Image diffusion filtering algorithm combined with variable exponent and convective constraint
A diffusion function based on mixed gradient and variable exponent is built by combining image characteristics in wavelet transform and spatial domains to solve the edge blurring problem of the traditional anisotropic diffusion model in image filtering and improve image filtering performance. A convective term is introduced as a constraint in the image diffusion model to control the strength of image diffusion, and an image adaptive diffusion filtering algorithm is developed. Experimental simulation using test images as research objects is performed to verify the effectiveness of the proposed algorithm. Results show that the proposed algorithm effectively inhibits the edge blurring effect during image diffusion and improves the visual quality of the filtered image.
KeywordsImage filtering Diffusion model Convective constraint Variable exponent Structural similarity
This work was supported by the Science and Technology Foundation of Jiangxi Provincial Education Department (No. GJJ170922) and the National Natural Science Foundation of China under No. 11461057.
- 15.Liu, J.H., She, K.: Image diffusion filtering based on dual tree complex wavelet and wave atoms. Acta Physica Sinica 60(12), 124203-1-10 (2011)Google Scholar
- 19.Dabov, K., Foi, A., Katkovnik, V., Egiazarian, K.: BM3D image denoising with shape-adaptive principal component analysis. In: Proc. Workshop on Signal Processing with Adaptive Sparse Structured Representations (SPARS’09), Saint-Malo (2009). https://www.cs.tut.fi/~foi/papers/SPARS09-BM3D-SAPCA.pdf