Dense-Haze: A Benchmark for Image Dehazing with Dense-Haze and Haze-Free Images

Single image dehazing is an ill-posed problem that has recently drawn important attention. Despite the significant increase in interest shown for dehazing over the past few years, the validation of the dehazing methods remains largely unsatisfactory, due to the lack of pairs of real hazy and corresponding haze-free reference images. To address this limitation, we introduce Dense-Haze - a novel dehazing dataset. Characterized by dense and homogeneous hazy scenes, Dense-Haze contains 33 pairs of real hazy and corresponding haze-free images of various outdoor scenes. The hazy scenes have been recorded by introducing real haze, generated by professional haze machines. The hazy and haze-free corresponding scenes contain the same visual content captured under the same illumination parameters. Dense-Haze dataset aims to push significantly the state-of-the-art in single-image dehazing by promoting robust methods for real and various hazy scenes. We also provide a comprehensive qualitative and quantitative evaluation of state-of-the-art single image dehazing techniques based on the Dense-Haze dataset. Not surprisingly, our study reveals that the existing dehazing techniques perform poorly for dense homogeneous hazy scenes and that there is still much room for improvement.

I. INTRODUCTION

Haze represents one of the atmospheric phenomena most challenging for camera sensors and vision applications. Haze is an often occurring meteorological phenomenon especially during autumn and spring in temperate climates being generated by particles that scatter and absorb the incident light. The visibility of such hazy scene is highly degraded and as a result they are characterized by poor contrast, and additional noise. For instance, the presence of haze has a great impact in the road traffic as it may severely reduce the visibility for drivers. As a result, restoring the contents in hazy images – process known as dehazing – is important for several outdoor image processing and computer vision applications such as ADAS, remote sensing and visual surveillance.