针对海上低纹理、大视差图像拼接过程中由于误对齐产生的伪影问题,提出一种基于点线特征配准和最佳接缝线融合的几何结构保视差图像拼接算法。在传统的基于点特征求解单应变换的基础上引入线段特征,并将潜在的共面局部线段合并为全局线...针对海上低纹理、大视差图像拼接过程中由于误对齐产生的伪影问题,提出一种基于点线特征配准和最佳接缝线融合的几何结构保视差图像拼接算法。在传统的基于点特征求解单应变换的基础上引入线段特征,并将潜在的共面局部线段合并为全局线段,为接缝线融合提供精准对齐的条件;在图像融合过程中,利用tanh度量的颜色差异和梯度差异以及引入显著性检测权重来设计接缝切割方法的能量函数,引导最佳接缝线避开图像中显著海上结构,从而确保结构边缘的连续性;使用SIFT flow方法校正拼接缝上的错位像素,实现几何结构准确的海上图像拼接。在20对不同场景数据上的拼接实验表明,与基准方法相比,所提算法的基于SSIM的接缝质量误差平均降低了44.6%,最多降低66.7%,基于ZNCC(zero mean normalized cross-correlation)的接缝质量误差平均降低了24.7%,最多降低了51.6%,能够有效地避免伪影,从而得到观感自然的拼接结果,满足无人船航行过程中对宽视场的需求。展开更多
The development of welding robots suitable for specially unstructured working enviroments has been become an important development direction of industrial robot application because large-scale welding structures have ...The development of welding robots suitable for specially unstructured working enviroments has been become an important development direction of industrial robot application because large-scale welding structures have been used more and more widely in modern industry. In this paper, an intelligent mobile robot for welding of ship deck with the function of autosearching weld line was presented. A wheeled motion mechanism and a cross adjustment slider are used for the welding robot body. A sensing system based on laser-PSD （position sensitive detector） displacement sensor was developed to obtain two dimensional deviation signals during seam tracking. A full-digital control system based on DSP and CPLD has also been realized to implement complex and high-performance control algorithms. Furthermore, the system has still the function of auto-searching weld line according to the characteristics information of weld groove and adjusting posture itself to the desired status preparing for welding. The experiment of auto-searching welding line shows that the robot has high tracing accuracy, and can satisfy the requirement of practical welding project.展开更多
文摘针对海上低纹理、大视差图像拼接过程中由于误对齐产生的伪影问题,提出一种基于点线特征配准和最佳接缝线融合的几何结构保视差图像拼接算法。在传统的基于点特征求解单应变换的基础上引入线段特征,并将潜在的共面局部线段合并为全局线段,为接缝线融合提供精准对齐的条件;在图像融合过程中,利用tanh度量的颜色差异和梯度差异以及引入显著性检测权重来设计接缝切割方法的能量函数,引导最佳接缝线避开图像中显著海上结构,从而确保结构边缘的连续性;使用SIFT flow方法校正拼接缝上的错位像素,实现几何结构准确的海上图像拼接。在20对不同场景数据上的拼接实验表明,与基准方法相比,所提算法的基于SSIM的接缝质量误差平均降低了44.6%,最多降低66.7%,基于ZNCC(zero mean normalized cross-correlation)的接缝质量误差平均降低了24.7%,最多降低了51.6%,能够有效地避免伪影,从而得到观感自然的拼接结果,满足无人船航行过程中对宽视场的需求。
文摘The development of welding robots suitable for specially unstructured working enviroments has been become an important development direction of industrial robot application because large-scale welding structures have been used more and more widely in modern industry. In this paper, an intelligent mobile robot for welding of ship deck with the function of autosearching weld line was presented. A wheeled motion mechanism and a cross adjustment slider are used for the welding robot body. A sensing system based on laser-PSD （position sensitive detector） displacement sensor was developed to obtain two dimensional deviation signals during seam tracking. A full-digital control system based on DSP and CPLD has also been realized to implement complex and high-performance control algorithms. Furthermore, the system has still the function of auto-searching weld line according to the characteristics information of weld groove and adjusting posture itself to the desired status preparing for welding. The experiment of auto-searching welding line shows that the robot has high tracing accuracy, and can satisfy the requirement of practical welding project.
