Lateral Vehicle Imaging for Back Seat Passenger Counting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing passenger counting systems struggle to accurately count individuals sitting on the back seat of a vehicle due to unclear input images from cameras positioned outside the vehicle.
Innovation Solution
A passenger counting system that captures images of the vehicle from the lateral and front directions, using image feature calculations to detect profiles and frontal faces, and determines the number of passengers based on these detections, thereby improving accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a camera is positioned outside the vehicle to capture images for passenger counting, then the system can count passengers non-invasively, but the image quality becomes insufficient to clearly capture faces and heads of persons on the back seat
Solution Approach 1:
The system transitions from traditional front-facing camera views to lateral direction imaging, capturing passenger profiles from the side. This dimensional change allows clear detection of back seat passengers who are occluded in frontal views, resolving the image clarity issue while maintaining counting accuracy
2Measurement precision
If the camera captures images from the front of the vehicle, then the system can detect faces and heads of front seat passengers, but it cannot correctly count persons sitting on the back seat
Solution Approach 1:
By positioning the camera to capture images from the lateral direction of the vehicle, the system obtains clear profile views of passengers including those on the back seat. This lateral perspective eliminates occlusion problems inherent in frontal views, enabling reliable detection of all passengers regardless of seating position
Solution Approach 2:
The system focuses on detecting specific local features (profiles of persons) that are most visible and distinctive from the lateral direction, rather than attempting to detect all features uniformly. This localized feature detection approach optimizes counting accuracy for the chosen imaging perspective
Data Source
AI summary
A passenger counting system that can count correctly the number of persons present in a vehicle, including persons sitting on the back seat, is provided. The passenger counting system includes an image capturing device 11 which is arranged outside the vehicle and which captures an image of the vehicle from the lateral direction of the vehicle, an image acquisition unit 31 that controls the image capturing device 11 and acquires an image in which the inside of the vehicle is captured as an input image, a profile detection unit 32 that detects a profile of a person from a partial image obtained from the input image based on an image feature amount calculated by using the partial image and outputs a detection result, and a passenger number determination unit 33 that determines the number of persons present in the vehicle based on the detection result.