The color demosaicing and image scaling based on improve Hamilton-Adams

The demosaicing process is the most critical step in image processing, this study presents a Hamilton-Adams-based algorithm for mosaic removal and image scaling. The algorithm consists of three steps: image pre-processing, image scaling, and improved Hamilton-Adams mosaic removal. The image pre-proc...

Full description

Saved in:
Bibliographic Details
Published inProceedings ... Asia-Pacific Signal and Information Processing Association Annual Summit and Conference APSIPA ASC ... (Online) pp. 892 - 897
Main Authors Peng, Yu-Wen, Hu, Chia-Yu, Chin, Yen-Ju, Chou, He-Sheng, Lin, Yuan-Jin, Liu, Yu-Lin, Chen, Shih-Lun, Chen, Tsung-Yi, Li, Kuo-Chen, Chen, Chiung-An
Format Conference Proceeding
LanguageEnglish
Published IEEE 31.10.2023
Subjects
Color filter array
Demosaicing
Hamilton-Adams
Image color analysis
Image edge detection
Image pre-processing
Image quality
Image scaling
Interpolation
Measurement
Noise reduction
Pipelines
Online AccessGet full text
ISSN2640-0103
DOI10.1109/APSIPAASC58517.2023.10317437

Cover

More Information
Summary:The demosaicing process is the most critical step in image processing, this study presents a Hamilton-Adams-based algorithm for mosaic removal and image scaling. The algorithm consists of three steps: image pre-processing, image scaling, and improved Hamilton-Adams mosaic removal. The image pre-processing stage employs a mean filter to reduce high-frequency signals at the image edges, effectively mitigating the zipper effect caused by image scaling and improving image quality. The improved Hamilton-Adams method incorporates high-quality linear interpolation to achieve more accurate edge computations. In terms of image quality assessment, this paper increased 1.13 dB of PSNR, 1.19 dB of CPSNR, and 0.03 of SSIM values compared to high-quality linear interpolation with bicubic interpolation.
ISSN:2640-0103
DOI:10.1109/APSIPAASC58517.2023.10317437