Passenger Cabin Imaging Using Combined NIR Wavelengths
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional driver monitoring systems in vehicles face challenges in identifying physical characteristics and differentiating between vehicle seat fabrics and driver clothing due to limited spectral information from near-infrared lighting, requiring complex and expensive optical coatings or filters.
Innovation Solution
A monitoring system that combines light rays from multiple sources operating at different spectral bandwidths to capture multiple spectral images, allowing for in-depth analysis and object identification within a passenger cabin without the need for multiple filters or coatings, using a processing unit to switch on/off light sources sequentially or simultaneously to form a combined light ray with a single wavelength.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If conventional near-infrared lighting at 850 nm or 940 nm is used, then the monitoring system can capture images of the driver, but the spectral information is very narrow and limited, making it difficult to identify physical characteristics and differentiate materials
Solution Approach 1:
The patent combines multiple light sources with different spectral bandwidths (e.g., 850 nm and 940 nm NIR LEDs) to illuminate the target simultaneously, capturing reflected light that contains combined spectral information from both sources in a single image, thereby widening the spectral information without adding complex optical components
Solution Approach 2:
A single imaging module is designed to capture images under combined multi-spectral illumination, making it universal for both driver monitoring and material identification tasks without requiring separate cameras or complex filter systems for different spectral bands
2Loss of information
If optical coatings or filters are added to capture multiple-band images, then spectral information can be improved, but the manufacturing cost and complexity increase significantly
Solution Approach 1:
The patent uses inexpensive, commercially available NIR LEDs at different wavelengths as light sources, replacing expensive optical coatings or filters with simple, easily manufactured LED components that can be readily integrated into the monitoring system
Solution Approach 2:
Instead of changing the physical structure of the imaging module with coatings or filters, the patent changes the illumination parameters by selecting LEDs with specific wavelengths (850 nm, 940 nm) to achieve multi-spectral imaging through software-controlled light source selection
3Measurement precision
If extensive image recognition algorithms are used to analyze images under narrow-band lighting, then object identification can be achieved, but the system complexity and processing requirements increase
Solution Approach 1:
The patent performs preliminary spectral differentiation at the illumination stage by using light sources with distinct wavelengths, so that materials with different spectral reflectance properties naturally differentiate themselves in the captured image, reducing the complexity of subsequent image recognition algorithms
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
Enables effective identification of objects and materials within the passenger cabin by capturing images with wider spectral information, facilitating the recognition of human skin and other objects, and reducing the complexity and cost of the imaging system.
Implementation Method 1
a plurality of light sources operable to emit light rays, and a processing unit, the processing unit operable to switch on/switch off each of the plurality of light sources, wherein each of the plurality of light sources is operable in a different spectral bandwidth
Implementation Method 2
Conventional monitoring system uses near-infrared (NIR) lighting at either 850 nm or 940 nm, which provides a very narrow bandwidth with limited spectral information. Apart from the challenge of recognizing physical characteristics of the driver, which often involves analyzing complex or multiple reflections appearing on images captured
Data Source
AI summary
A monitoring system for identifying objects in a passenger cabin of a motor vehicle is disclosed. The monitoring system includes an imaging module to capture multiple spectral images. The imaging module having a plurality of light sources for emitting light rays at different spectral bandwidth. The monitoring system further includes a processing unit operable to combine light rays emitted from at least two of the plurality of light sources into a combined light ray operating in a single wavelength, such that the imaging module is operable to capture a multiple spectral image of a passenger cabin for object identification. A method of identifying objections in a passenger cabin of a motor vehicle base on a multiple spectral image captured by a monitoring disclosed herein, a computer software product and a non-transitory medium prestored with the same is also disclosed.