Earticle

Image compression-decompression methods have become increasingly crucial in modern times, facilitating the transfer of high-quality images while minimizing file size and internet traffic. Historically, early image compression relied on rudimentary codecs, aiming to compress and decompress data with minimal loss of image quality. Recently, a novel compression framework leveraging colorization techniques has emerged. These methods, originally developed for infusing grayscale images with color, have found application in image compression, leading to colorization-based coding. Within this framework, the encoder plays a crucial role in automatically extracting representative pixels—referred to as color seeds—and transmitting them to the decoder. The decoder, utilizing colorization methods, reconstructs color information for the remaining pixels based on the transmitted data. In this paper, we propose a novel approach to image compression, wherein we decompose the compression task into grayscale image compression and colorization tasks. Unlike conventional colorization-based coding, our method focuses on the colorization process rather than the extraction of color seeds. Moreover, we employ the Denoising Diffusion Null-Space Model (DDNM) for colorization, ensuring high-quality color restoration and contributing to superior compression rates. Experimental results demonstrate that our method achieves higher-quality decompressed images compared to standard JPEG and JPEG2000 compression schemes, particularly in high compression rate scenarios.

영상 처리에서 노이즈 제거는 후속 처리의 정확도를 위한 핵심적인 전처리 단계이다. 기존 공간 도메인 필터는 노이즈 제거 시 에지 정보까지 훼손하는 한계가 있으며, 최근 활발하게 연구되는 딥러닝 기반 모델은 연산 자원을 많이 요구하여 실시간 처리에 제약이 따른다. 본 논문에서는 수학적 원리가 명확하고 연산량이 적은 편미분방정식 기반의 Perona-Malik 비등방성 확산 모델을 적용, 연산 속도가 제한되는 환경에서도 에지를 보존하면서 노이즈를 효과적으로 제거하는 알고리즘을 제안하고 그 성능을 분석한다. 열방정식에서 유도된 비선형 확산 모델을 유한차분법으로 이산화하고, 노이즈 수준에 따른 적응적 대비 파라미터 제어 전략과 최적의 반복 횟수 기준을 체계화하였다. 성능평가에 널리 사용되는 이미지 데이터셋(Set12, BSD68 등)에 4단계 노이즈를 적용하여 실험한 결과, 제안 기법은 𝜎=10 에서 PSNR 32.00 dB, 에지 보존 지수 0.9655를 기록하며 기존 필터 대비 우수한 성능을 보였다.

In image processing, denoising is a critical preprocessing step for accurate subsequent tasks. Conventional spatial filters often degrade edge information, while recent deep learning models demand massive computational resources, restricting real-time applications. This paper proposes a computationally efficient, edge-preserving denoising algorithm using the partial differential equation-based Perona-Malik anisotropic diffusion model, suitable for environments with limited processing speed. The heat equation-derived nonlinear diffusion model is discretized via the finite difference method. Furthermore, we systematize an adaptive contrast parameter control strategy based on noise levels alongside criteria for optimal iteration counts. Experiments applying four noise levels to benchmark datasets (e.g., Set12, BSD68) show that at 𝜎=10, the proposed method achieves a PSNR of 32.00 dB and an Edge Preservation Index of 0.9655, outperforming conventional filters.