Vehicle Seat Occupancy Detection Using Depth-Intensity CNN Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting seat occupancy in vehicles are inefficient and resource-intensive, particularly when distinguishing between different types of occupants or objects on the seat, leading to potential inaccuracies and increased computational demands.
Innovation Solution
A method utilizing a combination of depth and intensity data from imaging devices, processed by classifier algorithms and convolutional neural networks, specifically a Siamese convolutional neural network, to determine seat occupancy levels and types, optimizing data processing by threshold-based differentiation and adaptive reference image updates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If neural networks are applied to process both depth data and intensity data for seat occupancy detection, then detection accuracy is improved, but computational resource consumption increases
Solution Approach 1:
The patent divides the detection task into two separate processing paths: one for depth data and one for intensity data. Each path uses a dedicated neural network model trained for its specific data type, allowing selective and efficient processing rather than applying a single comprehensive model to all data
Solution Approach 2:
The system applies neural network processing only when necessary - using depth data processing when occupancy is detected and intensity data processing for supplementary information. This partial application of complex processing reduces overall computational load while maintaining accuracy where needed
2Reliability
If comprehensive data processing is applied to all imaging data, then detection reliability is improved, but processing time increases
Solution Approach 1:
The patent segments the imaging data into depth data and intensity data, processing each through specialized neural networks simultaneously. This parallel processing of divided data streams maintains comprehensive analysis while reducing sequential processing time
Solution Approach 2:
The system performs preliminary classification of the seating surface using depth data first, then uses this information to guide subsequent intensity data processing. This preliminary action allows the system to focus computational resources on relevant areas, reducing overall processing time while maintaining reliability
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
Enhances accuracy and reduces computational resources by selectively applying neural networks to relevant data channels, improving the detection of seat occupancy types while minimizing unnecessary processing and adapting to changes in seat conditions.
Implementation Method 1
The imaging device may be, for example, a time-of-flight camera, a stereo camera or a radar camera. The depth data represent, for each pixel, a depth information, which corresponds to a real distance in height
Implementation Method 2
the intensity data represent a luminescence for the corresponding pixel
Data Source
AI summary
Computer implemented method for detecting a type of seat occupancy, comprising capturing, by means of an imaging device, an image of a seat, the image comprising depth data and intensity data, performing, by means of a processor device, a classifier algorithm on the captured image to determine a level of occupancy, wherein, if the determination indicates that the level of occupancy is above a predetermined threshold, the method comprises processing, by means of the processor device, the depth data with a convolutional neural network, to determine a type of occupation and wherein, if the determination indicates that the level of occupancy is below a predetermined threshold, the method comprises processing, by means of the processor device, the intensity data with a convolutional neural network to determine a type of occupation.