Rear Vehicle Camera Object Drop Detection via Frame Comparison

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Drivers face challenges in monitoring objects loaded on vehicle carriers, as it is difficult to continuously observe these objects while driving, leading to potential drops without detection.

Innovation Solution

A method and apparatus using cameras mounted at the rear of vehicles to capture and compare image frames of the road, determining if an object has dropped by identifying differences in image sections, with a data processing device providing warnings through auditory or visual alerts.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a driver pays close attention to an object loaded on a carrier, then the object drop can be detected, but the driver's attention is diverted from the road

Engineering Contradiction:
Improveobject drop detectionVSAvoiddriver distraction
Core Design Contradiction:
ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The monitoring system performs automatic object detection and comparison of image frames without requiring driver intervention. The system serves itself by autonomously capturing images, processing them through the processor, comparing consecutive frames, and generating alerts, thereby eliminating the need for driver attention while maintaining reliable detection

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The manual visual monitoring by the driver is replaced with an automated optical-electronic system. Cameras capture images of the carrier and road, the processor electronically analyzes frame differences to detect object drops, and automated alerts are generated, substituting the mechanical human observation process with an electronic monitoring system

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If advanced image recognition technology is used to monitor objects, then detection accuracy is improved, but system complexity increases

Engineering Contradiction:
Improveobject drop detection accuracyVSAvoidimage recognition system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

Instead of using complex full-frame image recognition, the system applies partial action by focusing only on specific image sections that contain the carrier and potential objects. The processor compares only the relevant portions of consecutive image frames, achieving sufficient detection accuracy while minimizing computational complexity and processing requirements

Inventive Principle:
Principle #16Partial or excessive action

Solution Approach 2:

The monitoring system segments the image processing task by dividing the full image into specific sections of interest. The processor identifies and compares only the carrier-containing sections across consecutive frames, separating the essential monitoring function from unnecessary full-image analysis, thereby reducing complexity while maintaining detection effectiveness

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP4213113A1A scheme for monitoring an object loaded on a carrier of a vehicle
Publication Date: 2023.07.19 ALPS ALPINE CO LTD
  • EP4213113A1 patent drawingFigure 1(A)~1(B)
  • EP4213113A1 patent drawingFigure 2
  • EP4213113A1 patent drawingFigure 3A

AI summary

A method of monitoring, by a data processing device comprising a processor, an object loaded on a carrier of a vehicle is provided. The method comprises obtaining a plurality of image frames of a road on which the vehicle is travelling, captured by at least one camera mounted at a rear portion of the vehicle at different points of time; selecting a first image frame and at least one second image frame among the plurality of image frames such that the first image frame and the at least one second image frame have a first image section and at least one second image section, respectively, containing an image of the same part of the road; comparing the first image section and the at least one second image section; and determining if the object has dropped down on the road based on the comparison result.