Delta Image Sensor Pixel Storage Using Multi-Ramp ADC Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image sensors face challenges in efficiently storing and comparing previous illumination levels to detect changes, leading to issues like gradual storage degradation, sensitivity to mismatch, and reduced flexibility in light-to-electric conversion, particularly in dynamic vision sensors.
Innovation Solution
A delta image sensor design that combines digital conversion and storage in pixel circuits, using different ramps for analogue-to-digital conversion, allowing local evaluation of differences over time and location, and preparation and propagation of event information, enabling efficient detection of changed illumination levels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If analogue storage is used in dynamic vision sensors, then compact realisation is achieved, but gradual storage degradation and sensitivity to mismatch occur
Solution Approach 1:
The patent replaces the analogue storage mechanism (capacitor-based) with a digital storage mechanism (register-based). The analogue-to-digital conversion is performed using a ramp comparator circuit that converts the stored analogue value to a digital code, which is then stored in a digital register. This substitution eliminates the gradual degradation inherent in analogue storage while maintaining compact pixel circuit area through shared circuitry.
Solution Approach 2:
The patent changes the storage parameter from analogue voltage levels to digital binary codes. By converting the stored illumination level from an analogue quantity subject to drift and degradation into a discrete digital representation, the system achieves stable, non-degrading storage. The digital value remains constant over time and is immune to the hot-pixel effects that plague analogue storage.
2Reliability
If digital storage is used in pixel circuits, then storage fidelity is improved, but pixel circuit area increases
Solution Approach 1:
The patent merges multiple functions into shared circuits that serve multiple pixels. The ramp generator, comparator, and ADC circuitry are implemented as shared resources across pixel arrays, with each pixel contributing to the overall conversion process. This merging approach allows digital storage per pixel while keeping the total circuit area manageable through resource sharing.
Solution Approach 2:
The patent implements universal circuits that perform multiple functions: the ramp comparator serves both as an ADC for conversion and as a storage mechanism through its digital output. The same circuit infrastructure supports both analogue-to-digital conversion and digital storage operations, eliminating the need for separate dedicated circuits for each function and thereby reducing overall area.
3Productivity
If one ADC per pixel is implemented, then processing capability is improved, but device complexity and cost increase
Solution Approach 1:
The patent segments the ADC functionality into modular components that can be shared across multiple pixels. Rather than implementing a complete ADC in each pixel, the system divides the conversion process into stages with shared resources, reducing per-pixel complexity while maintaining overall processing capability.
Solution Approach 2:
The patent implements dynamic resource allocation where ADC circuits are activated and deactivated based on processing needs. The system can dynamically share ADC resources across different pixel groups or time periods, allowing high processing capability when needed while reducing complexity and power consumption during normal operation.
4Area of stationary object
If analogue-to-digital conversion is performed globally, then circuit area is reduced, but signal integrity and precision are degraded
Solution Approach 1:
The patent transitions from a purely spatial arrangement where each pixel has dedicated circuits to a hybrid approach that uses time-multiplexed sharing. By adding the time dimension to resource allocation, the system achieves both area reduction through sharing and precision maintenance through dedicated conversion periods for each pixel group.
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 results in an area- and cost-optimized sensor with high resolution, reduced power consumption, and improved fidelity in storage and processing, enabling efficient event-based dynamic vision sensing with reduced computational complexity and noise suppression.
Implementation Method 1
a sensor circuit comprising a photodetector configured to generate a sensor signal, VSIG, depending on a light signal illuminating the photodetector
Data Source
AI summary
A delta image sensor comprising a plurality of acquisition circuits corresponding to at least one pixel. Each acquisition circuit includes at least one sensor circuit comprising a photosensor to generate a sensor signal, VSIG, depending on a light signal illuminating the photosensor; at least one single slope analogue to digital conversion, A/D, circuit configured to convert a current VSIG to a digital signal, wherein the A/D circuit (12) is configured to use one of a plurality of ramps for the conversion; at least one digital storage circuit configured to store a representation of at least one digital signal corresponding to a previous VSIG; at least one digital comparison circuit configured to compare the level of the stored representation with the current VSIG to detect whether a changed level is present; and at least one digital output circuit configured to generate an event output, in response to the changed level.