Event Vision Sensor Motion Filtering for Low-Power Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vision sensors face issues with unnecessary event detection due to movement below a certain level, leading to inefficient power consumption and reduced accuracy in motion detection, especially in environments like snowy or rainy conditions.
Innovation Solution
A vision sensor with a pixel array and event detection circuit that generates event data, a processor to analyze this data, and a method to determine motion signals based on probability data and threshold comparisons, allowing the sensor to operate in monitoring and active modes to optimize power usage and reduce noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the vision sensor detects all events including small movements, then the detection sensitivity is improved, but the power consumption increases and false detections occur
Solution Approach 1:
The patent segments the detection process into two distinct modes: monitoring mode for low-power operation detecting only significant motion events, and active mode for comprehensive detection. This segmentation allows the system to achieve high detection sensitivity when needed while minimizing power consumption during normal operation by processing only essential event data.
Solution Approach 2:
The vision sensor dynamically switches between monitoring mode and active mode based on detection needs. The processor activates full detection capabilities only when motion is detected in monitoring mode, otherwise maintaining low-power operation. This dynamic adaptation resolves the contradiction by adjusting detection sensitivity according to operational context.
2Measurement precision
If the vision sensor processes all event data, then the detection accuracy is improved, but the processing time and computational resources increase
Solution Approach 1:
The patent applies partial action by processing only necessary event data rather than all possible data. In monitoring mode, only events exceeding a threshold are processed. In active mode, full processing occurs but only after motion is detected. This selective processing maintains detection accuracy while minimizing processing time and computational resource usage.
3Measurement precision
If the vision sensor operates in high sensitivity mode, then the motion detection capability is improved, but the noise from small movements increases
Solution Approach 1:
The patent implements preliminary anti-action by using the monitoring mode to detect and filter out small movements before they can generate noise in the active detection process. The monitoring mode acts as a preliminary filter, allowing only significant motion events to trigger full active mode processing, thereby preventing noise from minor movements from affecting the detection 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
Improves motion detection accuracy and reduces power consumption by selectively processing event data, minimizing resource usage and enhancing data clarity in various environments.
Implementation Method 1
A vision sensor, e.g., an active vision sensor, generates when an event (e.g., a change in light intensity) occurs, information about the event, that is, event data
Data Source
AI summary
According to the vision sensor, the operating method of the vision sensor, and the image processing device according to the inventive concepts, event data may be generated in response to the movement of an object, a plurality of pieces of comparison data may be generated by comparing each of a plurality of pieces of event data respectively corresponding to a plurality of sub regions with reference data, a motion count value may be updated by comparing each of the plurality of pieces of event data respectively corresponding to the plurality of sub regions with a first threshold value, and whether an object moves may be determined by comparing an updated final motion count value with a second threshold value.