Nighttime Haze Removal Based on New Imaging Model with Artificial Light Sources
Funds:
The National Natural Science Foundation of china (61372145, 61472274, 61632081)
-
摘要: 夜間有霧圖像光照不均勻，整體亮度較低，色偏嚴重，且人工光源周圍存在光暈?，F(xiàn)有的去霧模型和算法大多針對白天圖像，其并不適用于夜間場景，夜間圖像去霧頗具挑戰(zhàn)性。該文深入分析夜間有霧圖像的成像規(guī)律，建立含有人工光源的夜間霧天圖像成像新模型，并在此基礎上提出夜間圖像去霧新算法。針對夜間圖像光照不均問題，提出基于低通濾波的環(huán)境光估計方法，利用估計出的環(huán)境光可準確預測夜間場景傳輸率；針對目前夜間圖像去霧后存在光源光暈問題，提出根據(jù)圖像色度估計場景點屬于近光源區(qū)域的程度，使算法能自適應地處理光源區(qū)域和非光源區(qū)域；針對非一致色偏問題，利用直方圖匹配方法進行顏色校正。對大量圖像進行實驗，并與現(xiàn)有白天、夜晚圖像去霧算法進行比較，驗證了該文提出的夜間霧天圖像成像模型及去霧算法的有效性。Abstract: The non-uniform illumination, low brightness, serious color deviation and halo effects around artificial light sources lead to the difficulty in haze removal for night-time image. The existing dehazing methods are mostly designed for daytime image and not applicable to nighttime image. This paper focuses on researching nighttime image dehazing. A new nighttime haze model that accounts for the artificial varying light sources is introduced. Based on this new model, a new dehazing framework is proposed. Firstly, the atmospheric light is estimated based on the low pass filter method. This atmospheric light map can be used to predict the transmission of night scene accurately. Secondly, to solve the problem of halo effects around artificial light sources in existing dehazing methods, a method that estimates the distance between the object of the scene and the artificial light sources based on the image chromaticity is proposed. In this way, the scene objects near to the light source region and objects far away from the light source region can be processed respectively. Finally, as for the color cast, an efficient color correction algorithm based on the histogram matching is presented in this paper. Comparing with existing daytime and nighttime dehazing methods, the experimental results of a number of examples demonstrate the effectiveness of the proposed night-time haze model and the dehazing method.
-
NARASIMHAN S and NAYAR S. Vision and the atmosphere [J]. International Journal of Computer Vision, 2002, 48(3): 233-254. doi: 10.1023/A:1016328200723. LIU Haibo, YANG Jie, WU Zhengping, et al. A fast single image dehazing method based on dark channel prior and Retinex theory[J]. Acta Automatica Sinica, 2015, 41(7): 1264-1273. doi: 10.16383/j.aas.2015.c140748. LI Hui, XIE Weihao, and WANG Xingang. GPU implementation of multi-scale Retinex image enhancement algorithm[C]. IEEE/ACS International Conference of Computer Systems and Applications (AICCSA), Agadir, Morocco, 2016: 1-5. AI Y, TSAI P, YAO C, et al. Improved local histogram equalization with gradient-based weighting process for edge preservation[J]. Multimedia Tools and Applications, 2017, 76(1): 1585-1613. doi: 10.1007/s11042-015-3147-7. TAN R. Visibility in bad weather from a single image[C]. IEEE Conference on Computer Vision and Pattern Recognition(CVPR), Anchorage, USA, 2008: 1-8. HE Kaiming, SUN Jian, and TANG Xiaoou. Single image haze removal using dark channel prior[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, 33(12): 2341-2353. doi: 10.1109/TPAMI.2010.168. FATTAL R. Single image dehazing[J]. ACM Transactions on Graphics, 2008, 27(3): 1-9. doi: 10.1145/1399504.1360671. ZHANG Jing, CAO Yang, and WANG Zengfu. Nighttime haze removal based on a new imaging model[C]. IEEE International Conference on Image Processing, Quebec, Canada, 2015: 4557-4561. LI Yu, TAN R, and BROWN M. Nighttime haze removal with glow and multiple light colors[C]. IEEE International Conference on Computer Vision(ICCV), Santiago, Chile, 2015: 226-234. ANCUTI C, ANCUTI C, VLEESCHOUWER C, et al. Night- time dehazing by fusion[C]. IEEE International Conference on Image Processing(ICIP), Phoenix, USA , 2016: 2256-2260. LI Zhengguo, WEI Zhe, WEN Changyun, et al. Detail- enhanced multi-Scale exposure fusion[J]. IEEE Transactions on Image Processing, 2017, 26(3): 1243-1252. doi: 10.1109/ TIP.2017.2651366. ZHANG Jin, CAO Yang, FANG Shuai, et al. Fast haze removal for nighttime image using maximum reflectance prior [C]. IEEE Conference on Computer Vision and Pattern Recognition(CVPR), Honolulu, USA, 2017: 7016-7024. PARK D, HAN D K, and KO H. Nighttime image dehazing using local atmospheric selection rule and weighted entropy for visible-light systems[J]. Optical Engineering, 2017, 56(5): 050501. doi: 10.1117/1.OE.56.5.050501. NARASIMHAN S and NAYAR S. Shedding light on the weather[C]. IEEE Conference on Computer Vision and Pattern Recognition(CVPR), Madison, USA , 2003: 665-672. HE Kaiming, SUN Jian, and TANG Xiaoou. Guided image filtering[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(6): 1397-1409. doi: 10.1109/ TPAMI.2012.213. REINHARD E, STARK M, SHIRLEY P, et al. Photographic tone reproduction for digital images[J]. ACM Transactions on Graphics, 2002, 21(3): 267-276. doi: 10.1145/566570.566575. FINLAYSON G and TREZZI E. Shades of gray and colour constancy[C]. Color and Imaging Conference(CIC), Scottsdale, USA, 2004: 37-41. FU Xueyang, ZHUANG Peixian, HUANG Yue, et al. A Retinex-based enhancing approach for single underwater image[C]. IEEE International Conference on Image Processing (ICIP), Paris, France, 2015: 4572-4576. SHEN Dinggang. Image registration by local histogram matching[J]. Pattern Recognition, 2007, 40(4): 1161-1172. doi: 10.1016/j.patcog.2006.08.012. 李大鵬, 禹晶, 肖創(chuàng)柏, 等. 圖像去霧的無參考客觀質(zhì)量評測方法[J]. 中國圖象圖形學報, 2011, 16(9): 1753-1757. doi: 10.11834/jig.20110928. LI Dapeng, YU Jing, XIAO Chuangbai, et al. No-reference quality assessment method for defogged images[J]. Journal of Image and Graphics, 2011, 16(9): 1753-1757. doi: 10.11834/ jig.20110928. -
計量
- 文章訪問數(shù): 1991
- HTML全文瀏覽量: 388
- PDF下載量: 219
- 被引次數(shù): 0
下載:
