An improved two-dimensional variational mode decomposition algorithm and its application in oil pipeline image. Issue 1 (1st January 2020)

• Record Type: Journal Article
• Authors: Gao, Hongyu; Ma, Liyuan; Dong, Hongli; Lu, Jingyi; Li, Gongfa
• Is Part Of: Systems science & control engineering. Volume 8:Issue 1(2020)
• Page Start: 297
• Page End: 307
• Publication Date: 2020-01-01
• Subjects: Two-dimensional variational mode decomposition -- intrinsic mode function -- Hausdorff distance -- image denoising -- wavelet denoising; System theory -- Periodicals; Automatic control -- Periodicals; 003.05
• DOI: 10.1080/21642583.2020.1756523
• Languages: English
• ISSNs: 2164-2583

Abstract

Abstract : In this paper, an image denoising algorithm is presented based on the two-dimensional variational mode decomposition (2D-VMD) and the Hausdorff distance (HD). The procedure of the developed algorithm is that: (1) use the 2D-VMD to decompose the image into a number of intrinsic mode functions (IMFs); (2) use the HD and the probability density function (PDF) to distinguish the signal-dominant IMFs (S-D IMFs) and the noise-dominant IMFs (N-D IMFs), and then use the wavelet threshold denoising method to eliminate the noise in the N-D IMFs; and (3) use the denoised IMFs and the S-D IMFs to reconstruct the image and obtain the denoised image. The effectiveness of the proposed algorithm is verified on the oil pipeline images. Simulation results show that, compared with the median filtering, the wavelet threshold filtering, and the Rudin Osher Fatemi, the proposed algorithm has a better denoising effect in both subjective and objective evaluation.