Please use this identifier to cite or link to this item: http://repositorio.ufla.br/jspui/handle/1/49828
Title: Light-field imaging reconstruction using deep learning enabling intelligent autonomous transportation system
Keywords: Light-field imaging
Deep learning framework
Low computational complexity
Autonomous transportation systems
Intelligent transportation systems
Aprendizado profundo
Baixa complexidade computacional
Sistemas autônomos de transporte
Sistemas de transportes inteligentes
Issue Date: May-2021
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Citation: SILVA, J. C. et al. Light-field imaging reconstruction using deep learning enabling intelligent autonomous transportation system. IEEE Transactions on Intelligent Transportation Systems, [S.I.], v. 23, n. 2, p. 1587-1595, Feb. 2022. DOI: 10.1109/TITS.2021.3079644.
Abstract: Light-field (LF) cameras, also known as plenoptic cameras, permit the recording of the 4D LF distribution of target scenes. However, many times, surface errors of a microlens array (MLA) are responsible for degradation in the images captured by a plenoptic camera. Additionally, the limited pixel count of the sensor can cause missing parallax information. The aforementioned issues are crucial for creating accurate maps for Intelligent Autonomous Transport System (IATS), because they cause loss of LF information, and need to be addressed. To tackle this problem, a learning-based framework by directly simulating the LF distribution is proposed. A high-dimensional convolution layer with densely sampled LFs in 4D space and considering a soft activation function based on ReLU segmentation correction is used to generate a superresolution (SR) LF image, improving the convergence rate in the deep learning network. Experimental results show that our proposed LF image reconstruction framework outperforms the existing state-of-the-art approaches; specifically, it is effective for learning the LF distribution and generating high-quality LF images. Different image quality assessment methods are used to evaluate the performance of the proposed framework, such as PSNR, SSIM, IWSSIM, FSIM, GFM, MDFM, and HDR-VDP. Additionally, the computational efficiency was evaluated in terms of number of parameters and FLOPs, and experimental results demonstrated that our proposed framework reached the highest performance in most of the datasets used.
URI: https://ieeexplore.ieee.org/document/9442895
http://repositorio.ufla.br/jspui/handle/1/49828
Appears in Collections:DCC - Artigos publicados em periódicos

Files in This Item:
There are no files associated with this item.


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Admin Tools