Lucas Ortega - Academia.edu (original) (raw)
Related Authors
University of Electronic Science and Technology of China
Uploads
Papers by Lucas Ortega
Blucher Engineering Proceedings, 2020
Currently, millions of photos are captured daily, and several factors can influence the quality o... more Currently, millions of photos are captured daily, and several factors can influence the quality of an image, causing distortions. Research has shown that there are several ways to remove defects from images. This study aims to comparatively analyze the potential of Deep Learning techniques to improve scanned images with shadow, glare, crumpled paper, external lightning, change of perspective and wave distortion defects. Based on a review of the literature on recent deep learning architectures, we have selected three, which were trained and refined to improve the quality of the images. The results indicate that the nets were able to attenuate and remove some defects. On this basis, these initial experiments demonstrate that deep learning models are promising for the studied defects.
Blucher Engineering Proceedings, 2020
Currently, millions of photos are captured daily, and several factors can influence the quality o... more Currently, millions of photos are captured daily, and several factors can influence the quality of an image, causing distortions. Research has shown that there are several ways to remove defects from images. This study aims to comparatively analyze the potential of Deep Learning techniques to improve scanned images with shadow, glare, crumpled paper, external lightning, change of perspective and wave distortion defects. Based on a review of the literature on recent deep learning architectures, we have selected three, which were trained and refined to improve the quality of the images. The results indicate that the nets were able to attenuate and remove some defects. On this basis, these initial experiments demonstrate that deep learning models are promising for the studied defects.