Unveiling contrast in darkness

At nighttime, reduced visibility could cause foreground and background images to appear to blend together. However, ambient light is always present in the natural environment, and as a consequence, it creates some contrast in darkness. In this paper, we formulate a visual analytic method that automatically unveils the contrast of dark images (i.e. nighttime), revealing the "hidden" contents. We utilize the traits of image representations obtained from computer vision techniques through a learning based inversion algorithm, eliminating the reliance on night vision camera and at the same time minimizing the need for human intervention (i.e. manual fine-tuning the gamma correction using Adobe Photoshop software). Experiments using the new Malaya Pedestrian in the Dark (MyPD) dataset that we have collected from the website Flickr, and in comparison with conventional methods such as image integral and gamma correction, it shows the efficacy of the proposed method. Additionally, we show the potential of this framework in applications that could benefit public safety.