ID |
原文 |
译文 |
26055 |
其次,边界复原子网修复阴影边界实现友好的过渡; |
Thirdly, the shadow boundary are naturally recovered by the boundary completion sub-net. |
26056 |
最后,使用自适应衰减因子引导图像进行细节恢复以得到纹理丰富的结果。 |
Finally, the shadow removal result is obtained using a detail recovering method guided by adaptive attenuation factor. |
26057 |
实验结果表明所提方法可以有效地提高阴影去除效果。 |
Experimental results show that the proposed method can improve the removal performance effectively. |
26058 |
针对热轧带钢表面缺陷检测中存在的检测速度慢、检测精度低等问题,提出了一种改进的 YOLOv3 算法模型。 |
To solve the problem of slow speed and low accuracy in the surface defect detection of hot rolled strips, an improved YOLOv3 algorithm is proposed. |
26059 |
使用加权 K-means 聚类算法来优化确定先验框参数,提高先验框(priors anchor)与特征图层(feature map)的匹配度; |
Firstly, the weighting K-means clustering algorithm is put forward to optimize priors anchor's parameters, which can improve the match between priors anchor and feature map. |
26060 |
同时,调整 YOLOv3 算法的网络结构,融合浅层特征与深层特征,形成新的大尺度检测图层,提高网络对带钢表面缺陷的检测精度。 |
Secondly, the improved net-work structure of the YOLOv3 algorithm is proposed to improve the detection accuracy, whose shallow features and deep features are combined to form the new large-scale inspection layer. |
26061 |
实验结果表明,改进后的 YOLOv3 算法在 NEU-DET 数据集上平均精度均值达到了 80% ,较原有的YOLOv3 算法提高了 11% ; |
The experiments are carried out on the NEU-DET data-set, the results show that the average accuracy of the improved YOLOv3 algorithm is 80% , which is 11% higher than that ofthe original algorithm; |
26062 |
同时检测速度保持在 50fps,优于目前其它深度学习带钢表面缺陷检测算法。 |
the detection speed is 50fps, which is faster than other strip surface defect detection algorithms based on deep learning. |
26063 |
提出了基于贝塞尔波束的大景深毫米波介质透镜天线的设计方法。 |
In this study, we propose an approach to design large depth of field (DOF) millimeter wave (MMW) lens antenna based on Bessel beam. |
26064 |
利用轴锥镜生成贝塞尔波束,并根据成像要求分别设计了基于贝塞尔波束的毫米波介质透镜天线和基于高斯波束的毫米波介质透镜天线。 |
The Bessel beam is generated by using axicon. And the MMW dielectric lens antenna basedon Bessel beam or Gaussian beam are designed respectively to satisfy the requirement of imaging. |