Vehicle Occupant Counting With Certainty-Based Result Review
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for computing the number of occupants in a vehicle often result in errors, making it impractical for humans to verify the accuracy of the computed results due to the vast amount of information processed by computers.
Innovation Solution
A program that processes sensing data from cameras, microphones, or millimeter wave radars to compute the number of occupants and their certainty factors, allowing for the identification and visual confirmation of potentially erroneous results by distinguishing them from accurate ones.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a computer processes enormous information to compute the number of occupants in a vehicle, then the computation result is obtained, but it becomes impractical for humans to verify the accuracy of the computation result
Solution Approach 1:
The patent extracts only the necessary information for verification from the enormous processed data. The computation result extraction unit identifies and extracts specific computation results that need verification, separating them from the vast amount of processed information. This allows human verifiers to focus only on critical data points rather than reviewing all processed information, thus maintaining computation efficiency while making verification feasible.
Solution Approach 2:
The verification process is segmented into multiple stages: automatic computation by the computer, extraction of key results requiring verification, and targeted human verification. This segmentation divides the overwhelming task of verifying all processed information into manageable portions, allowing humans to efficiently verify only the extracted computation results while the computer handles the extensive data processing.
2Reliability
If all processed information is visually confirmed by human, then verification accuracy is improved, but the workload and time consumption become excessive
Solution Approach 1:
Instead of verifying all processed information, the patent applies partial verification by extracting only the computation results that require human verification. The computation result extraction unit identifies specific results based on predetermined conditions, allowing human verifiers to focus on critical data points. This partial verification approach maintains sufficient reliability for decision-making while dramatically reducing the time and workload compared to complete verification of all processed information.
3Measurement precision
If the computer processes enormous information for occupant count computation, then comprehensive data analysis is achieved, but the complexity of information management increases
Solution Approach 1:
The computation result extraction unit serves as an intermediary between the computer's extensive data processing and human verification. It acts as a filter that receives the enormous processed information, applies predetermined conditions to identify relevant computation results, and outputs only the necessary results for human review. This intermediary simplifies information management by bridging the gap between comprehensive automated processing and targeted human verification, reducing the complexity of managing and navigating vast amounts of processed data.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An information processing apparatus (10) includes a computation unit (110), a selection unit (120), and a processing unit (130). The computation unit (110) analyzes a sensing result acquired by sensing a vehicle running on a road, and thereby computes the number of occupants in the vehicle and a certainty factor of the number of the occupants. As one example, the computation unit (110) computes reliability in the case of determining that a person exists related to each seat of the vehicle, and uses the reliability for each seat to compute a certainty factor related to each possible number as the number of the occupants in the vehicle. The selection unit (120) selects a sensing result with the computed certainty factor of the number of occupants which does not satisfy a predetermined criterion as an analysis result. The processing unit (130) allows the sensing result selected by the selection unit (120) to be in the state of being distinguishable from the other sensing results.