Alternating Eye and Hand Detection in Vehicle Cameras
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for recognizing eyes and hands in motor vehicles require multiple cameras and computers, leading to increased hardware and computing resource demands, and struggle to efficiently detect the distinct movements of eyes and hands simultaneously.
Innovation Solution
A system that uses a single camera and computer to alternately process sequences of images using separate algorithms for eye and hand recognition, with a predetermined ratio of images allocated to each algorithm, allowing for efficient detection of eye and hand movements without requiring extensive resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If separate computers and sensors are used for eye detection and hand detection, then detection accuracy for both eyes and hands is improved, but hardware complexity and computing resource demands increase
Solution Approach 1:
The patent combines separate eye detection and hand detection systems into a single integrated system. One camera captures images containing both eyes and hands, and one computer processes all images using alternating algorithms - first an eye detection algorithm, then a hand detection algorithm. This merging reduces hardware complexity while maintaining detection accuracy through specialized processing sequences.
Solution Approach 2:
The patent segments the image processing task into distinct phases: eye detection phase and hand detection phase. The computer alternates between applying an eye detection algorithm to identify eye positions and a hand detection algorithm to identify hand positions and gestures. This segmentation allows each algorithm to be optimized for its specific task while sharing the same hardware resources.
2Power
If separate computers are used for eye and hand detection, then processing capability is improved, but computing resource consumption increases
Solution Approach 1:
The computer employs periodic action by alternating between eye detection processing and hand detection processing in a systematic sequence. Rather than continuously running both algorithms simultaneously on separate systems, the single computer cycles through eye detection on one set of images, then hand detection on subsequent images, optimizing resource utilization while maintaining comprehensive detection capability.
3Quantity of substance
If a single camera is used for both eye and hand detection, then hardware cost is reduced, but detection reliability may deteriorate
Solution Approach 1:
The patent applies parameter changes by dynamically adjusting processing parameters based on detection needs. The system modifies algorithm parameters, processing sequences, and resource allocation based on whether eyes or hands are being detected, allowing a single camera system to achieve reliable detection results for both body parts through adaptive parameter optimization.
Data Source
Figure 1
Figure 2~3
AI summary
The invention relates to a system (100) and a method (500, 600), in particular for a motor vehicle (10), for detecting eyes and hands in a sequence of images captured by a camera (110) of the system (100), wherein the system (100) comprises a computer (150), which is designed to feed, in alternation, a first partial sequence comprising a predefinable first number of successive images to a first algorithm, configured in particular in the computer (150), for eye detection and a second partial sequence comprising a predefinable second number of successive images to a second algorithm, configured in particular in the computer (150), for hand detection, wherein the first algorithm, according to the first partial sequence, identifies whether the images of the first partial sequence comprise an image of an eye, and wherein the second algorithm, according to the second partial sequence, identifies whether the images of the second partial sequence comprise an image of a hand.