Event-Based Vision Sensor Object Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing event-based vision sensors lack sufficient peripheral technologies to effectively recognize vibrating objects, despite their advantages in high-speed operation and low power consumption.
Innovation Solution
A computer system and method that utilizes an event-based vision sensor to recognize vibrating objects by generating event signals at the timing of vibration occurrence, incorporating a processor to acquire and analyze these signals in conjunction with other sensors like RGB cameras and dToF sensors to enhance object recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If an event-based vision sensor is used to detect vibrating objects, then detection speed and power efficiency are improved, but detection accuracy for thin or difficult-to-detect objects deteriorates
Solution Approach 1:
The patent combines multiple sensors (event-based vision sensor, frame-based vision sensor, and other sensors) to integrate their respective strengths. The event-based sensor provides high-speed detection capability while the frame-based sensor complements it with higher resolution information, thereby maintaining both fast detection speed and accurate detection of thin objects
Solution Approach 2:
The system performs multiple functions by using the event-based vision sensor for both high-speed vibration detection and object recognition. By integrating event signals with frame-based images and other sensor data, the system achieves multi-functional capability that addresses both speed and accuracy requirements
2Use of energy by moving object
If an event-based vision sensor operates at high speed, then power consumption is reduced, but peripheral technology sufficiency deteriorates
Solution Approach 1:
The system performs preliminary processing by integrating event signals with frame-based images and other sensor data before final object recognition. This preliminary action prepares the data in advance, reducing the computational burden on subsequent processing stages and simplifying the overall peripheral technology requirements
Solution Approach 2:
The patent introduces an intermediary processing stage that combines event signals with frame-based images and other sensor data. This intermediary layer acts as a bridge between the high-speed event-based sensor and the object recognition system, managing the complexity of peripheral technologies while maintaining energy efficiency
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
Enhances the ability to detect and recognize vibrating objects, particularly thin or difficult-to-detect objects like nets, with improved accuracy and reduced latency, using event signals to analyze luminance changes caused by vibration.
Implementation Method 1
event signals are asynchronously generated by pixels that detect a change in intensity of incident light, specifically, a change in luminance of a surface of a subject
Data Source
AI summary
Disclosed is a computer system for recognizing an object. The computer system includes a memory for storing a program code and a processor for performing operations in accordance with the program code. The operations include acquiring information indicating the occurrence of vibration in the object and recognizing the object according to an event signal generated by an event-based vision sensor at the timing of the occurrence of the vibration.