Automated Construction Quality Inspection via Image Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The quality of work in construction sites is not managed efficiently due to varying proficiency levels among workers, leading to delays and potential undetected substandard work, as conventional inspection methods are time-consuming and resource-intensive.
Innovation Solution
An information processing method that compares design information with measured information from images of completed work, providing notifications when differences fall within specific acceptable ranges, allowing for focused quality checks and reducing unnecessary processing and power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional inspection methods are used to check work quality, then all work can be inspected, but it takes a great deal of time and leads to delays in construction
Solution Approach 1:
The patent replaces manual mechanical inspection with an automated image processing system that captures images of work objects, extracts features, and compares them against design information to automatically determine quality compliance, thereby eliminating time-consuming manual inspection while maintaining comprehensive quality coverage
Solution Approach 2:
The patent creates digital copies of the work through image capture and processing, generating measured information that replicates the physical work characteristics. This digital replica can be analyzed without physically interfering with or delaying the construction process
2Reliability
If conventional inspection methods are used to check work quality, then all work can be inspected, but significant hardware resources or power consumption are required
Solution Approach 1:
The patent extracts only the essential features from captured images that are relevant to quality assessment, rather than processing entire high-resolution images. This feature extraction approach significantly reduces computational load and power consumption while maintaining inspection effectiveness
Solution Approach 2:
The patent performs partial processing by focusing only on critical features and areas that impact quality compliance. Rather than analyzing every pixel or detail of the work, the system selectively processes only the necessary portions to determine whether work meets quality standards
3Measurement precision
If inspection is performed on all objects, then quality can be detected, but it may not be possible to detect an object having quality less than a certain quality due to resource constraints
Solution Approach 1:
The patent applies different levels of inspection scrutiny to different work objects or features based on their importance. Critical features that significantly impact quality are inspected with higher precision, while less critical aspects receive proportionate attention, optimizing the balance between detection accuracy and processing throughput
Data Source
AI summary
An information processing method includes: obtaining design information regarding work, the design information including a reference value and an acceptable range that includes the reference value; identifying a difference between the design information obtained and measured information regarding the work by comparing the design information with the measured information, the measured information being obtained from an image obtained by capturing an object on which the work has been performed; and providing a notification to a user when the difference identified falls within a given range that is in a neighborhood of an upper limit or a lower limit of the acceptable range.