As a promising technique to enhance the spatial reso- lution of remote sensing imagery, sub-pixel mapping is processed based on the spatial dependence theory with the assumption that the land cover is spatially depend...As a promising technique to enhance the spatial reso- lution of remote sensing imagery, sub-pixel mapping is processed based on the spatial dependence theory with the assumption that the land cover is spatially dependent both within pixels and be- tween them. The spatial attraction is used as a tool to describe the dependence. First, the spatial attractions between pixels, sub- pixel/pixel spatial attraction model （SPSAM）, are described by the modified SPSAM （MSPSAM） that estimates the attractions accord- ing to the distribution of sub-pixels within neighboring pixels. Then a mixed spatial attraction model （MSAM） for sub-pixel mapping is proposed that integrates the spatial attractions both within pix- els and between them. According to the expression of the MSAM maximumising the spatial attraction, the genetic algorithm is em- ployed to search the optimum solution and generate the sub-pixel mapping results. Experiments show that compared with SPSAM, MSPSAM and pixel swapping algorithm modified by initialization from SPSAM （MPS）, MSAM can provide higher accuracy and more rational sub-pixel mapping results.展开更多
针对露天矿生产场景中存在着目标像素低、小目标众多、背景复杂等问题,在YOLOv5s的基础上提出一种多尺度和超分辨率网络(multiscale and super-resolution network,MS_Net)。在特征融合模块,将PANet的三尺度检测升级为四尺度检测,提高...针对露天矿生产场景中存在着目标像素低、小目标众多、背景复杂等问题,在YOLOv5s的基础上提出一种多尺度和超分辨率网络(multiscale and super-resolution network,MS_Net)。在特征融合模块,将PANet的三尺度检测升级为四尺度检测,提高网络的多尺度学习能力,并使用子像素卷积作为上采样方法;提出一种多层融合(multi layer fusion,MLF)模块,融合了PANet 3个输出层的特征,得到一个具有丰富语义信息和空间信息的特征图;在预测层中,使用SIoU作为定位损失函数,优化模型的参数。实验结果表明:MS_Net网络在PASCALVOC数据集上mAP为79.4%,FPS为59;在矿山数据集上mAP为80.2%,FPS为64.5,模型可快速、准确、高效地对露天矿中的目标进行识别检测。展开更多
基金supported by the National Natural Science Foundation of China (60802059)the Foundation for the Doctoral Program of Higher Education of China (200802171003)
文摘As a promising technique to enhance the spatial reso- lution of remote sensing imagery, sub-pixel mapping is processed based on the spatial dependence theory with the assumption that the land cover is spatially dependent both within pixels and be- tween them. The spatial attraction is used as a tool to describe the dependence. First, the spatial attractions between pixels, sub- pixel/pixel spatial attraction model （SPSAM）, are described by the modified SPSAM （MSPSAM） that estimates the attractions accord- ing to the distribution of sub-pixels within neighboring pixels. Then a mixed spatial attraction model （MSAM） for sub-pixel mapping is proposed that integrates the spatial attractions both within pix- els and between them. According to the expression of the MSAM maximumising the spatial attraction, the genetic algorithm is em- ployed to search the optimum solution and generate the sub-pixel mapping results. Experiments show that compared with SPSAM, MSPSAM and pixel swapping algorithm modified by initialization from SPSAM （MPS）, MSAM can provide higher accuracy and more rational sub-pixel mapping results.
