Color Image Demosaicking Using a 3-Stage Convolutional Neural Network Structure

Cui, Kai; Jin, Zhi; Steinbach, Eckehard

doi:10.1109/icip.2018.8451020

Cited by 36 publications

(41 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The proposed method is based on multi-stage CNN demosaicking [5], [6]. This section explains these methods.…”

Section: Related Workmentioning

confidence: 99%

“…Cui et al extends the multi-stage processing and proposed a 3-stage method [6]. The first stage processes only the G plane from a pre-processed full-color image, the second stage processes RG and BG pairs, and the third stage obtains the final full-color outputs.…”

Section: Related Workmentioning

confidence: 99%

“…In particular, Convolutional Neural Network (CNN) plays an essential role in many signal processing fields. Many CNN-based methods have been proposed in image demosaicking [1]- [6] and they obtain better performance than conventional demosaicking algorithms. For example, Tan et.al.…”

Section: Introductionmentioning

confidence: 99%

“…However, for simple CNN-based demosaicking methods which only use one loss function and one network, it is hard to control the processing in their networks. To obtain better performance, this paper focuses on multi-stage CNN algorithms [5], [6]. Compared with the simple CNN-based methods, they utilize multiple networks and calculate losses between the outputs of every network and the ground truth data.…”

Section: Introductionmentioning

confidence: 99%

“…Hence, it is difficult to approximate the demosaicking process from Bayer images with simple CNN networks. To avoid this problem, conventional methods adopt non-CNN pre-processing [1], [5], [6]. However, Tan et al [1] shows that the pre-processing can negatively influence the CNN's outputs.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Multi-Stage Dense CNN Demosaicking With Downsampling and Re-Indexing Structure

Yamaguchi

Ikehara

2020

IEEE Access

View full text Add to dashboard Cite

Multi-stage CNN is composed of several networks and loss functions. Compared to a normal single-stage CNN, a multi-stage CNN can control the processing in each network. In image demosaicking, it is utilized to follow the flow of conventional non-CNN demosaicking algorithms. This paper proposes a 4-stage CNN demosaicking method. The first stage obtains a pre-estimated full color image from an input Bayer image instead of non-CNN pre-processing. The second stage processes the color planes independently while the third stage processes the combination for two colors. The final stage obtains the full-color outputs from all outputs of the former stages. To achieve better performance, the proposed method introduces dense structures for both inside and outside of every network. Another contribution is dealing with the discontinuity between a pixel and its neighboring pixels in Bayer images by a downsampling and re-indexing structure for feature maps. These contributions lead to higher accuracy and prevent false colors. Some results through demosaicking experiments show that the proposed method can achieve better quality images with less false colors and better PSNR than state-of-the-art CNN demosaicking methods.

show abstract

“…The proposed method is based on multi-stage CNN demosaicking [5], [6]. This section explains these methods.…”

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%