CN-122023845-A - Photovoltaic station construction progress identification method and device based on unmanned aerial vehicle inspection

CN122023845ACN 122023845 ACN122023845 ACN 122023845ACN-122023845-A

Abstract

The invention relates to the technical field of construction of photovoltaic stations and discloses a photovoltaic station construction progress identification method and device based on unmanned aerial vehicle inspection, wherein the method comprises the steps of utilizing unmanned aerial vehicle inspection to collect an orthographic image of a target photovoltaic station, and carrying out subarray division on the orthographic image of the target photovoltaic station to obtain a plurality of subarray images; the method comprises the steps of carrying out target segmentation on each subarray image to obtain a photovoltaic group string corresponding to each subarray image, carrying out image recognition on each photovoltaic group string to obtain the overall quantity of the photovoltaic panels, and comparing the overall quantity of the photovoltaic panels with the planned quantity of drawings of the photovoltaic panels to obtain the construction progress of the photovoltaic station. The invention realizes accurate and efficient identification and statistics of the construction progress of the photovoltaic station.

Inventors

JIANG JUN
Ge Peijuan
ZOU YIBIN
XU TAO
WANG YANG
CHENG HAIFENG
XIAO JIANDONG
WU DI
ZHANG LING
XU MINGXI
WANG CHENYUE

Assignees

上海勘测设计研究院有限公司

Dates

Publication Date: 20260512
Application Date: 20260127

Claims (10)

1. The utility model provides a photovoltaic station construction progress recognition method based on unmanned aerial vehicle inspection, which is characterized in that the method includes: acquiring an orthographic image of a target photovoltaic field station by using unmanned aerial vehicle inspection, and carrying out subarray division on the orthographic image of the target photovoltaic field station to obtain a plurality of subarray images; performing target segmentation on each subarray image to obtain a photovoltaic group string corresponding to each subarray image; carrying out image recognition on each photovoltaic group string to obtain the overall quantity of the photovoltaic panels; And comparing the overall quantity of the photovoltaic panels with the number of drawing plans of the photovoltaic panels to obtain the construction progress of the photovoltaic station.
2. The method of claim 1, wherein the acquiring the orthographic image of the target photovoltaic field station with the drone inspection, and the subarray dividing the orthographic image of the target photovoltaic field station to obtain a plurality of subarray images, comprises: Performing image preprocessing on the orthographic image of the target photovoltaic field station to obtain a preprocessed image; And carrying out image subarray cutting on the preprocessed image to obtain a plurality of subarray images.
3. The method of claim 1, wherein the performing the target segmentation on each of the subarray images to obtain the photovoltaic group string corresponding to each of the subarray images comprises: extracting local features of the subarray images to obtain multi-scale local features; Global feature enhancement is carried out on the multi-scale local features to obtain enhanced multi-scale features; carrying out multi-scale feature fusion on the enhanced multi-scale features to obtain a fusion feature map; based on the fusion feature map, a detection frame and a segmentation mask of the photovoltaic group string are respectively generated by using a detection head and a segmentation head; And carrying out post-processing on the detection frame and the segmentation mask of the photovoltaic group string to obtain the photovoltaic group string corresponding to each subarray image.
4. A method according to claim 3, wherein the performing global feature enhancement on the multi-scale local feature results in an enhanced multi-scale feature, comprising: Carrying out feature serialization treatment on the multi-scale local features to obtain a feature total sequence; Coding the characteristic total sequence to obtain a coded characteristic sequence; And carrying out feature reconstruction and residual fusion on the coded feature sequence to obtain the enhanced multi-scale feature.
5. The method of claim 1, wherein said performing image recognition on each of said strings of photovoltaic groups results in a global number of photovoltaic panels, comprising: performing image preprocessing on each photovoltaic group string to obtain preprocessed images; carrying out morphological operation denoising on the preprocessed image to obtain a denoised image; performing edge detection on the denoised image to obtain the edge profile of the photovoltaic panel; traversing the edge outline of the photovoltaic panel, and screening and identifying the edge outline of the photovoltaic panel to obtain the global quantity of the photovoltaic panels.
6. The method of claim 1, wherein comparing the global number of photovoltaic panels to the planned number of drawings of photovoltaic panels to obtain a photovoltaic station construction progress comprises: acquiring actual geographic coordinate information of a photovoltaic group string of a drawing plan based on the orthographic image of the target photovoltaic station and the number of the drawing plans; Acquiring geographic coordinate information of the identified photovoltaic group strings based on the global quantity of the photovoltaic panels; And carrying out coordinate matching on the actual geographic coordinate information of the photovoltaic string of the drawing plan and the geographic coordinate information of the identified photovoltaic string, and determining the construction progress of the photovoltaic station based on a coordinate matching result.
7. Photovoltaic station construction progress recognition device based on unmanned aerial vehicle patrols and examines, its characterized in that, the device includes: the subarray division module is used for acquiring an orthographic image of a target photovoltaic field station by using unmanned aerial vehicle inspection, and subarray division is carried out on the orthographic image of the target photovoltaic field station to obtain a plurality of subarray images; The target segmentation module is used for carrying out target segmentation on each subarray image to obtain a photovoltaic group string corresponding to each subarray image; the image recognition module is used for carrying out image recognition on each photovoltaic group string to obtain the overall quantity of the photovoltaic panels; and the comparison module is used for comparing the overall quantity of the photovoltaic panels with the number of drawing plans of the photovoltaic panels to obtain the construction progress of the photovoltaic station.
8. An electronic device, comprising: A memory and a processor, said memory and said processor being communicatively coupled to each other, said memory having stored therein computer instructions, the processor executes the computer instructions to perform the unmanned aerial vehicle inspection-based photovoltaic yard construction progress recognition method of any one of claims 1 to 6.
9. A computer-readable storage medium, wherein computer instructions for causing a computer to execute the photovoltaic station construction progress recognition method based on unmanned aerial vehicle inspection according to any one of claims 1 to 6 are stored on the computer-readable storage medium.
10. A computer program product comprising computer instructions for causing a computer to perform the unmanned aerial vehicle inspection-based photovoltaic yard construction progress identification method of any of claims 1 to 6.

Description

Photovoltaic station construction progress identification method and device based on unmanned aerial vehicle inspection Technical Field The invention relates to the technical field of photovoltaic station construction, in particular to a photovoltaic station construction progress identification method and device based on unmanned aerial vehicle inspection. Background In the construction process of the photovoltaic station, accurate statistics of construction progress is a key for ensuring that engineering is smoothly carried out and settled. The related photovoltaic station construction progress statistical method mainly depends on manual field investigation, and the method not only consumes a great deal of manpower and time, but also has low efficiency, is easily influenced by artificial and environmental factors, and causes errors in statistical results. In addition, the photovoltaic station generally occupies a large area, and the manual investigation is difficult to comprehensively and timely master the construction condition of the whole station. Disclosure of Invention The invention provides a photovoltaic station construction progress identification method and device based on unmanned aerial vehicle inspection, which are used for solving the problems of low efficiency and low precision of photovoltaic station construction progress statistics. In a first aspect, the invention provides a photovoltaic station construction progress identification method based on unmanned aerial vehicle inspection, which comprises the following steps: Acquiring an orthographic image of a target photovoltaic field station by using unmanned aerial vehicle inspection, and carrying out subarray division on the orthographic image of the target photovoltaic field station to obtain a plurality of subarray images; performing target segmentation on each subarray image to obtain a photovoltaic group string corresponding to each subarray image; carrying out image recognition on each photovoltaic group string to obtain the overall quantity of the photovoltaic panels; And comparing the overall quantity of the photovoltaic panels with the planned quantity of the drawings of the photovoltaic panels to obtain the construction progress of the photovoltaic station. According to the photovoltaic station construction progress identification method based on unmanned aerial vehicle inspection, the orthographic image of the target photovoltaic station is acquired through unmanned aerial vehicle inspection, and technologies such as image subarray segmentation, target segmentation and image identification are combined, so that the number of photovoltaic panels in the photovoltaic group string can be accurately identified and calculated, and further, compared with the number of drawing plans, the construction progress identification is realized. In an alternative embodiment, the method for acquiring the orthographic image of the target photovoltaic field station by using the unmanned aerial vehicle for inspection, and sub-array dividing the orthographic image of the target photovoltaic field station to obtain a plurality of sub-array images includes: Performing image preprocessing on the orthographic image of the target photovoltaic field station to obtain a preprocessed image; and carrying out image subarray cutting on the preprocessed image to obtain a plurality of subarray images. According to the photovoltaic station construction progress identification method based on unmanned aerial vehicle inspection, the large-size aerial photo of unmanned aerial vehicle inspection is split into the small-size subarrays, so that the calculated amount of a single image is greatly reduced, and the rapid inspection of a large-scale photovoltaic power station can be realized. In an alternative embodiment, performing object segmentation on each subarray image to obtain a photovoltaic group string corresponding to each subarray image, including: extracting local features of the subarray images to obtain multi-scale local features; global feature enhancement is carried out on the multi-scale local features to obtain enhanced multi-scale features; carrying out multi-scale feature fusion on the enhanced multi-scale features to obtain a fusion feature map; based on the fusion feature map, respectively utilizing a detection head and a segmentation head to generate a detection frame and a segmentation mask of the photovoltaic group string; And carrying out post-processing on the detection frame and the segmentation mask of the photovoltaic group string to obtain the photovoltaic group string corresponding to each sub-array image. According to the photovoltaic station construction progress identification method based on unmanned aerial vehicle inspection, after the local features of the sub-array images are extracted, the global features of the multi-scale local features are enhanced, and then the enhanced multi-scale features are subjected to multi-scale feature fusio