PanicleNeRF：利用智能手机实现低成本、高精度的水稻穗田间表型分析
PanicleNeRF: low-cost, high-precision in-field phenotypingof rice panicles with smartphone
AI总结：
提出 PanicleNeRF，结合 SAM、YOLOv8 与 NeRF，用智能手机实现田间水稻穗三维重建和性状提取，分割、点云及产量相关指标表现优异。
摘要：
水稻穗性状显著影响籽粒产量，因此成为水稻表型研究的主要目标。然而，现有大多数技术局限于受控的室内环境，难以捕捉自然生长条件下的水稻穗性状。为此，我们开发了 PanicleNeRF，这是一种利用智能手机在田间高精度、低成本重建水稻穗三维（3D）模型的新方法。所提出的方法结合了大模型 Segment Anything Model（SAM）和小模型 You Only Look Once version 8（YOLOv8），以实现水稻穗图像的高精度分割。随后，利用带有二维分割结果的图像，采用 NeRF 技术进行三维重建。最后，对生成的点云进行处理，成功提取稻穗性状。结果表明，PanicleNeRF 有效完成了二维图像分割任务，平均 F1 Score 达到 86.9%，平均 Intersection over Union（IoU）达到 79.8%，并且边界重叠（BO）性能相比 YOLOv8 几乎提升一倍。在点云质量方面，PanicleNeRF 显著优于传统的 SfM-MVS（structure-from-motion and multi-view stereo，运动恢复结构与多视图立体）方法，例如 COLMAP 和 Metashape。随后准确提取了稻穗长度，其中籼稻的相对均方根误差（rRMSE）为 2.94%，粳稻为 1.75%。由三维点云估算的稻穗体积与籽粒数（籼稻 R²=0.85，粳稻 R²=0.82）和籽粒质量（籼稻 0.80，粳稻 0.76）具有很强的相关性。该方法为高通量水稻穗田间表型分析提供了一种低成本解决方案，可提高水稻育种效率。
英文摘要：
The rice panicle traits significantly influence grain yield, making them a primary target for rice phenotyping studies. However, most existing techniques are limited to controlled indoor environments and difficult to capture the rice panicle traits under natural growth conditions. Here, we developed PanicleNeRF, a novel method that enables high-precision and low-cost reconstruction of rice panicle three-dimensional (3D) models in the field using smartphone. The proposed method combined the large model Segment Anything Model (SAM) and the small model You Only Look Once version 8 (YOLOv8) to achieve high-precision segmentation of rice panicle images. The NeRF technique was then employed for 3D reconstruction using the images with 2D segmentation. Finally, the resulting point clouds are processed to successfully extract panicle traits. The results show that PanicleNeRF effectively addressed the 2D image segmentation task, achieving a mean F1 Score of 86.9% and a mean Intersection over Union (IoU) of 79.8%, with nearly double the boundary overlap (BO) performance compared to YOLOv8. As for point cloud quality, PanicleNeRF significantly outperformed traditional SfM-MVS (structure-from-motion and multi-view stereo) methods, such as COLMAP and Metashape. The panicle length was then accurately extracted with the rRMSE of 2.94% for indica and 1.75% for japonica rice. The panicle volume estimated from 3D point clouds strongly correlated with the grain number (R2 = 0.85 for indica and 0.82 for japonica) and grain mass (0.80 for indica and 0.76 for japonica). This method provides a low-cost solution for high-throughput in-field phenotyping of rice panicles, accelerating the efficiency of rice breeding.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
