Solid-State Image Sensor Selective AD Conversion for Power Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Asynchronous solid-state image sensors (DVS) generate two-bit data per pixel, which is insufficient for high-quality image recognition requiring three-bit or more data, and adding circuits to enhance image quality increases power consumption.
Innovation Solution
A solid-state image sensor with pixels generating analog signals through photoelectric conversion, accompanied by an analog-digital conversion unit that converts signals from pixels with light amount changes outside a predetermined range into digital signals, using selection units and multiple analog-digital converters to optimize data conversion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If asynchronous solid-state image sensor (DVS) generates two-bit data per pixel to detect address events, then detection speed is improved, but image quality is insufficient for high-quality image recognition requiring three-bit or more data
Solution Approach 1:
The patent applies partial action by performing analog-to-digital conversion only for pixels where address events are detected, rather than converting all pixel signals. This selective conversion approach maintains high detection speed while providing sufficient image quality for recognition tasks, avoiding the excessive action of full-frame conversion that would consume unnecessary power and time.
Solution Approach 2:
The patent segments the image processing pipeline into two distinct paths: one for rapid address event detection (generating two-bit data) and another for selective high-quality conversion (generating three-bit or more data) for specific pixels where events occur. This segmentation allows the system to maintain both high detection speed and adequate image quality by applying different processing levels where needed.
2Measurement precision
If circuits are added to enhance image quality to three-bit or more data while detecting address events, then image quality is improved, but power consumption increases
Solution Approach 1:
The patent implements partial action by activating the analog-to-digital conversion circuitry only when and where address events are detected, rather than continuously converting all pixel data. This selective operation dramatically reduces power consumption compared to full-frame conversion while still providing high-quality three-bit or more data for pixels that contribute to the captured image.
Solution Approach 2:
The patent extracts only the necessary pixel signals that contain address events from the full pixel array and routes them to the analog-to-digital conversion unit. By taking out only the relevant data for conversion rather than processing the entire frame, the system achieves high image quality for detected events while minimizing power consumption by leaving the rest of the pixel data in lower-resolution format.
3Measurement precision
If all pixel signals are converted to digital signals for high-quality image capture, then image quality is improved, but processing time and power consumption increase
Solution Approach 1:
The patent applies partial action by converting only the subset of pixel signals where address events are detected, rather than converting the entire pixel array. This selective conversion approach maintains adequate processing time by reducing the total number of conversion operations while still providing high-quality digital data for the pixels that contribute to the final captured image.
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 allows for high-quality image capture while detecting address events with reduced power consumption by selectively converting only necessary pixel signals, minimizing AD conversion and enhancing image processing efficiency.
Implementation Method 1
a plurality of pixels each configured to generate an analog signal by photoelectric conversion
Data Source
AI summary
In a solid-state image sensor that detects presence or absence of an address event, power consumption when capturing an image is reduced.The solid-state image sensor includes a plurality of pixels and an analog-digital conversion unit. In the solid-state image sensor, each of the plurality of pixels generates an analog signal by photoelectric conversion. Moreover, in the solid-state image sensor, the analog-digital conversion unit converts the analog signal of a pixel into a digital signal, the pixel having an amount of change in incident light amount that falls outside a predetermined range, of the plurality of pixels.