An Analysis of Dehazing Methods with Detail Fusion Combination Technique

Authors

  • Sarudin Kari Razak Faculty of Technology and Informatics, Advanced Informatics School, Universiti Teknologi Malaysia
  • Abdullah Bade Faculty of Science and Natural Resources, Universiti Malaysia Sabah
  • Mohd Shafry Mohd Rahim School of Computing, Faculty of Engineering, Universiti Teknologi Malaysia
  • Noor Asma Husain School of Computing, Faculty of Engineering, Universiti Teknologi Malaysia

Keywords:

image processing, dehazing method, ground truth, image quality assessment

Abstract

The proposed Detail Fusion Combination Technique (DFC) has an objective to reduce the haze in a single image whilst preserving the image quality. The result of the proposed technique has been compared with other state-of-the-art dehazing approaches and evaluated with image quality assess- ment. Various image quality assessment methods such as the Mean Squared Er- ror (MSE), Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM) have been used to test the veracity of the result. Each resultant image has been evaluated with its corresponding ground truth counterpart respectively. Additionally, auxiliary datasets without ground truth images have been used with the proposed dehazing method in order to show the image quality qualita- tively.

Published

2020-07-10