Dynamic Vision Sensor Static Object Detection via Intensity Modulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Dynamic Vision Sensor (DVS) cameras struggle to capture static objects, leading to blind spots in applications like human tracking and SLAM, as they are sensitive only to dynamic objects and lose track of slowly moving or still objects.
Innovation Solution
A dynamic vision sensing system is developed, incorporating an AI recognition module and a static mode triggering module that uses an optical module with a GPA or liquid lens to induce intensity changes, allowing the camera to capture static objects by switching between focusing and defocusing states, enabling the detection and classification of actions and recognition of static objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If DVS camera operates in dynamic mode to capture moving objects with high sensitivity, then motion detection capability is improved, but static object detection capability deteriorates
Solution Approach 1:
The system dynamically switches between two operational modes: dynamic mode for capturing moving objects and static mode for capturing stationary objects. The static mode triggering module detects when an object becomes static and triggers a mode switch, allowing the DVS camera to adapt its detection characteristics based on the current state of the target object.
Solution Approach 2:
The system changes the operational parameters of the DVS camera by triggering an intensity change in the environment when static mode is activated. This parameter change allows the camera to detect static objects by capturing intensity variations caused by environmental changes or object state transitions, rather than relying solely on motion-induced intensity changes.
2Use of energy by moving object
If DVS camera captures only dynamic information to reduce power consumption, then energy efficiency is improved, but object recognition completeness deteriorates
Solution Approach 1:
The system employs periodic action by only activating the static mode triggering mechanism when necessary - specifically when the AI recognition module detects that an object of interest has become static or disappeared. This periodic activation maintains low power consumption during normal dynamic operation while periodically checking for and capturing static object information when needed.
Solution Approach 2:
The AI recognition module performs self-service by autonomously monitoring the detected objects and automatically triggering the static mode when it detects that an object has become static or disappeared. This self-triggering mechanism eliminates the need for continuous high-power operation while ensuring that static objects are captured when relevant, maintaining information completeness without sacrificing energy efficiency.
3Measurement precision
If AI recognition module continuously monitors environment for static objects, then object detection accuracy is improved, but system complexity increases
Solution Approach 1:
The AI recognition module performs preliminary action by continuously monitoring and classifying actions of detected objects to determine their state (moving vs. static). This preliminary classification allows the system to proactively identify when an object has become static and trigger the appropriate capture mode before the object is lost, improving detection accuracy without requiring complex continuous static monitoring mechanisms.
Solution Approach 2:
The action classification module serves as an intermediary between the object detection module and the static mode triggering module. It processes the detected objects and their actions, and based on its classification, triggers the static mode when appropriate. This intermediary layer simplifies the overall system architecture by using existing action classification capabilities to infer static object states, rather than implementing a separate complex monitoring system.
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
The system effectively captures static objects with intensity changes at their boundaries, expanding DVS camera functionality beyond dynamic object tracking, enabling applications in video surveillance, accident detection, and other industries with improved motion recognition and low power consumption.
Implementation Method 1
The lens focusing power of the liquid lens is changed periodically to establish a small shift in focusing as to induce an intensity change at the dynamic vision sensor sensor behind the lens module for static scenery
Implementation Method 2
In the case of the liquid lens electro-optical mechanical module, the liquid lens electro-optical mechanical module for example may contain a liquid lens placed at a frontmost position before the lens module
Implementation Method 3
a circular polarizer adapted to filter a polarization from an incoming light to the circular polarizer at a predetermined handedness
Implementation Method 4
a GPA (Geometric Phase Axicon) module located after the circular polarizer on an optical path. The GPA module is adapted to alter the focus of an environmental scene inputted to the circular polarizer
Data Source
AI summary
A dynamic vision sensing system, which includes a dynamic vision sensor, a AI recognition module connected to the dynamic vision sensor, and a static mode triggering module coupled to the Ai recognition module. The static mode triggering module is adapted to trigger an intensity change in an environment captured by the dynamic vision sensor to observe a static object of interest, a property that conventional dynamic vision sensor failed to capture. The motion recognition module upon detecting a change of a motion in the environment is adapted to send a command to the static mode triggering module to trigger the intensity change. An AI controlled electro-optical system is therefore proposed to capture object of interest even if the object is static by triggering effective intensity change of the static object at sensing end, hence eliminate blind spot of existing DVS system.