Neural Network Road Defect Detection System

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Poor road infrastructure conditions, particularly potholes, lead to safety issues, resource loss, and traffic congestion, necessitating an efficient and cost-effective solution for data collection and analysis to identify and address these problems.

Innovation Solution

A data collection system comprising a camera, location module, and sensors, coupled with neural network models, is used to identify problematic road segments, including potholes, cracks, and pavement markings, and transmit this information for remote processing and classification.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If traditional manual inspection methods are used to assess road conditions, then detailed examination can be performed, but the process is time-consuming and labor-intensive

Engineering Contradiction:

Improveroad inspection efficiencyVSAvoidinspection time

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The patent replaces manual mechanical inspection with an automated optical-mechanical system consisting of a camera mounted on a vehicle, GPS receiver for location tracking, and a processor that automatically captures images and identifies road defects. This substitution eliminates the need for manual inspection while maintaining detection capability, thereby improving productivity and reducing time loss.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If extensive road inspection is conducted to identify all defects, then comprehensive data is collected, but the cost and resources required increase significantly

Engineering Contradiction:

Improvedefect detection accuracyVSAvoidresource consumption

Core Design Contradiction:

Measurement precisionVSLoss of energy

Solution Approach 1:

The system captures images at a high frame rate (60 fps) which exceeds the minimum necessary for defect detection. This excessive action ensures that no defect is missed due to timing or positioning, while the automated processing selectively identifies and flags only the relevant defects for further analysis, maintaining high detection accuracy without requiring exhaustive manual examination of every captured image.

Inventive Principle:

Principle #16Partial or excessive action

Solution Approach 2:

The processor automatically analyzes captured images to identify road defects without requiring manual review of each image. The system self-services by autonomously detecting defects, determining their locations, and generating reports, thereby reducing the need for extensive human resources while maintaining comprehensive defect detection capability.

Inventive Principle:

Principle #25Self-service

3Measurement precision

If multiple sensors and processing units are added to improve data collection accuracy, then defect identification improves, but system complexity increases

Engineering Contradiction:

Improveroad defect identification accuracyVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent combines multiple functional components into an integrated system: a camera for image capture, a GPS receiver for location data, and a processor that performs both image analysis and defect identification. These components are merged into a single coordinated system that collects and processes data simultaneously, improving measurement precision while managing system complexity through integration rather than separate distributed components.

Inventive Principle:

Principle #5Merging (Combining)

4Measurement precision

If high-resolution imaging is used to detect small defects, then detection capability improves, but data processing requirements and storage needs increase

Engineering Contradiction:

Improvesmall defect detection capabilityVSAvoiddata volume

Core Design Contradiction:

Measurement precisionVSQuantity of substance

Solution Approach 1:

The processor extracts and identifies only the relevant defect information from the captured images, separating the essential defect data from the rest of the image data. By extracting only the critical defect characteristics and locations rather than processing and storing all image data, the system maintains high detection capability for small defects while significantly reducing the volume of data that requires storage and further processing.

Inventive Principle:

Principle #2Taking out (Extraction)

Data Source

PatentUS11538256B2Systems and methods for data collection and performance monitoring of transportation infrastructure

Publication Date: 2022.12.27 ROWAN UNIVERSITY

AI summary

The present invention provides a data collection system comprising: a camera; a location module; a plurality of sensors; and a first processor communicatively coupled to the camera and the location module, the first processor programmed to: obtain a plurality of frames from the camera; obtain a plurality of locations from the location module; obtain a plurality of data measurements from the plurality of sensors; apply a previously trained first neural network model for identifying problematic road segments to frames captured by the camera; and if the first neural network model indicates that a frame is a problematic road segment, save the frame in association with a location provided by the location module.