基于人工智能的多源遥感数据融合在电网勘测应用研究

Application of multi-source remote sensing data fusion based on artificial intelligence in power grid survey

随着卫星遥感、空间科技等技术的不断发展,通过卫星采集多源遥感数据已被广泛应用到诸多行业。将多源数据进行融合,生成信息更丰富、质量更高的图像,能更清晰地分析地物地貌地形情况,已成为近年来图像智能处理领域的技术研究前沿方向。文章基于人工智能卷积网络和注意力机制,提出一种统一融合网络,将不同光谱和空间属性的遥感源数据进行有效融合,生成具有精确光谱信息和清晰空间细节的高分辨率图像,为电网勘测选址选线等业务提供了辅助支撑的有效手段。实验结果表明,文章研究结果比现有典型方法具备更好的融合效果。

With the continuous development of satellite remote sensing,space technology,and other technologies,collecting multi-source remote sensing data with satellites has been widely applied in many industries.The fusion of multi-source data to generate images with richer and higher quality information,and the clearer analysis of terrain and topography have become the forefront of technological research in the field of image intelligent processing in recent years.A unified fusion network based on artificial intelligence convolutional networks and attention mechanisms is proposed,which can effectively fuse remote sensing source data with different spectral and spatial attributes,and generate high-resolution images with accurate spectral information and clear spatial details.It provides an effective means of auxiliary support for power grid survey,site selection,and other business operations.The experimental results indicate that this researsh results has better fusion performance than existing typical methods.