Rule-Based Fuzzy Decision System for Power Line Image Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing power line inspection processes are time-consuming due to the manual selection of candidate images from a large volume of data captured during inspections, which lacks efficient strategies to streamline this task.
Innovation Solution
A computer-implemented method that uses a rule-based fuzzy decision system to select candidate images for power line inspections. This method combines geometric information with artificial intelligence algorithms, evaluating two independent criteria to determine whether an image contains a power tower and selecting or discarding images based on these evaluations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual review processes are used to select candidate images, then inspection accuracy can be maintained, but the time required for reviewing images increases significantly
Solution Approach 1:
The image selection process is divided into multiple independent criteria: geometric criterion (checking if tower GPS coordinates fall within image FOV), AI-based criterion (CNN classification), and rule-based fuzzy decision system. Each criterion processes different aspects of image evaluation separately, then combines results to make final selection, improving both speed and accuracy.
Solution Approach 2:
A rule-based fuzzy decision system acts as an intermediary between the geometric criterion and AI-based criterion. It combines results from both criteria using fuzzy logic rules, allowing the system to leverage strengths of each approach while mitigating their individual weaknesses in image selection.
2Device complexity
If a single-decision criterion process is used for image selection, then the process is simpler to implement, but the reliability of image selection decreases
Solution Approach 1:
The selection process is segmented into distinct criteria: geometric criterion for quick filtering, AI-based criterion for pattern recognition, and rule-based fuzzy decision system for final integration. This segmentation allows each component to be optimized independently while maintaining overall system reliability.
Solution Approach 2:
Different criteria are applied to different aspects of image evaluation: geometric criterion handles spatial positioning, AI-based criterion handles visual pattern recognition, and fuzzy decision system handles integration logic. Each part has specialized quality optimized for its specific function, improving overall reliability.
3Speed
If CNN output is used without verification to track towers, then real-time processing is achieved, but the accuracy of tower detection may be compromised
Solution Approach 1:
Geometric criteria are applied preliminarily to filter images before detailed AI analysis. By pre-screening images based on GPS coordinates and field-of-view calculations, the system reduces the number of images requiring full CNN processing, maintaining real-time speed while improving detection accuracy through multiple verification stages.
Solution Approach 2:
The system implements feedback mechanisms where geometric criterion results and rule-based fuzzy decision outcomes are fed back to verify and correct CNN detections. This multi-stage verification ensures real-time processing while maintaining high detection accuracy through iterative refinement.
Data Source
AI summary
A method, system, and computer programs for selecting candidate images for a power line inspection process are proposed. The method comprises receiving an image of an infrastructure acquired by a camera included in a flying object; tagging the received image with GNSS metadata including a GPS location of the camera and a pose of the camera based on IMU values of the flying object; executing a first-decision criterion that determines whether a GPS location of a power tower is comprised within a 2D projection captured in the image using a list of possible GPS power tower locations and a FOV, the pose, and/or the GPS location of the camera; executing a second-decision criterion that determines whether a power tower is included in the tagged image using the tagged image; selecting/discarding the received image as candidate image by executing a ruled based fuzzy decision system that combines the result of the two criterions.