Light field (LF) cameras capture scenes from multiple perspectives, providing comprehensive image information. However, the resulting data often suffers from low image quality, and current research generally exhibits limitations in underutilizing multi-focus characteristics. The insufficient exploitation of deep feature capabilities directly constrains the visual enhancement effects of LF imaging. To address this issues, converting sub-aperture images (SAIs) is proposed into a focal stack (FS) using depth information and analyzing them. Leveraging the varying depth-of-field characteristics of FS, depth maps are generated through specifically designed filters to classify images as focused or defocused. A selective super-resolution (SR) network is then applied to enhance focused images, where a dimensionality stretching strategy is employed to integrate low-resolution FS with degradation maps corresponding to deterioration parameters as input to convolutional neural networks. Experiments were conducted on both public and self-built light field datasets. The results show that the proposed method has an average increase of 0.21% in the image information entropy for the foreground and 0.22% for the background compared to the LF-IINet method, and an average increase of 0.21% for the foreground and 0.28% for the background compared to the Distg-SSR method. In terms of the high-frequency energy of the image, the proposed method has an average increase of 10.78% for the foreground and 10.56% for the background compared to the LF-IINet method, and an average increase of 11.33% for the foreground and 11.02% for the background compared to the Distg-SSR method. Through depth-guided focused enhancement and defocused blurring strategies, this approach achieves a balance between detail enhancement and noise suppression. The framework provides an efficient and reliable solution for advancing LF imaging applications in computational photography.
LIUYanyangshuo, PANJinxiao, PANYihua, et al. Light field photography simulation of three-dimensional scene[J]. Journal of Test and Measurement Technology, 2018, 32(2): 119-124. (in Chinese)
[3]
ZHUH, GUOM, LIH, et al. Revisiting spatio-angular trade-off in light field cameras and extended applications in super-resolution[J]. IEEE Transactions on Visualization and Computer Graphics, 2021, 27(6): 3019-3033.
[4]
NG R, LEVOYM, BRÉDIFM, et al. Light field photography with a hand-held plenoptic camera[D]. Palo Alto: Stanford University, 2005.
[5]
VENKATARAMANK, LELESCUD, DUPARRÉJ, et al. Picam: an ultra-thin high performance monolithic camera array[J]. ACM Transactions on Graphics (TOG), 2013, 32(6): 1-13.
[6]
WANGY, YANGJ, GUOY, et al. Selective light field refocusing for camera arrays using bokeh rendering and superresolution[J]. IEEE Signal Processing Letters, 2019, 26(1): 204-208.
[7]
BISHOPT E, FAVAROP. The light field camera: extended depth of field, aliasing, and superresolution[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(5): 972-986.
[8]
WANGT C, EFROSA A, RAMAMOORTHIR. Depth estimation with occlusion modeling using light-field cameras[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2016, 38(11): 2170-2181.
[9]
WILLIEM, PARKI K, LEEK M. Robust light field depth estimation using occlusion-noise aware data costs[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018, 40(10): 2484-2497.
[10]
JEONH G, PARKJ, CHOEG, et al. Depth from a light field image with learning-based matching costs[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2019, 41(2): 297-310.
[11]
ZHOUM, DINGY, JIY, et al. Shape and reflectance reconstruction using concentric multi-spectral light field[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020, 42(7): 1594-1605.
[12]
WANGT C, CHANDRAKERM, EFROSA A, et al. SVBRDF-invariant shape and reflectance estimation from a light-field camera[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018, 40(3): 740-754.
[13]
TAOM W, SUJ C, WANGT C, et al. Depth estimation and specular removal for glossy surfaces using point and line consistency with light-field cameras[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2016, 38(6): 1155-1169.
[14]
ALPEROVICHA, JOHANNSENO, STRECKEM, et al. Light field intrinsics with a deep encoder-decoder network[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018: 9145-9154.
[15]
KIMC, ZIMMERH, PRITCHY, et al. Scene reconstruction from high spatio-angular resolution light fields[J]. ACM Transactions on Graphics, 2013, 32(4): 1-12.
[16]
LIN, YEJ, JIY, et al. Saliency detection on light field[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(8): 1605-1616.
[17]
SRINIVASANP P, NG R, RAMAMOORTHIR. Light field blind motion deblurring[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017: 2354-2362.
[18]
SCHAMBACHM, HEIZMANNM. A multispectral light field dataset and framework for light field deep learning[J]. IEEE Access, 2020, 8: 193492-193502.
[19]
TAKAHASHIK, KOBAYASHIY, FUJIIT. From focal stack to tensor light-field display[J]. IEEE Transactions on Image Processing, 2018, 27(9): 4571-4584.
[20]
YOONY, JEONH G, YOOD, et al. Learning a deep convolutional network for light-field image super-resolution[C]//2015 IEEE International Conference on Computer Vision Workshop (ICCVW), 2015: 57-65.
[21]
YOONY, JEONH G, YOOD, et al. Light-field image super-resolution using convolutional neural network[J]. IEEE Signal Processing Letters, 2017, 24(6): 848-852.
[22]
WUG, ZHAOM, WANGL, et al. Light field reconstruction using deep convolutional network on EPI[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017: 1638-1646.
[23]
WUG, LIUY, DAIQ, et al. Learning sheared EPI structure for light field reconstruction[J]. IEEE Transactions on Image Processing, 2019, 28(7): 3261-3273.
[24]
WUG, WANGY, LIUY, et al. Spatial-angular attention network for light field reconstruction[J]. IEEE Transactions on Image Processing, 2021, 30: 8999-9013.
[25]
SHINC, JEONH G, YOONY, et al. EPINET: a fully-convolutional neural network using epipolar geometry for depth from light field images[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018: 4748-4757.
[26]
ZHANGS, LINY, SHENGH. Residual networks for light field image super-resolution[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019: 11038-11047.
[27]
LIUG, YUEH, WUJ, et al. Intra-inter view interaction network for light field image super-resolution[J]. IEEE Transactions on Multimedia, 2021, 25: 256-266.
[28]
WANGY, WANGL, WUG, et al. Disentangling light fields for super-resolution and disparity estimation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023, 45(1): 425-443.
[29]
ZHANGK, LIANGJ, VAN GOOLL, et al. Designing a practical degradation model for deep blind image super-resolution[C]//2021 IEEE/CVF International Conference on Computer Vision (ICCV), 2021: 4771-4780.
[30]
LIPPMANNG. Épreuves réversibles donnant la sensation du relief[J]. Journal de Physique Théorique et Appliquée, 1908, 7(1): 821-825.
[31]
LIANGC K, LINT H, WONGB Y, et al. Programmable aperture photography[J]. ACM Transactions on Graphics, 2008, 27(3): 1-10.
RERABEKM, EBRAHIMIT. New light field image Datastet[C]//2016 8th International Conference on Quality of Multimedia Experience (QoMEX), 2016: 1-8.
[34]
MAGGIORIE, TARABALKAY, CHARPIATG, et al. Can semantic labeling methods generalize to any city?The inria aerial image labeling benchmark[C]//2017 IEEE International geoscience and remote sensing symposium (IGARSS), 2017: 3226-3229.