In-Vehicle Passerby Detection Using Interior-Exterior Image Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing driver-assistance technologies struggle to accurately detect passersby outside a vehicle when images captured include both interior and exterior views, often leading to erroneous detection of occupants as passersby.
Innovation Solution
A driver-assistance apparatus that captures images of the vehicle's interior and exterior, uses image recognition to estimate characteristics such as distance, relative speed, and height, and outputs warnings for detected passersby based on predetermined criteria.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Difficulty of detecting and measuring
If a camera disposed inside the vehicle captures images including both interior and exterior views, then the system can detect passersby around the vehicle, but it causes erroneous detection of occupants as passersby
Solution Approach 1:
The system applies different detection criteria to different spatial regions within the captured image. By identifying the interior region of the vehicle and applying exclusion logic specifically to that zone, the system maintains high sensitivity for exterior passerby detection while eliminating false positives from interior occupants. This localized differentiation resolves the contradiction between comprehensive detection coverage and detection accuracy.
2Area of stationary object
If the camera imaging range includes the interior of the vehicle, then the system can capture a broader field of view, but it increases the complexity of distinguishing passersby from occupants
Solution Approach 1:
The detection system segments the captured image into distinct regions: interior vehicle space and exterior environment. By dividing the imaging area into these functional zones and applying region-specific detection rules, the system handles the complexity of wide-area monitoring through modular processing. This segmentation approach reduces algorithmic complexity while maintaining comprehensive coverage.
Solution Approach 2:
The system introduces an intermediary classification step that first identifies whether a detected person belongs to the interior or exterior region before applying passerby detection logic. This intermediate classification layer simplifies the overall detection process by filtering out interior occupants early in the processing pipeline, reducing the computational burden on subsequent passerby identification algorithms.
Data Source
AI summary
A driver-assistance apparatus to be attached to a moving body captures an image including an interior of the moving body and a subject outside the moving body, recognizes, by using the captured image, a person in the image, and estimates one or more characteristics of the recognized person in the image. The apparatus detects, as a passerby, the person in the image having each of the one or more estimated characteristics satisfying a predetermined criterion; and causes an output device to output a warning about a presence of the detected passerby. The one or more characteristics include at least one of a distance from the driver-assistance apparatus to the person, a relative moving speed of the person with respect to the moving body, and a height of the person.