This paper addresses the challenge of image restoration (IR) by integrating deep convolutional neural networks (CNNs) as denoiser priors into model-based optimization methods. The authors propose a set of fast and effective CNN denoisers, which are trained to handle various noise levels and color images. These denoisers are then integrated into a model-based optimization framework using variable splitting techniques, such as half-quadratic splitting (HQS), to solve other inverse problems like image deblurring and super-resolution. The experimental results demonstrate that the learned CNN denoisers not only achieve promising Gaussian denoising results but also significantly improve the performance of IR tasks. The proposed method is flexible, efficient, and effective, making it a strong competitor to both model-based optimization methods and discriminative learning methods. The paper also discusses the advantages of using CNNs for denoiser priors, including their speed, performance, and ability to handle color images.This paper addresses the challenge of image restoration (IR) by integrating deep convolutional neural networks (CNNs) as denoiser priors into model-based optimization methods. The authors propose a set of fast and effective CNN denoisers, which are trained to handle various noise levels and color images. These denoisers are then integrated into a model-based optimization framework using variable splitting techniques, such as half-quadratic splitting (HQS), to solve other inverse problems like image deblurring and super-resolution. The experimental results demonstrate that the learned CNN denoisers not only achieve promising Gaussian denoising results but also significantly improve the performance of IR tasks. The proposed method is flexible, efficient, and effective, making it a strong competitor to both model-based optimization methods and discriminative learning methods. The paper also discusses the advantages of using CNNs for denoiser priors, including their speed, performance, and ability to handle color images.