ASSESSMENT OF ATMOSPHERIC CORRECTION METHODS FOR OPTIMIZING HAZY SATELLITE IMAGERIES
DOI:
https://doi.org/10.31172/jmg.v15i3.217Keywords:
Atmospheric effects, Haze elimination, rural vegetation area, urban non vegetated area, avnir, landsatAbstract
Tujuan penelitian ini untuk menguji kesesuaian tiga jenis metode koreksi haze terhadap kejelasan obyek permukaan di wilayah tutupan vegetasi dan non vegetasi, berkenaan menghilangkan haze di wilayah citra satelit optis yang memiliki karakteristik tertentu dan diduga proses pembentukan partikel hazenya berbeda. Sehingga daerah penelitian dibagi menjadi wilayah rural yang diasumsikan sebagai daerah vegetasi dan urban sebagai non vegetasi. Pedesaan terpilih kecamatan Balaraja dan Perkotaan terpilih kecamatan Penjaringan. Tiap lokasi menggunakan Avnir-2 dan Landsat 7. Untuk mendapatkan hasil pengurangan kabut di kedua lokasi tersebut digunakan metode Dark Object Substraction (DOS), Virtual Cloud Point (VCP) dan histogram Match (HM) dengan persamaan nilai optimasi kabut HOT = DNbluesin(∂)-DNredcos(∂). hasil penelitian ini sebagai berikut: dalam hal AVNIR-Rural, VCP memiliki hasil yang baik di Band-1 sedangkan HM memiliki hasil yang baik pada band-2, 3 dan 4 sehingga dalam kasus AVNIR-Rural dapat diterapkan HM. Dalam hal AVNIR-Urban, DOS memiliki hasil yang baik pada band-1, 2 dan 3. Sementara HM memiliki hasil yang baik pada band 4, sehingga dalam kasus AVNIR-Urban dapat diterapkan DOS. Dalam kasus Landsat-Rural, DOS memiliki hasil yang baik pada band-1, 2 dan 6, Sementara VCP memiliki hasil yang baik pada band 4 dan 5. Sehingga dalam kasus Landsat-Rural dapat diterapkan DOS. Dalam hal Landsat-Urban, DOS memiliki hasil yang baik pada band-1, 2 dan 6 sedangkan VCP memiliki hasil yang baik pada band-3, 4, dan 5. Sehingga dalam hal Landsat-Urban dapat diterapkan VCP. Semakin baik citra hasil koreksi semakin kecil nilai optimasi kabut, nilai rata–rata terkecil adalah 106,547 dengan VCP di Landsat-Rural.
The purpose of this research is to examine the suitability of three types of haze correction methods toward the distinctness of surface objects in land cover. Considering the formation of haze, therefore, the main research is divided into both region namely rural assumed as vegetation and urban assumed as non-vegetation area. Region of interest for rural selected Balaraja and urban selected Penjaringan. Haze imagery reduction utilized techniques such as Dark Object Subtraction, Virtual Cloud Point and Histogram Match. By applying an equation of Haze Optimized Transformation HOT = DNbluesin(∂)-DNredcos(∂), the main result of this research includes: in the case of AVNIR-Rural, VCP has good results on Band 1 while the HM has good results on band 2, 3 and 4, therefore in the case of Avnir-Rural can be applied to HM. in the case of AVNIR-Urban, DOS has good result on band 1, 2 and 3 meanwhile HM has good results on band 4, therefore in the case of AVNIR-Urban can be applied to DOS. In the case of Landsat-Rural, DOS has a good result on band 1, 2 and 6 meanwhile VCP has good results on band 4 and 5 and the smallest average value of HOT is 106.547 by VCP, therefore in the case of Lansat-Rural can be applied to DOS and VCP. In the case of Landsat-Urban, DOS has a good result on band 1, 2 and 6 meanwhile VCP has good results on band 3, 4 and 5, therefore in the case of Landsat-Urban can be applied to VCP.
Downloads
Published
How to Cite
Issue
Section
License
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work.