A Framework for Objective Evaluation of Single Image De-Hazing Techniques

Alessandro Artusi,Konstantinos A Raftopoulos

doi:10.1109/access.2021.3082207

Abstract

Real-world environment, where images are acquired with digital camera, may be subject to sever climatic conditions such as haze that may drastically reduce the quality performance of sophisticated computer vision algorithms used for various tasks, e.g., tracking, detection, classification etc. Even though several single image de-hazing techniques have been recently proposed with many deep-learning approaches among them, a general statistical framework that would permit an objective performance evaluation has not been independently introduced yet. In this manuscript, certain performance metrics that emphasize different aspects of image quality, output ranges and polarity, are identified and combined into a single performance indicator derived in an unbiased manner. A general methodology is thus introduced, as a framework for objective performance evaluation of current and future dehazing tasks, through an extensive comparison of 15 single image de-hazing techniques over a vast range of image data sets. The proposed unified framework shows several advantages in evaluating diverse and perceptually meaningful image features but also in elucidating future directions for improvement in image dehazing tasks.

Highlights

Image capturing is the first step in an imaging pipeline and plays an important role in providing input with an acceptable level of quality that will not compromise the performance of existing computer vision and image processing tasks, i.e., classification, object detection, details extraction etc
The quality of the captured image may be degraded by noise generated by the device sensor and by environmental conditions such as fog, rain drops, haze, illuminations conditions etc. affecting visibility of objects, details etc. and drastically compromising the recognition capability of computer vision and image processing tasks
Methods that use image prior features to constraint the solution of the scattering model and methods that use a black-box approach where deep-learning is used to limit the constraints required by the prior features approaches

Summary

A Framework for Objective Evaluation of Single Image De-Hazing Techniques

The work of Alessandro Artusi was supported in part by the European Union’s Horizon 2020 Research and Innovation Programme under Grant 739578, and in part by the Government of the Republic of Cyprus through the Deputy Ministry of Research, Innovation and Digital Policy.

INTRODUCTION

EXPERIMENTS AND ANALYSIS

PERFORMANCE EVALUATION

CONCLUSION