Event-Based Camera Feature Detection Hardware
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional computer vision systems in mobile devices require significant processing resources and power for detecting features and objects, leading to higher power consumption and reduced battery life due to the need for extensive processing and data transfer from frame-based cameras.
Innovation Solution
Implementing an event-based camera system with dedicated computer vision computation hardware that performs localized feature computation and event detection within the sensor array, reducing the need for extensive processing by the application processor and minimizing data transfer.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional frame-based camera processing is used, then complete image data is captured and processed, but processing resources and power consumption increase significantly
Solution Approach 1:
The patent segments the image processing task by implementing event detection logic within the sensor array that operates independently on individual pixels or small groups of pixels. Each sensor element detects changes in intensity and generates events locally, rather than processing complete frames centrally. This segmentation reduces the data volume requiring high-power processing while maintaining feature detection capability through distributed, low-power event generation.
2Productivity
If extensive processing is performed on the application processor, then feature detection is achieved, but processing load and power consumption increase
Solution Approach 1:
The patent extracts the event detection function from the application processor and implements it directly within the sensor array hardware. By taking out this specific processing task from the central processor and placing it at the sensor level, the system achieves feature detection capability while significantly reducing the processing load on the application processor. The extracted event data requires minimal further processing compared to complete frame data.
3Loss of information
If frame-based camera data is transferred extensively, then all image information is available, but data transfer volume and power consumption increase
Solution Approach 1:
The patent applies local quality by having each sensor element independently detect intensity changes and generate events based on local conditions rather than transferring all pixel data centrally. This localized event generation ensures that only relevant change information is transmitted, maintaining the essential feature detection information while dramatically reducing data transfer volume and associated energy loss compared to transferring complete frame data.
Data Source
AI summary
Certain techniques are provided for triggering events based on detecting changes in CV features, such as edges, corners etc., by generating computed results based on sensor readings. Yet other techniques also provided for detecting CV features once an event is detecting based on a change in a sensor reading for a sensor element. In certain aspects, the event detection logic and the feature calculation CV vision operations may be performed, individually, or in combination in circuitry on a sensor or in software/firmware on a computing device.