High Dynamic Range Sensor with Locally Selectable Integration Times
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current night vision systems are limited by a dynamic range of about 16 bits, which is insufficient for capturing the wide range of light conditions encountered in urban environments, where light intensity can vary by 6 or 7 orders of magnitude, leading to image degradation and the need for multiple scans of a scene.
Innovation Solution
A high dynamic range sensor assembly with locally selectable integration times and miniature programmed microcore processors that adjust integration times based on sensed light intensity, allowing for a single scan with extended dynamic range and improved image capture.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single user adjustable exposure control is used to set a nominal range for present light conditions, then the dynamic range is limited to about 16 bits of sample resolution, but the system cannot automatically capture a wider range of night vision scene information including dark areas within relatively bright scenes
Solution Approach 1:
The sensor array is divided into multiple independently controllable pixel sets, each with its own integration time control. This segmentation allows different regions of the scene to be captured with different integration times, enabling simultaneous capture of both bright and dark areas within the same scene, thus achieving automatic wide dynamic range capture without manual intervention.
Solution Approach 2:
The system dynamically adjusts the integration time for each pixel set based on the local light intensity conditions. The processor automatically controls the integration time of each pixel set to optimize the capture of scene information, transitioning from static single exposure control to dynamic adaptive multi-exposure control, thereby achieving 20 bits or more of effective dynamic range.
2Measurement precision
If the sensor is scanned twice per field with different integration times, then the dynamic range is extended, but the frame rate is reduced and image degradation occurs due to scene changes between scans
Solution Approach 1:
Instead of performing sequential scans, the sensor array is segmented into multiple pixel sets that can operate simultaneously with different integration times. This parallel operation maintains the original frame rate while achieving extended dynamic range, eliminating the image degradation caused by scene changes between sequential scans.
Solution Approach 2:
The patent adds a spatial dimension to the integration time control by organizing pixels into multiple independently controllable sets across the sensor array. This allows different integration times to be applied simultaneously at different spatial locations, transforming the problem from temporal multiplexing (sequential scans) to spatial parallelism, thereby maintaining frame rate while extending dynamic range.
3Measurement precision
If multiple pixel sets with independently controllable integration times are implemented, then the dynamic range is extended to 20 bits or more, but the device complexity increases
Solution Approach 1:
Multiple pixel sets with independently controllable integration times are merged into a single sensor array structure. The shared readout circuitry and integrated processor control multiple pixel sets simultaneously, reducing the overall complexity compared to having separate sensors for each integration time. This merging approach achieves extended dynamic range while maintaining a compact unified sensor design.
Solution Approach 2:
The sensor assembly is designed with universal components that serve multiple functions. The same sensor array structure and readout circuitry are used across multiple pixel sets, each capable of operating with different integration times. This multi-functionality reduces the need for separate dedicated components for each integration time setting, thereby controlling device complexity while achieving extended dynamic range.
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
The solution enables the capture of images with light intensity conditions varying by 20 bits or more, overcoming image degradation and achieving significantly extended dynamic range without the need for multiple scans, thus enhancing the performance of night vision systems and other wide dynamic range optical imaging applications.
Implementation Method 1
A typical light sensor is a P-N junction that generates photocurrent in proportion to the intensity of the light that impinges on the P-N junction
Data Source
AI summary
A high dynamic range sensor assembly includes a plurality of sensing sets that are organized into a sensing array. Each of the sensing sets includes a set of sensing elements for sensing physical phenomena. Each set of sensing elements has a locally selectable integration time. An analog-to-digital (A/D) converter operatively connected to the set of sensing elements acquires and converts an analog signal from each of the sensing elements into a digital signal. A processor operatively connected to the A/D converter and to the set of sensing elements manages the selectable integration time for the set of sensing elements and analyzes the digital signals from each of the sensing elements in the set of sensing elements. The digital signals from each of the sensing elements are measured by the processor and an integration scaling factor for the set of sensing elements is computed and controlled by the processor to adjust the integration time. The integration scaling factor for the set of sensing elements is mathematically combined with a value of the digital signal from the A/D converter to form a larger data word than what is generated by the A/D converter. The larger data word is utilized to represent a magnitude of each of the sensing elements. If a substantial number of A/D values have saturated, the integration time is decreased; and, if a substantial number of A/D values are below a predetermined threshold, the integration time is increased.