Crane Human Detection Camera Alignment for Shifted View Regions

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing human detection systems in work vehicles, such as cranes, face challenges in determining whether the image capture region has shifted from its reference position, leading to inaccurate obstacle detection due to the obstruction caused by telescopic booms, which obstruct the operator's visual field.

Innovation Solution

A human detection system that utilizes a camera, monitor, and control device to continuously detect and visually indicate if the detection region has shifted by comparing histogram features of edge angles and luminance gradients using SVM learning, displaying reference and assisting figures to aid in adjusting the camera position.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a mask region is defined by coordinates in the monitored image to exclude the vehicle body from obstacle detection, then erroneous determination of obstacles is reduced, but the system cannot adapt when the image capture region shifts, leading to inaccurate obstacle detection

Engineering Contradiction:
Improveobstacle detection accuracyVSAvoidadaptation to image capture region shift
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The system performs preliminary action by detecting shifts in the image capture region before they cause detection errors. The shift detection unit continuously monitors the camera position and compares it against reference coordinates, enabling proactive adjustment of the mask region or notification to the operator before inaccurate obstacle detection occurs.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements feedback by using the detected image capture region shift information to adjust the mask region coordinates. The shift detection unit provides feedback about camera position changes, and this feedback is used to dynamically update the mask region definition, ensuring the vehicle body continues to be correctly excluded from obstacle detection despite camera movement.

Inventive Principle:
Principle #23Feedback

2Ease of operation

If the image capture region of the television camera shifts, then the monitored region may include areas that should be mask regions or exclude areas that should be detection regions, but quickly adjusting the attachment position is difficult without proper indication

Engineering Contradiction:
Improvecamera position adjustmentVSAvoidtime for position adjustment
Core Design Contradiction:
Ease of operationVSLoss of time

Solution Approach 1:

The notification unit provides immediate feedback to the operator when an image capture region shift is detected. This feedback mechanism enables quick awareness of the shift condition, allowing the operator to及时调整 the camera position before detection accuracy deteriorates, thereby reducing both adjustment time and operational difficulty.

Inventive Principle:
Principle #23Feedback

3Measurement precision

If the detection region is continuously monitored for shifts, then accurate determination of region displacement is achieved, but the system complexity increases

Engineering Contradiction:
Improveshift detection accuracyVSAvoidsystem structure
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system replaces complex mechanical adjustment mechanisms with an information processing approach. Instead of requiring complex mechanical systems to maintain perfect camera alignment, the invention uses image processing and coordinate comparison to detect shifts and provides information feedback, substituting mechanical complexity with computational simplicity.

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

Solution Approach 2:

The shift detection unit acts as an intermediary between the camera and the obstacle detection system. It monitors camera position changes and translates them into actionable information for the notification unit or mask region adjustment, simplifying the overall system architecture by introducing a dedicated intermediate component rather than requiring direct complex interactions between existing elements.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP3595298B1Human detection system for work vehicle, and work vehicle equipped with same
Publication Date: 2024.12.18 TADANO LTD
  • EP3595298B1 patent drawingFigure 1
  • EP3595298B1 patent drawingFigure 2~3
  • EP3595298B1 patent drawingFigure 4A~4B

AI summary

Provided are: a human detection system for a work vehicle, which can assist in a determination regarding whether a detection region in which image data is acquired by an image data acquisition section has deviated from a reference region which is a photography region at a reference position; and a work vehicle equipped with the same. A human detection system 18 comprises: a camera 19 which serves as an image data acquisition section that acquires detection image data Dd of a detection region D, the detection region D being a prescribed region including a portion of a crane 1 which is a work vehicle; a deviation determination unit 21c in which image data of the crane 1 included in the detection image data Dd acquired by the camera 19 in a state of being disposed at a reference position is set as reference image data Sd of the detection region D, and in which the detection image data Dd of the detection region D acquired at an arbitrary time by the camera 19 is compared to the reference image data Sd to determine whether the position of the camera 19 has deviated from the reference position; and a reporting unit 21d that reports that the position of the camera 19 has deviated.