Partial cloud cover is a severe problem in the optical remote sensing images, it seriously affected the interpretation and the use of remote sensing images. So how to remove or to reduce haze region from remote sensing images, and how to restore the ground information below thin cloud regions, will become the essential part in improving the availability of the applications of remote sensing images. In this paper, an improvement of haze removal method based on BSHTI (background suppressed haze thickness index) cloud detection method was introduced. We took HJ small satellite image which was shot by CCD camera as experimental data. Comparing the results, we can find that the improved method is better than other haze removal methods. The results show that the approach not only effectively reduce the interference effects of thin clouds, but also reduce the loss of remote sensing image's information, and maintaining the original image's definition. So this method is an effective and feasible haze removal method. At last, we study and realization the parallel implementation for haze remove. Experiments show that the parallel approach can improve the rate of haze removal.
[1]
Bobby R. Hunt,et al.
A new approach to removing cloud cover from satellite imagery
,
1984,
Comput. Vis. Graph. Image Process..
[2]
J. Cihlar,et al.
An image transform to characterize and compensate for spatial variations in thin cloud contamination of Landsat images
,
2002
.
[3]
Yong Du,et al.
Haze detection and removal in high resolution satellite image with wavelet analysis
,
2002,
IEEE Trans. Geosci. Remote. Sens..
[4]
Li Cunjun,et al.
Automatic detection and removal of thin haze based on own features of Landsat image
,
2006
.
[5]
Xiaofeng Yang,et al.
An Operational Improvement of Haze/Clear Line Identification from Satellite Imagery Based on Multi-Resolution Segmentation
,
2008,
2008 Second International Symposium on Intelligent Information Technology Application.
[6]
Xiaofeng Yang,et al.
Haze Removal from the Visible Bands of CBERS Remote Sensing Data
,
2009,
2009 International Conference on Industrial and Information Systems.