中值滤波算法是一种常见的图像去噪算法,其性能受滤波器窗口大小的影响较大,因此提出一种能够根据噪声点自动调节滤波窗口大小的改进中值滤波算法,可以有效克服中值滤波不能调整窗口大小的缺点,避免图像边缘模糊。通过利用Matlab进行仿...中值滤波算法是一种常见的图像去噪算法,其性能受滤波器窗口大小的影响较大,因此提出一种能够根据噪声点自动调节滤波窗口大小的改进中值滤波算法,可以有效克服中值滤波不能调整窗口大小的缺点,避免图像边缘模糊。通过利用Matlab进行仿真实验,将所提算法与中值滤波和维纳滤波算法进行对比,并计算峰值信噪比(Peak Signal to Noise Ratio,PSNR)来分析去噪结果。实验表明,与中值滤波和维纳滤波相比,改进中值滤波算法能够更加快速准确地去除图像中的噪声。因此,该改进中值滤波算法是一种有效的图像去噪算法。展开更多
We measure the surface roughness of the mechanical parts based on digital holography. A digital off- axis hologram recording setup for reflective samples is built. Firstly, the height reconstruction error 2.3% of the ...We measure the surface roughness of the mechanical parts based on digital holography. A digital off- axis hologram recording setup for reflective samples is built. Firstly, the height reconstruction error 2.3% of the setup is calibrated by using the quartz step height standard (VLSI- SHS-880QC). Then, the standard scribed-line model and the grinding roughness specimen are selected as the test samples and their surface roughness are 0.095 6 jam and 0.025 3 jam, with errors 6.3%, 0.9%, respectively. The results are in good agreement with the given roughness parameters. At last, we also analyze the window effect of the filter on the roughness measurement value based on digital holography. In conclu- sion, the paper demonstrated effectively that the digital holography could provide the surface feature for the rough- ness measurement with high accuracy.展开更多
文摘中值滤波算法是一种常见的图像去噪算法,其性能受滤波器窗口大小的影响较大,因此提出一种能够根据噪声点自动调节滤波窗口大小的改进中值滤波算法,可以有效克服中值滤波不能调整窗口大小的缺点,避免图像边缘模糊。通过利用Matlab进行仿真实验,将所提算法与中值滤波和维纳滤波算法进行对比,并计算峰值信噪比(Peak Signal to Noise Ratio,PSNR)来分析去噪结果。实验表明,与中值滤波和维纳滤波相比,改进中值滤波算法能够更加快速准确地去除图像中的噪声。因此,该改进中值滤波算法是一种有效的图像去噪算法。
基金The Young Scientists Fund of the Natural Science Foundation of China(Grant No. 61107004)
文摘We measure the surface roughness of the mechanical parts based on digital holography. A digital off- axis hologram recording setup for reflective samples is built. Firstly, the height reconstruction error 2.3% of the setup is calibrated by using the quartz step height standard (VLSI- SHS-880QC). Then, the standard scribed-line model and the grinding roughness specimen are selected as the test samples and their surface roughness are 0.095 6 jam and 0.025 3 jam, with errors 6.3%, 0.9%, respectively. The results are in good agreement with the given roughness parameters. At last, we also analyze the window effect of the filter on the roughness measurement value based on digital holography. In conclu- sion, the paper demonstrated effectively that the digital holography could provide the surface feature for the rough- ness measurement with high accuracy.