Monocular Camera Depth Mapping for Confined-Space Inspection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Inspecting confined spaces such as pipes and manholes is difficult, dangerous, and costly using existing methods like tethered CCTV rovers and laser line profiling tools, which are slow and expensive.
Innovation Solution
A method using machine learning to derive depth calculations from monocular images or videos, enabling safe, cost-effective, and accurate inspection and mapping of confined spaces by processing images with a depth perception algorithm and overlaying depth data onto captured images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If tethered CCTV rovers and laser line profiling tools are used for inspection, then measurement accuracy is improved, but inspection speed and cost worsen
Solution Approach 1:
The patent replaces mechanical measurement systems (laser line profilers, tethered CCTV rovers) with a monocular camera system combined with machine learning algorithms. The depth perception algorithm processes images from the camera to calculate depth information, eliminating the need for complex mechanical profiling equipment while maintaining measurement capability.
Solution Approach 2:
The system creates a digital copy of the physical environment by capturing images with the monocular camera and generating depth maps through machine learning. This digital representation allows for measurement and analysis without requiring physical measurement tools to be present during inspection.
2Measurement precision
If tethered CCTV rovers are used for inspection, then measurement capability is improved, but safety and cost worsen
Solution Approach 1:
The patent replaces physical inspection equipment that requires entry into confined spaces with a monocular camera system that can capture images and generate depth information remotely. This eliminates the safety risks associated with sending personnel or tethered equipment into dangerous environments while preserving measurement capabilities.
Solution Approach 2:
The system performs self-measurement by using the monocular camera to capture images and the machine learning algorithm to automatically calculate depth and identify defects. This eliminates the need for external measurement tools or human intervention in hazardous areas.
3Measurement precision
If laser line profiling tools are used, then depth measurement accuracy is improved, but device complexity and cost worsen
Solution Approach 1:
The patent replaces complex optical-mechanical systems (laser line profilers) with a computational approach using a monocular camera and machine learning algorithms. The depth perception is achieved through software processing rather than physical laser projection and mechanical scanning, significantly reducing device complexity.
Solution Approach 2:
The system changes the fundamental parameter for depth measurement from optical phase or time-of-flight (used in laser systems) to pixel intensity patterns and geometric relationships in monocular images. The machine learning algorithm learns to infer depth from these 2D image parameters, simplifying the measurement approach.
Data Source
AI summary
Methods for providing depth-related information for images captured by a monocular-type camera are provided herein. A camera can move through a confined space capturing video. An image from the video can be captured and processed using a machine learning algorithm that identifies a depth of field from the camera point of view. The location of the captured image can be used along with known or pre-learned dimension data for the location, to identify the depth of field distances to objects found in the image. This depth data can be overlaid onto the image, and an operator can use this information to measure anomalies found in the image.