Automatic generation method of photovoltaic panel template based on surface line primitive association model
A photovoltaic panel and associated model technology, applied in image analysis, image enhancement, instrumentation, etc., can solve problems such as template failure, area analysis interference, difficulty in ensuring rule universality, etc., to achieve complete automation and avoid precision defects.
Active Publication Date: 2020-04-03
NANJING NORMAL UNIVERSITY
View PDF1 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0007] For small artificial ground objects such as photovoltaic panels (Photovoltaic panel, PVP), if the conventional method of OBIA is used for extraction, that is, after the image is segmented, the patterns that meet the rules are extracted for output. The disadvantages are: 1) It is difficult to guarantee Separation of photovoltaic panels and other ground features; some non-photovoltaic panel blocks may meet the above rules, but if more stringent conditions are imposed on the rules, it is difficult to ensure the universality of the rules, resulting in loss of panels; this is also the classification of OBIA rules 2) Due to the unavoidable under- and over-segmentation errors in segmentation, some panel shapes cannot meet the requirements of the rules and cannot be extracted; 3) The positioning accuracy of some panel segmentation edges is partially inaccurate due to interference from various factors , the shape of the output result is inaccurate, which interferes with the application of subsequent results (area analysis, drawing)
However, the technical problems in the template matching technology in practical applications include: 1) usually the user is required to manually provide a specific template
Template generation methods include: users provide sketches (sketch templates), or more commonly, manually draw from images (image templates); There are differences) are closely related, and image changes may cause the template to fail, so it is often necessary to re-select the template on the new image; 3) Even in the same image, objects on a large-scale remote sensing image are often affected by camera angles or placement positions. There are deformations due to other reasons, or the difference in tone due to exposure, etc., which will bring difficulties to the template setting
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0046] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.
[0047] Such as image 3 As shown, the automatic generation of the photovoltaic panel template based on the surface line primitive association model disclosed in the embodiment of the present invention mainly includes the following steps:
[0048] Step 1: Perform image segmentation and line detection on the image to obtain surface and line primitives. Carry out image segmentation based on hard boundary constraints and two-stage merging on a remote sensing image, and phase grouping line detection to obtain surface (segmentation patch) and line (edge line) primitives. Specifically include:
[0049] (1) Image segmentation with hard boundary constraints and two-stage merging
[0050] Firstly, for a multi-spectral remote sensing image, the watershed segmentation and edge allocation under Canny edge constraints are performed to obtain sub-units, ...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a method for automatically generating photovoltaic panel templates based on a region-line primitive association framework (RLPAF). The method mainly includes the steps of firstperforming remote sensing image segmentation and line detection to obtain region and line primitives; then analyzing spectral and shape features of photovoltaic panels, and setting a rule set for photovoltaic panel extraction to perform preliminary extraction of photovoltaic panels; and on this basis, combining the features of the region and line primitives, modeling a region-line association relationship of suspected photovoltaic panel primitive locations, next, extracting a photovoltaic panel templates using a region-line primitive optimal goodness of fit index to generate a preliminary panel template set, finally, performing Gaussian distribution modeling of areas of the preliminarily generated template set, and removing templates with an area outlier template to obtain a standard template set. Based on the RLPAF, the invention proposes the concept of the region-line primitive optimal goodness of fit index, templates can be automatically generated, and the morphological accuracy defects of extraction of photovoltaic panel targets by the conventional OBIA of first segmentation and then classification can be avoided.
Description
technical field [0001] The present invention relates to a method for target recognition in remote sensing images, specifically a method for automatically generating templates for photovoltaic panels based on surface-line primitive association models, which is applied to photovoltaic panels on high-resolution remote sensing images based on template matching Automatic and accurate extraction belongs to the field of remote sensing image processing and target recognition. Background technique [0002] Object-based image analysis (OBIA), which generally adopts the technical framework of remote sensing image analysis that first divides and then classifies, has been considered as a "paradigm technology" for realizing information extraction of medium and high-resolution remote sensing images. Compared with traditional For Pixel-based image analysis (PBIA) technology, OBIA has a stronger technical performance than PBIA due to its extremely rich features. Existing OBIAs generally use...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & AuthorityPatents(China)
IPC IPC(8): G06T7/11G06T7/13G06T7/136G06T7/187G06T7/60
CPCG06T7/11G06T7/13G06T7/136G06T7/187G06T7/60G06T2207/10032
Inventor汪闽孙宇颉崔齐
OwnerNANJING NORMAL UNIVERSITY