analog FPAs, then device simplicity is maintained, but dynamic range capability is limited
Solution Approach 1:
The system transitions from a static fixed well depth to a dynamic adjustable integration time mechanism. By varying the integration time parameter, the system can adapt the effective well depth to match the dynamic range requirements of different scenes. This dynamic adjustment allows the same hardware to achieve extended dynamic range capability without fundamentally changing the device architecture, thus maintaining relative simplicity while improving versatility.
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
Enables real-time generation of high dynamic range images with an effective bit depth greater than the m-bit counter alone, reducing information loss and supporting video frame rates of 60 Hz or more by effectively managing counter rollover and estimating true signal values.
Implementation Method 1
the detector element detects incident photons during a first integration period
Data Source
AI summary
When imaging bright objects, a conventional detector array can saturate, making it difficult to produce an image with a dynamic range that equals the scene's dynamic range. Conversely, a digital focal plane array (DFPA) with one or more m-bit counters can produce an image whose dynamic range is greater than the native dynamic range. In one example, the DFPA acquires a first image over a relatively brief integration period at a relatively low gain setting. The DFPA then acquires a second image over longer integration period and/or a higher gain setting. During this second integration period, counters may roll over, possibly several times, to capture a residue modulus 2m of the number of counts (as opposed to the actual number of counts). A processor in or coupled to the DFPA generates a high-dynamic range image based on the first image and the residues modulus 2m.