Journal
MULTIMEDIA TOOLS AND APPLICATIONS
Volume 76, Issue 13, Pages 15027-15048Publisher
SPRINGER
DOI: 10.1007/s11042-017-4453-z
Keywords
Low illumination night-time image; Image enhancement; Negative image; Dark channel prior
Categories
Funding
- National Natural Science Foundation of China [61202198, 61401355, 41601353]
- China Scholarship Council [201608610048]
- Key Laboratory Foundation of Shaanxi Education Department [14JS072]
- Nature Science Foundation of Science Department of PeiLin count at Xi'an [GX1619]
Ask authors/readers for more resources
Images captured in low illumination conditions usually suffer from a poor visibility which has an important affect on the performance of computer vision systems. Thereby low illumination image enhancement is critical for image-related applications. In this paper, we proposed a photographic negative imaging inspired method for enhancing the images taken under low illumination environments. It consists of three basic components. First, the input night-time image is reversed to obtain its corresponding negative image (which is comparable with the latent image of a photographic film). Second, a rectification on the negative image is performed (Which is comparable with the chemical development in the process of film processing) by using an image dehazing method. This operation is inspired by the observation that the negative image looks like a hazed image and thereby it can be enhanced using a dehazing method. Third, the rectified negative image is reversed to obtain the final enhanced image (Which is comparable with the operation of fixing in the later stage of film processing.). Experiments over a large quantity of low contrast night-time images show that the proposed method is effective for enhancing low illumination images. Compared with six state-of-the-art image enhancement methods, the proposed method is superior to them in both enhancing image quality and decreasing time cost.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available