Spatial Resolution Enhancement for Halftone Images Using Convolutional Neural Networks

Silva Novaes, Guilherme Apolinário; Kim, Hae Yong

doi:10.1109/csci58124.2022.00247

Cited by 1 publication

(3 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We published a preliminary version of this work at a conference [11], but the present work presents several aspects that were not described there. In our previous work, we trained the well-known enhanced deep super resolution (EDSR) model [2] on single-channel halftone images to increase the resolution.…”

Section: B Contributions Of This Workmentioning

confidence: 99%

“…In our previous work [11], we used whole images to train the model without any type of data augmentation. Using a batch size of 2 and full 2000×1124 images, training a single model took approximately 12 hours.…”

Section: E Choosing the Patch And Batch Sizesmentioning

confidence: 99%

“…We used EDSR model with configuration similar to described in [9], [11]. This model consisted of 16 residual blocks, using the Mean Absolute Error (MAE) with a single up-sampling layer for doubling the image resolution.…”

Section: Halftone-edsr In Color Imagesmentioning

confidence: 99%

See 2 more Smart Citations

Super-Resolution of Color Halftone Images Using Convolutional Neural Networks

Novaes,

Kim

2024

IEEE Access

View full text Add to dashboard Cite

In this work, we investigate how to increase the resolution of color halftone images using convolutional neural networks (CNNs). As far as we know, this is the first work that increases resolution of color halftone images using a CNN-based solution. For this task, we first train the well-known Enhanced Deep Super-Resolution (EDSR) network with halftone images to obtain the Halftone-EDSR model. We argue that it is not possible to use conventional data augmentation techniques in this problem, due to the peculiar texture of halftone images. We present cropping as a viable data augmentation technique. Using cropping and image patches as training samples, we substantially speed up the training and get better quality models. We compare the independent channel model (which increases the resolution of each of the CMYK channels independently and then merges them) with the joint channel model (which increases the resolution of all four image channels at once) and conclude that the the latter is superior to the first. We experimentally demonstrate that the proposed Halftone-EDSR is superior to all previous techniques, both for pre-print and post-print halftone images. However, Halftone-EDSR can generate upsampled images with Moiré patterns. To minimize Moiré patterns, we propose a new network model called Halftone-Net. We use the Fast Fourier Transform, followed by a CNN, to detect the strong presence of Moiré patterns in halftone images and demonstrate that Halftone-Net generates fewer images with strong Moiré patterns than Halftone-EDSR.

show abstract

Section: B Contributions Of This Workmentioning

confidence: 99%

Section: E Choosing the Patch And Batch Sizesmentioning

confidence: 99%