Peripheral Zone Movement Recognition via Feature Point Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to accurately recognize movement of objects in the peripheral zone of a camera's field of view, especially when only a part of the object is within the view, leading to inefficiencies in controlling electronic devices.
Innovation Solution
The method involves dividing the camera's field of view into a middle zone and a peripheral zone, extracting feature points from objects in the peripheral zone, tracking these points to recognize movement, and assigning different control signals based on the location and type of movement within the field of view.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If the entire field of view is used for gesture recognition, then the coverage area is increased, but the recognition accuracy in peripheral zones deteriorates
Solution Approach 1:
The field of view is divided into multiple zones (central zone and peripheral zones) with different recognition strategies. The central zone uses full object recognition while peripheral zones use feature point-based recognition, allowing the system to maintain broad coverage while ensuring accurate recognition in critical areas.
Solution Approach 2:
Different recognition methods are applied to different spatial zones within the field of view. The central zone receives high-precision full object recognition while peripheral zones use optimized feature point tracking, ensuring that each zone's recognition quality matches its importance and visual information availability.
2Measurement precision
If feature point extraction is used for peripheral zone recognition, then the recognition accuracy is improved, but the processing complexity increases
Solution Approach 1:
Instead of performing full object recognition on all zones, the system applies feature point extraction only to peripheral zones where partial object visibility is common. This partial application of the more complex method only where needed reduces overall processing complexity while maintaining accuracy where required.
Solution Approach 2:
The recognition process is segmented into different methods for different zones: full object recognition for the central zone and feature point-based recognition for peripheral zones. This segmentation allows the system to manage processing complexity by applying computationally intensive methods only where necessary.
3Measurement precision
If different control signals are assigned based on object location, then the control precision is improved, but the system complexity increases
Solution Approach 1:
Different control signals and recognition strategies are assigned to different spatial zones within the field of view. The central zone triggers one type of control response while peripheral zones trigger different control responses, ensuring that control precision matches the spatial context and importance of each zone.
Data Source
AI summary
A method of controlling an electronic device by recognizing movement of an object includes obtaining at least one image including an image of the object; dividing the obtained at least one image into a middle zone and a peripheral zone; extracting one or more feature points of the object that are within the peripheral zone; recognizing movement of the object based on the extracted one or more feature points; and controlling the electronic device based on the recognized movement.