Photovoltaic Module Detection via Coarse-Fine Contour Masking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for detecting photovoltaic modules on images, such as threshold detection and neural network methods, fail to accurately detect module contours and positions, leading to inaccurate maintenance and potential energy production losses.
Innovation Solution
A method involving a computer-based process that includes coarse detection by a model, contour mask determination through image processing, and fine detection using contour masks to accurately identify photovoltaic modules and their contours on images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If threshold detection methods are used, then detection speed is improved, but contour detection precision deteriorates
Solution Approach 1:
The detection process is divided into two independent stages: rough detection using threshold methods for speed, and fine detection using contour analysis for precision. Each stage handles different aspects of the detection task, allowing both speed and precision requirements to be met simultaneously.
Solution Approach 2:
The method dynamically adjusts the detection approach by first applying a quick threshold-based rough detection, then selectively applying more computationally intensive contour analysis only to regions of interest identified in the first stage. This dynamic two-stage approach optimizes both speed and precision.
2Measurement precision
If neural network detection methods are used, then detection accuracy is improved, but computational complexity increases
Solution Approach 1:
The complex detection task is segmented into rough detection (low complexity) and fine detection (higher complexity) stages. By dividing the problem, the overall computational burden is reduced while maintaining high accuracy through the specialized fine detection stage.
Solution Approach 2:
Instead of applying full neural network processing to the entire image, the method applies sophisticated contour analysis only to specific regions identified by the rough detection stage. This partial application of complex processing reduces overall computational complexity while maintaining detection accuracy.
3Productivity
If rough detection with enclosing shapes is used, then processing speed is improved, but contour accuracy deteriorates
Solution Approach 1:
The detection process segments contour detection into rough enclosing shape generation (fast) and fine contour refinement (accurate) stages. The rough stage provides quick bounding boxes, while the fine stage delivers precise contours only where needed.
Solution Approach 2:
The rough detection stage performs preliminary identification of module locations using simple enclosing shapes, preparing the groundwork for the subsequent fine detection stage. This preliminary action filters out unnecessary regions, allowing detailed contour analysis to focus only on relevant areas.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present invention relates to a method for detecting photovoltaic modules on an image, the method comprising: a. receiving an initial image of photovoltaic modules, b. coarsely detecting, by a model, at least one photovoltaic module imaged entirely on the initial image and highlighting said photovoltaic module on the initial image by an enclosing shape, c. determining a contour mask for each photovoltaic module detected on the initial image as a function of the enclosing shape, and d. applying each determined contour mask to the initial image to finely detect the corresponding photovoltaic module.