No-reference image quality assessment using Prewitt magnitude based on convolutional neural networks

  • Li J
  • Zou L
  • Yan J
 et al. 
  • 12


    Mendeley users who have this article in their library.
  • 16


    Citations of this article.


No-reference image quality assessment is of great importance to numerous image processing applications, and various methods have been widely studied with promis-ing results. These methods exploit handcrafted features in the transformation or space domain that are discriminated for image degradations. However, abundant a priori knowl-edge is required to extract these handcrafted features. The convolutional neural network (CNN) is recently introduced into the no-reference image quality assessment, which inte-grates feature learning and regression into one optimization process. Therefore, the network structure generates an effec-tive model for estimating image quality. However, the image quality score obtained by the CNN is based on the mean of all of the image patch scores without considering the human visual system, such as edges and contour of images. In this paper, we combine the CNN and the Prewitt magnitude of segmented images and obtain the image quality score using the mean of all the products of the image patch scores and weights based on the result of segmented images. Experi-mental results on various image distortion types demonstrate that the proposed algorithm achieves good performance.

Author-supplied keywords

  • Convolutional neural networks (CNNs)
  • Graph-based image segmentation
  • No-reference image quality assessment
  • Prewitt magnitude

Get free article suggestions today

Mendeley saves you time finding and organizing research

Sign up here
Already have an account ?Sign in

Find this document

Get full text


  • Jie Li

  • Lian Zou

  • Jia Yan

  • Dexiang Deng

  • Tao Qu

  • Guihui Xie

Cite this document

Choose a citation style from the tabs below

Save time finding and organizing research with Mendeley

Sign up for free