Event-Based Image Sensor Pixel Circuit for Temporal Resolution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image sensors face limitations in temporal and spatial resolution due to frame-based acquisition, leading to high redundancy in data, increased complexity, cost, and power consumption, especially in dynamic scenes, which affects various vision applications.
Innovation Solution
An event-based image sensor with pixel circuits that detect changes in light intensity independently, using a current comparator and storage cells to selectively record and transmit only changes, eliminating temporal redundancy and enabling asynchronous data readout.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If frame-based acquisition is used to capture image data, then complete image information is acquired, but temporal redundancy increases and temporal resolution is limited
Solution Approach 1:
The patent extracts only the essential change information from each pixel (whether intensity changed and by how much) rather than transmitting complete frame data. This selective extraction eliminates temporal redundancy while preserving all meaningful visual information about scene dynamics.
Solution Approach 2:
The patent implements dynamic, event-driven data transmission where pixels asynchronously generate output only when changes occur, rather than following a fixed frame rate. This dynamic approach adapts to scene activity levels, providing high temporal resolution for changing regions while minimizing data transmission for static areas.
2Reliability
If full frames of image data are acquired and processed, then complete visual information is obtained, but data volume and processing complexity increase
Solution Approach 1:
The patent extracts only the essential change information from each pixel (whether intensity changed and by how much) rather than transmitting complete frame data. This selective extraction eliminates temporal redundancy while preserving all meaningful visual information about scene dynamics.
Solution Approach 2:
The patent segments the image processing task to the pixel level, where each pixel independently detects and encodes its own changes. This segmentation enables parallel processing and reduces the need for centralized processing of entire frames, thereby reducing overall data volume and processing complexity.
3Loss of time
If higher frame rates are used to improve temporal resolution, then scene dynamics are captured better, but power consumption and data transmission requirements surge
Solution Approach 1:
The patent implements dynamic, event-driven data transmission where pixels asynchronously generate output only when changes occur, rather than following a fixed frame rate. This dynamic approach adapts to scene activity levels, providing high temporal resolution for changing regions while minimizing data transmission for static areas.
Solution Approach 2:
The patent maintains continuous monitoring of light intensity at each pixel, ready to detect changes immediately, without the need for periodic frame-based sampling. This continuous action ensures no temporal information is lost while avoiding the energy waste of transmitting data at fixed intervals when no changes occur.
4Quantity of substance
If frame differencing is applied to reduce redundancy, then data volume decreases, but temporal resolution remains limited by frame rate
Solution Approach 1:
The patent implements dynamic, event-driven data transmission where pixels asynchronously generate output only when changes occur, rather than following a fixed frame rate. This dynamic approach adapts to scene activity levels, providing high temporal resolution for changing regions while minimizing data transmission for static areas.
Solution Approach 2:
The patent performs change detection at the pixel level before data transmission, using local intensity comparison logic within each pixel circuit. This preliminary action filters out redundant data early in the sensing process, reducing the burden on subsequent processing stages while maintaining high temporal resolution.
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 reduces data volume significantly while maintaining high information content, improving temporal resolution, reducing system power and complexity, and enhancing imaging quality by eliminating spatial divergence between change detection and exposure measurement.
Implementation Method 1
a photoreceptor circuit (10) configured for delivering on an output a photoreceptor current derived from light
Data Source
AI summary
An event-based image sensor is provided that includes a plurality of pixel circuits. Each pixel circuit includes a photoreceptor circuit with a light-sensitive element configured for delivering at an output a photoreceptor current, an analog bus, and a bank of current memory cells connected to the analog bus. An output of the photoreceptor circuit is selectively connected to the analog bus. Each current memory cell is adapted to store an electric current flowing in the analog bus and to deliver an electric current stored within the current memory cell. Each pixel circuit may further include a current comparator including a current sign detector connected to the analog bus and configured for comparing a value of a compared electric current resulting from a difference between the photoreceptor current and a previous photoreceptor current stored in a current memory cell. At least one storage cell may also be provided that has a state conditioned by the output of the current sign detector.