SUPERVISED HAZE TRANSFORM METHOD FOR HAZE DETECTION AND REMOVAL IN SPOT 6/7 IMAGERY
| Main Authors: | Sulyantara, D. Heri, Ulfa, Kurnia, Prabowo, Yudhi, Siwi, Sukentyas Estuti, Brahmantara, Randy Prima |
|---|---|
| Other Authors: | Pusat Teknologi dan Data Penginderaan Jauh LAPAN |
| Format: | Article info application/pdf eJournal |
| Bahasa: | eng |
| Terbitan: |
Indonesian National Institute of Aeronautics and Space (LAPAN)
, 2019
|
| Subjects: | |
| Online Access: |
http://jurnal.lapan.go.id/index.php/jurnal_inderaja/article/view/3060 http://jurnal.lapan.go.id/index.php/jurnal_inderaja/article/view/3060/2438 |
Daftar Isi:
- Multispectral satellite images often contaminated by haze and cirrus. It will reduce the accuracy of data interpretation. There are some haze detection and removal methods that have been developed by the experts. However, haze detection and removal still remains become the one of the challenges for optical multispectral data correction. This paper purposed to remove haze on SPOT 6/7 imagery by determining the haze and cloud threshold using the Supervised Haze Transform (SHT). The method is used on hazy SPOT 6/7 imagery. This method is developed based on the reflectance slope of blue and red visible bands by taking the training sample in a clear, a little haze, and lots of haze area of vegetation and bare soil. The vegetation and bare soil in this case are attempted to have the same land cover. Visual comparison is showed by the result of haze detection using manual threshold and mean threshold. The best result haze detection by using the manual threshold. So that, The best result haze detection was the using of manual threshold. So that, the haze removal in SPOT 6/7 imagery can not be done massifly. Therefore, it is necessary to have an algorithm by inputting the image and corresponding threshold to produce a haze-free image more quickly and efficiently.