Klasifikasi Habitat Bentik Perairan Dangkal Dari Citra Worldview-2 Menggunakan Data in-Situ Dan Drone

Sugara, Ayub; Siregar, Vincentius P.; Agus, Syamsul Bahri

doi:10.29244/jitkt.v12i1.26448

Cited by 10 publications

(10 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Sampai saat ini penamaan kelas habitat bentik belum memiliki standardisasi yang baku, sehingga disesuaikan dengan komposisi penyusun habitat bentik dominan yang teramati di lapangan. Beberapa penelitian sebelumnya telah banyak dilakukan dan menghasilkan skema klasifikasi atau jumlah kelas yang berbedabeda (Phinn et al, 2011;Zhang et al, 2013;Wahidin et al, 2015;Sugara et al, 2020;Siregar et al, 2020).…”

Section: Skema Klasifikasiunclassified

Klasifikasi Habitat Bentik Atol Kaledupa Taman Nasional Wakatobi Dengan Algoritma Support Vector Machine

Setiawan¹,

Siregar²,

Susilo³

et al. 2023

J. Ilmu dan Teknologi Kelautan Tropis

Self Cite

View full text Add to dashboard Cite

Atol Kaledupa merupakan salah satu kawasan yang dijadikan sebagai zona perlindungan bahari dan zona pemanfaatan lokal di Taman Nasional Wakatobi. Informasi spasial habitat bentik sangat terbatas sehingga penelitian ini diharapkan menjadi pendukung dalam strategi dan upaya pelestarian keanekaragaman hayati laut Atol Kaledupa. Penelitian ini bertujuan untuk memetakan habitat bentik Atol Kaledupa dengan menggunakan metode klasifikasi terbimbing berbasis piksel dan objek/OBIA dengan algoritma support vector machine (SVM). Data yang digunakan adalah citra satelit Sentinel-2 dengan resolusi spasial 10 x10 m yang diakuisisi pada tanggal 4 Nopember 2019. Pengamatan habitat bentik dilakukan secara langsung di lokasi penelitian dengan meletakkan transek kuadran dan mengambil titik pada area habitat yang dominan atau homogen. Transek yang digunakan berukuran 100 x 100 cm2. Klasifikasi citra menggunakan input themathic layer dari data lapangan. Hasil klasifikasi habitat bentik dikelompokan menjadi 6 kelas. Habitat bentik dapat dipetakan menggunakan metode OBIA dengan tingkat akurasi sebesar 78,1% sedangkan klasifikasi berbasis piksel memiliki akurasi keseluruhan 61,8%. Klasifikasi habitat bentik dengan algoritma SVM menggunakan metode OBIA memberikan informasi yang lebih baik dibandingkan dengan metode berbasis piksel.

show abstract

Section: Skema Klasifikasiunclassified

Klasifikasi Habitat Bentik Atol Kaledupa Taman Nasional Wakatobi Dengan Algoritma Support Vector Machine

Setiawan¹,

Siregar²,

Susilo³

et al. 2023

J. Ilmu dan Teknologi Kelautan Tropis

Self Cite

View full text Add to dashboard Cite

show abstract

“…[24] reported that atmospheric correction could increase classification accuracy. Further, images without atmospheric correction are thought to have underestimated spectral reflectance value [25]. It was proven by increasing the visual quality of the images with the correction compared to those without correction [16], [25].…”

Section: Atmospheric and Water Column Correctionmentioning

confidence: 99%

Identification of seagrass beds using high-resolution satellite imagery to basic inventory data supporting coastal ecosystem management (Case study: Lancang Island)

Sugara,

Kurniawati,

Nurholis

et al. 2023

IOP Conf. Ser.: Earth Environ. Sci.

View full text Add to dashboard Cite

Seagrass habitat is an important element of the coastal marine ecosystem, as it influences all major fields in the coastal environments. Due to anthropogenic and natural pressure, such identification of habitat distribution becomes crucial. Remote sensing can provide the best solution regarding time and cost to extract seagrass beds’ spatial and temporal information. When the low spatial resolution satellite image provides low accuracy and high uncertainty of seagrass distribution, we assess the high spatial resolution image product, the WorldView-2 image, to identify the seagrass bed distribution in Lancang Island. We confidently found that higher resolution images could leverage the mapping quality, especially on the seagrass beds, which could be highly sensitive to miss-detection. We can produce 83,75% overall accuracy using pixel-based classification; this is better among other low-resolution images for Instants, Landsat, and Sentinel. Our results suggest that the WorldView-2 image could perform better on seagrass beds identification, yet further improvement on more robust classification methods and algorithms could amplify the mapping and monitoring strategy. Based on this, it can be used as basic data to support future coastal ecosystem management on Lancang Island.

show abstract

“…Rendahnya akurasi peta dan kualitas hasil akhir citra dapat disebabkan oleh tutupan awan (Putro & Handayani, 2015). Selain itu, akurasi peta juga dipengaruhi oleh jumlah kelas, metode pengambilan, dan jumlah titik ground truth sampel penguji (Sugara, Siregar, & Agus, 2020). Persentase terumbu karang di transek 1 sampai transek 20 dapat dilihat pada Tabel 3.…”

Section: Analisis Akurasi Peta Sebaran Terumbu Karangunclassified

Pemetaan terumbu karang dengan citra satelit Sentinel-2 dan analisis kondisi karang di kawasan Pantai Pasir Putih, Situbondo Jawa Timur

Fuad

Ramadhani

Dewi

et al. 2022

JPG

View full text Add to dashboard Cite

This research was conducted in Pasir Putih, Situbondo district, East Java with the aim of mapping coral reef ecosystems and analyzing the condition of coral diversity to support learning Geography in remote sensing courses. Coral reef maps were extracted from Sentinel-2 satellite imagery by applying a water column correction algorithm. Mapping of coral reef areas resulted in 3 categories, namely live coral with an area of 32,860 Ha, dead coral 31,690 Ha, and sand 42,670 Ha with a level of 72.2 percent which is included in the good category. Testing the accuracy of the extracted map using the confusion matrix method based on 90 field survey data. The accuracy of the resulting map met the accuracy standard specified in SNI No. 7716:2011 for Mapping of Habitats in Shallow Sea Waters. Furthermore, coral reefs were observed in 20 locations using the Line Intercept Transect (LIT) method. The results showed that 13 of the 20 transect areas were found to be in poor condition with less than 25 percent live coral cover. The most common types of coral life forms found at the study site were Coral Massive, Acropora Branching, and Acropora Tabulate. Coral life forms that are very rarely encountered are Acropora Submassive and Coral Mushroom. This research is very relevant in remote sensing courses for learning Geography.

show abstract

Klasifikasi Habitat Bentik Perairan Dangkal Dari Citra Worldview-2 Menggunakan Data in-Situ Dan Drone

Cited by 10 publications

References 19 publications

Klasifikasi Habitat Bentik Atol Kaledupa Taman Nasional Wakatobi Dengan Algoritma Support Vector Machine

Klasifikasi Habitat Bentik Atol Kaledupa Taman Nasional Wakatobi Dengan Algoritma Support Vector Machine

Identification of seagrass beds using high-resolution satellite imagery to basic inventory data supporting coastal ecosystem management (Case study: Lancang Island)

Pemetaan terumbu karang dengan citra satelit Sentinel-2 dan analisis kondisi karang di kawasan Pantai Pasir Putih, Situbondo Jawa Timur

Contact Info

Product

Resources

About