基于深度卷积神经网络的激光三维图像重建方法

Laser 3D image reconstruction method based on depth convolution neural network

针对当前的激光三维图像重建方法存在的精度低、效果差,耗时长等难题,为了提高激光三维图像重建效果,提出基于深度卷积神经网络的激光三维图像重建方法。首先采集待重建激光三维图像,采用去噪算法对激光三维图像进行去噪操作,并对去噪处理后的激光三维图像进行增强操作,改善激光三维图像视觉效果,然后采用深度卷积神经网络对激光三维图像进行激光三维图像重建,最后进行了多幅激光三维图像重建仿真测试,结果表明,深度卷积神经网络的激光三维图像重建精度超过93%,重建后激光三维图像质量得到提升,激光三维图像重建时间控制在40 ms以内,可以快速实现激光三维图像重建结果,同时重建后的激光三维图像整体效果要明显优于当前经典重建方法,具有更加广泛的应用前景。

In order to improve the accuracy of three-dimensional image reconstruction based on laser convolution,it is difficult to improve the effect of three-dimensional image reconstruction.Firstly,the laser three-dimensional image to be reconstructed is collected,the filtering algorithm is used to denoise the laser three-dimensional image,and the denoised laser three-dimensional image is enhanced to improve the visual effect of the laser three-dimensional image.Then the deep convolution neural network is used to reconstruct the laser three-dimensional image.Finally,the reconstruction simulation test of multiple laser three-dimensional images is carried out.The results show that,laser 3D image reconstruction accuracy of deep convolution neural network exceeds 93%.The quality of laser 3D image after reconstruction is improved.The laser 3D image reconstruction time is controlled within 40ms,which can quickly realize the laser 3D image reconstruction results.At the same time,the overall effect of laser 3D image after reconstruction is significantly better than the current classical reconstruction methods,and has a wider application prospect.