This paper provides a comprehensive survey of diffusion model-based techniques in low-level vision tasks, addressing the gap in existing literature. Diffusion models, characterized by their forward diffusion and reverse denoising processes, have emerged as powerful tools for generating high-quality and diverse samples, particularly in low-level vision tasks. The paper begins with an introduction to diffusion models, including three generic frameworks: denoising diffusion probabilistic models (DDPMs), noise-conditional score networks (NCSNs), and stochastic differential equations (SDEs). It then categorizes existing DM-based methods into supervised and zero-shot approaches, and further into specific tasks such as super-resolution, inpainting, deblurring, dehazing, low-light image enhancement, and image fusion. The paper also explores the application of DMs in extended domains like medical imaging, remote sensing, and video analysis. Additionally, it provides an overview of commonly used benchmarks and evaluation metrics, conducts a thorough experimental analysis, and identifies limitations and future research directions. The survey aims to facilitate a deeper understanding of diffusion models in low-level vision tasks and inspire further research in the field.This paper provides a comprehensive survey of diffusion model-based techniques in low-level vision tasks, addressing the gap in existing literature. Diffusion models, characterized by their forward diffusion and reverse denoising processes, have emerged as powerful tools for generating high-quality and diverse samples, particularly in low-level vision tasks. The paper begins with an introduction to diffusion models, including three generic frameworks: denoising diffusion probabilistic models (DDPMs), noise-conditional score networks (NCSNs), and stochastic differential equations (SDEs). It then categorizes existing DM-based methods into supervised and zero-shot approaches, and further into specific tasks such as super-resolution, inpainting, deblurring, dehazing, low-light image enhancement, and image fusion. The paper also explores the application of DMs in extended domains like medical imaging, remote sensing, and video analysis. Additionally, it provides an overview of commonly used benchmarks and evaluation metrics, conducts a thorough experimental analysis, and identifies limitations and future research directions. The survey aims to facilitate a deeper understanding of diffusion models in low-level vision tasks and inspire further research in the field.