Image Object Detection With Learned Locations for Low-Visibility Monitoring

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing systems struggle to continuously monitor and identify objects in images during periods of low visibility, such as night or adverse weather, especially in hazardous environments like wellsites, without requiring constant human presence.

Innovation Solution

An intelligent image analytics system using object detection models and automated learning to cluster detected objects by centroids, calculate geometric averages, and determine if they meet a minimum lighting threshold, enabling persistent object location even in low visibility conditions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If automated learning with clustering is implemented to learn object locations, then monitoring reliability during low visibility improves, but system complexity increases

Engineering Contradiction:

Improvemonitoring reliabilityVSAvoidsystem complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system performs preliminary actions by clustering detected objects and calculating geometric averages during periods of good visibility to establish baseline object locations. This preliminary learning phase enables the system to later identify objects during low visibility conditions without requiring complex real-time processing, thus improving reliability while managing system complexity.

Inventive Principle:

Principle #10Preliminary action

2Productivity

If continuous monitoring is implemented without human presence, then productivity improves, but measurement precision deteriorates during low visibility

Engineering Contradiction:

Improvemonitoring efficiencyVSAvoidobject detection accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The system uses feedback by comparing current image data against learned object locations from clustering. During low visibility conditions, the system references previously learned geometric averages and centroid positions to maintain measurement precision. This feedback mechanism allows continuous automated monitoring while compensating for reduced detection accuracy in adverse conditions.

Inventive Principle:

Principle #23Feedback

3Area of stationary object

If object detection is performed in low visibility conditions, then monitoring coverage improves, but detection precision worsens

Engineering Contradiction:

Improvemonitoring coverageVSAvoidobject detection precision

Core Design Contradiction:

Area of stationary objectVSMeasurement precision

Solution Approach 1:

The system transitions from relying solely on visual image data to incorporating spatial dimension information through clustering and geometric average calculations. By learning object locations across multiple dimensions (x-coordinates, y-coordinates) during visible periods, the system can maintain detection precision in low visibility conditions while expanding monitoring coverage to areas that would otherwise be inaccessible.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

Data Source

PatentUS20250218185A1Systems, methods, and computer program products for object detection and analysis of an image

Publication Date: 2025.07.03 DEVON ENERGY CORP

AI summary

Systems, methods, and computer program products of intelligent image analysis using object detection models to identify objects and locate and detect features in an image are disclosed. The systems, methods, and computer program products include automated learning to identify the location of an object to enable continuous identification and location of an object in an image during periods when the object may be difficult to recognize or during low visibility conditions.