Laser & Optoelectronics Progress, Volume. 57, Issue 16, 161023(2020)
Single Image Dehazing Based on Superpixel Segmentation Combined with Dark-Bright Channels
Fig. 1. Improvement effect comparison of filtering window. (a) Original hazy image; (b) dark-channel filtering window image; (c) proposed superpixel filtering window image; (d) enlarged image of dark-channel filtering window; (e) enlarged image of proposed superpixel filtering window
Fig. 2. Dark-channel and bright-channel images. (a)(b)(c)(d) Original images; (e)(f)(g)(h) dark-channel images; (i)(j)(k)(l) bright-channel images
Fig. 3. Atmospheric light locations determined by different methods. (a) Hazy images; (b) proposed sky segmentation images; (c) position maps of atmospheric light determined by proposed method; (d) position maps of atmospheric light determined by dark-channel algorithm
Fig. 4. Transmission optimization. (a) Original image; (b) transmission image without optimization; (c) transmission image optimized using guided filter; (d) transmission image optimized using proposed method
Fig. 5. Dehazing results of hazy images with sky areas and depth-of-field changes by different methods. (a) Hazy images; (b) method in Ref. [8]; (c) method in Ref. [9]; (d) method in Ref. [10]; (e) method in Ref. [15]; (f) proposed method
Fig. 6. Dehazing results of hazy images with close views and visions by different methods. (a) Hazy images; (b) method in Ref. [8]; (c) method in Ref. [9]; (d) method in Ref. [10]; (e) method in Ref. [15]; (f) proposed method
|
|
Get Citation
Copy Citation Text
Yong Chen, Chentao Lu. Single Image Dehazing Based on Superpixel Segmentation Combined with Dark-Bright Channels[J]. Laser & Optoelectronics Progress, 2020, 57(16): 161023
Category: Image Processing
Received: Feb. 4, 2020
Accepted: Mar. 10, 2020
Published Online: Aug. 5, 2020
The Author Email: Chen Yong (edukeylab@126.com)