湖南电力 ›› 2023, Vol. 43 ›› Issue (4): 74-79.doi: 10.3969/j.issn.1008-0198.2023.04.010

• 新技术及应用 • 上一篇    下一篇

航拍光伏组件图像的畸变校正方法研究

王佳文1, 朱永灿1,2, 王帅1, 李科锋1   

  1. 1.西安工程大学电子信息学院,陕西 西安 710048;
    2.西安电气设备互联传感与智能诊断重点实验室,陕西 西安 710048
  • 收稿日期:2023-04-04 修回日期:2023-05-04 出版日期:2023-08-25 发布日期:2023-09-07
  • 通信作者: 王佳文(1998),女,硕士研究生,通信作者,研究方向为电力系统状态监测与故障诊断。
  • 基金资助:
    陕西省创新能力支持项目(2022KJXX-41),西安重点科技项目(2022JH-RGZN-0005)

Research on Distortion Correction Method of Aerial Photovoltaic Module Images

WANG Jiawen1, ZHU Yongcan1,2, WANG Shuai1, LI Kefeng1   

  1. 1. College of Electronics and Information, Xi'an Polytechnic University, Xi'an 710048, China;
    2. Xi'an Key Laboratory of Interconnected Sensing and Intelligent Diagnosis for Electrical Equipment, Xi'an 710048, China
  • Received:2023-04-04 Revised:2023-05-04 Online:2023-08-25 Published:2023-09-07

摘要: 航拍光伏组件图像由于拍摄角度问题经常存在畸变,其中直线、角度和比例等元素都会受到扭曲,从而影响光伏组件的测量、识别和分类等分析过程。为了解决这一问题,研究了一种无需标定的航拍光伏组件图像畸变校正方法。首先,利用特征点提取算法从光伏组件图像中提取出关键的4个特征点,然后使用特征点匹配算法对这些点进行匹配,获取畸变前后的对应关系。其次,利用计算几何的理论,通过对这些对应点进行三角测量和坐标变换,计算畸变校正所需的参数T。最后,根据参数T对图像进行畸变校正,得到校正后的光伏组件图像。该算法不仅提高了图片的视觉校正效果,还提高了数据20%~55%的准确率,在航拍光伏组件图像的异物识别、运维检修等方面有较高的应用价值。

关键词: 光伏组件, 航拍图像, 畸变校正, 后向映射

Abstract: Aerial photography of photovoltaic components often suffers from distortion due to shooting angle issues, which affects the measurement, recognition, and classification of elements like straight lines, angles, and proportions. To address this problem, this paper proposes an undistorted aerial photovoltaic component image correction method that requires no calibration. The approach first extracts four critical feature points from the photovoltaic component image using a feature point extraction algorithm and then matches them using a feature point matching algorithm to obtain their corresponding relationships before and after distortion. Secondly, the theoretical principles of computational geometry are utilized to compute the parameters T necessary for distortion correction through triangulation and coordinate transformation of the corresponding points. Finally, distortion correction is carried out on the image using parameter T to obtain the corrected photovoltaic component image. This algorithm not only improves the visual correction effect of the image but also increases the accuracy of data by 20%~55%.This method has a high application value in the foreign object recognition, operation and maintenance of the aerial photovoltaics module image.

Key words: photovoltaic module, aerial images, distortion correction, backward mapping

中图分类号: