arrow
Volume 33, Issue 3
Saturation-Value Blind Color Image Deblurring with Geometric Spatial-Feature Prior

Hao Zhang, Yingying Fang, Hok Shing Wong, Lihua Li & Tieyong Zeng

Commun. Comput. Phys., 33 (2023), pp. 795-823.

Published online: 2023-04

Export citation
  • Abstract

Blind deblurring for color images has long been a challenging computer vision task. The intrinsic color structures within image channels have typically been disregarded in many excellent works. We investigate employing regularizations in the hue, saturation, and value (HSV) color space via the quaternion framework in order to better retain the internal relationship among the multiple channels and reduce color distortions and color artifacts. We observe that a geometric spatial-feature prior utilized in the intermediate latent image successfully enhances the kernel accuracy for the blind deblurring variational models, preserving the salient edges while decreasing the unfavorable structures. Motivated by this, we develop a saturation-value geometric spatial-feature prior in the HSV color space via the quaternion framework for blind color image deblurring, which facilitates blur kernel estimation. An alternating optimization strategy combined with a primal-dual projected gradient method can effectively solve this novel proposed model. Extensive experimental results show that our model outperforms state-of-the-art methods in blind color image deblurring by a wide margin, demonstrating the effectiveness of the proposed model.

  • AMS Subject Headings

65J22, 65K10, 65T50

  • Copyright

COPYRIGHT: © Global Science Press

  • Email address
  • BibTex
  • RIS
  • TXT
@Article{CiCP-33-795, author = {Zhang , HaoFang , YingyingWong , Hok ShingLi , Lihua and Zeng , Tieyong}, title = {Saturation-Value Blind Color Image Deblurring with Geometric Spatial-Feature Prior}, journal = {Communications in Computational Physics}, year = {2023}, volume = {33}, number = {3}, pages = {795--823}, abstract = {

Blind deblurring for color images has long been a challenging computer vision task. The intrinsic color structures within image channels have typically been disregarded in many excellent works. We investigate employing regularizations in the hue, saturation, and value (HSV) color space via the quaternion framework in order to better retain the internal relationship among the multiple channels and reduce color distortions and color artifacts. We observe that a geometric spatial-feature prior utilized in the intermediate latent image successfully enhances the kernel accuracy for the blind deblurring variational models, preserving the salient edges while decreasing the unfavorable structures. Motivated by this, we develop a saturation-value geometric spatial-feature prior in the HSV color space via the quaternion framework for blind color image deblurring, which facilitates blur kernel estimation. An alternating optimization strategy combined with a primal-dual projected gradient method can effectively solve this novel proposed model. Extensive experimental results show that our model outperforms state-of-the-art methods in blind color image deblurring by a wide margin, demonstrating the effectiveness of the proposed model.

}, issn = {1991-7120}, doi = {https://doi.org/10.4208/cicp.OA-2022-0226}, url = {http://global-sci.org/intro/article_detail/cicp/21660.html} }
TY - JOUR T1 - Saturation-Value Blind Color Image Deblurring with Geometric Spatial-Feature Prior AU - Zhang , Hao AU - Fang , Yingying AU - Wong , Hok Shing AU - Li , Lihua AU - Zeng , Tieyong JO - Communications in Computational Physics VL - 3 SP - 795 EP - 823 PY - 2023 DA - 2023/04 SN - 33 DO - http://doi.org/10.4208/cicp.OA-2022-0226 UR - https://global-sci.org/intro/article_detail/cicp/21660.html KW - Blind color image deblurring, quaternion, geometric spatial-feature prior, color space. AB -

Blind deblurring for color images has long been a challenging computer vision task. The intrinsic color structures within image channels have typically been disregarded in many excellent works. We investigate employing regularizations in the hue, saturation, and value (HSV) color space via the quaternion framework in order to better retain the internal relationship among the multiple channels and reduce color distortions and color artifacts. We observe that a geometric spatial-feature prior utilized in the intermediate latent image successfully enhances the kernel accuracy for the blind deblurring variational models, preserving the salient edges while decreasing the unfavorable structures. Motivated by this, we develop a saturation-value geometric spatial-feature prior in the HSV color space via the quaternion framework for blind color image deblurring, which facilitates blur kernel estimation. An alternating optimization strategy combined with a primal-dual projected gradient method can effectively solve this novel proposed model. Extensive experimental results show that our model outperforms state-of-the-art methods in blind color image deblurring by a wide margin, demonstrating the effectiveness of the proposed model.

Hao Zhang, Yingying Fang, Hok Shing Wong, Lihua Li & Tieyong Zeng. (2023). Saturation-Value Blind Color Image Deblurring with Geometric Spatial-Feature Prior. Communications in Computational Physics. 33 (3). 795-823. doi:10.4208/cicp.OA-2022-0226
Copy to clipboard
The citation has been copied to your clipboard