INTEGRASI METODE KLASIFIKASI BERBASIS OBJEK CITRA QUICKBIRD DAN SISTEM INFORMASI GEOGRAFIS UNTUK PEMODELAN SPASIAL ZONASI KEBUTUHAN ANGKUTAN UMUM MASSAL DI KOTA BANDUNG PROVINSI JAWA BARAT

LILI SOMANTRI, Prof. Dr. Totok Gunawan, M.S.

2014 | Disertasi | S3 Penginderaan Jauh

Bandung merupakan kota yang mengalami masalah kemacetan lalu lintas. Salah satu solusi mengurangi kemacetan adalah dengan penataan transportasi angkutan umum massal. Penataan angkutan umum massal memerlukan informasi zonasi wilayah yang membutuhkan angkutan umum massal tersebut. Teknologi penginderaan jauh dan sistem informasi geografis diharapkan dapat membantu dalam penentuan zonasi kebutuhan angkutan umum massal di Kota Bandung berupa penyediaan data dan informasi yang akurat. Citra penginderaan jauh yang digunakan, yaitu Citra Quickbird Multispektral dan Quickbird Pan-Sharpened. Adapun penelitian ini bertujuan 1) mengembangkan data pemanfaatan lahan hasil ekstraksi klasifikasi berbasis objek pada citra Quickbird dalam pemodelan spasial zonasi kebutuhan angkutan umum massal di Kota Bandung, 2) mengembangkan pemodelan spasial dengan Sistem Informasi Geografis untuk zonasi kebutuhan angkutan umum massal dan penentuan rute angkutan umum massal di Kota Bandung. Metode yang digunakan dalam penelitian ini, meliputi a) metode penginderaan jauh untuk memperoleh data parameter penentuan kebutuhan angkutan umum massal melalui teknik ekstraksi klasifikasi berbasis objek (object based image approach) dan interpretasi visual; b) pemodelan spasial menggunakan model indeks dengan pengharkatan (scoring) dan matrik dua dimensi untuk menentukan zonasi kebutuhan angkutan umum massal, d) penentuan rute angkutan umum massal menggunakan network analyst pada sistem informasi geografis. Kesimpulan yang diperoleh adalah metode klasifikasi berbasis objek (OBIA) mampu mengekstrak data pemanfaatan lahan dari Citra Quickbird Multispektral dengan 9 kategori, yaitu permukiman teratur, permukiman tidak teratur, perdagangan, industri, jasa, hutan kota, lahan terbuka, sawah, dan tegalan. Akurasi data pemanfaatan lahan hasil OBIA, yaitu 42.74% dibandingkan dengan akurasi interpretasi visual citra Quickbird Pan-Sharpened, yaitu 98.40%. Metode OBIA belum mampu menghasilkan data pemanfaatan lahan yang diperlukan dalam penentuan zonasi angkutan umum massal. Untuk pemodelan spasial zonasi kebutuhan angkutan umum massal menggunakan data hasil interpretasi visual citra Quickbird Pan-Sharpened. Data pemanfaatan lahan digunakan untuk penentuan zona bangkitan (permukiman teratur, tidak teratur) dan tarikan pergerakan (perdagangan, jasa, industri,dan pendidikan) Zonasi kebutuhan angkutan umum massal berdasarkan bangkitan pergerakan, tarikan pergerakan, kondisi penduduk, dan jaringan jalan. Kecamatan yang seluruh kelurahannya termasuk kategori sangat membutuhkan angkutan massal yaitu Coblong, Sukajadi, Kiaracondong, Mandalajati, Ujungberung, dan Cibiru. Berdasarkan tingkat pelayanan, jalan yang sangat membutuhkan angkutan umum massal yaitu arteri primer, arteri sekunder, dan kolektor sekunder. Berdasarkan zonasi kebutuhan angkutan umum massal direkomendasikan empat rute angkutan massal busway dan satu rute angkutan massal monorail.

Bandung is a city with traffic congestion problems. One of the solutions to reduce the congestion is to manage public transportation. Such mass public transportation management requires information on the zonation of territories that require the mass public transportation. It can be expected that both remote sensing technology and geographic information system can be useful in determining the zonation of territories with the need for mass public transportation in Bandung city, such as the provision of accurate data and information. The remote sensing images used were Multispectral Quickbird and Pan-Sharpened Quickbird images. The purposes of this study are 1) to develop the data of land use result the objectbased classification extraction method for modeling the spatial zonation of territories with the need for mass public transportation in Bandung city, and 2) to develop modeling the zonation through geographic information system for spatial zonation of the need and route mass public transportation in Bandung city. Methods used in the study were 1) a remote sensing method to gain the data of parameters in determining the territories with the need for mass public transportation through the object-based image approach and visual interpretation, 2) the spatial modeling of zonation by using an index model with scoring and two dimension matrix of determining the zonation of territories with the need for mass public transportation, 3) the determination of route for mass public transportation by using network analyst in geographic information system. Based on the results of the study, it can be concluded that the object-based classification method (object-based image approach-OBIA) was able to extract the data of land use from Multispectral Quickbird images with 9 category, that are regular settlement, irregular settlement, trades, industries, service, urban forest, open land, wet land, and dried land. The accuracy of lands as the result of OBIA was 42.74% compared to those of the visual interpretation of Pan-Sharpened Quickbird was 98.40%. OBIA method from Multispectral image not able to produce the data of land utilization could be used in the spatial modeling of the zonation. Modeling used the data of land utilization from visual interpretation Pan-Sharpened Quickbird image. The data of land utilization used to determine the generating zone (regular settlement, irregular settlement), and attracting zone of population movement (trades, service, industry, and education). The zonation of territories with the need for mass public transportation was based on generation of population movement, the attraction of population movement, population conditions, and road network. Sub districts with all urban villages being included into the category of requiring the mass transportation were Coblong, Sukajadi, Kiaracondong, Mandalajati, Ujungberung, and Cibiru. Based on the level of road service, zones requiring the mass public transportation were primary arterial roads, secondary arterial roads, and secondary collector roads. Based on the zonation of territories with the need for mass public transportation recommend four mass public transportation routes for busway, and one for monorail.

Kata Kunci : Zonasi Angkutan Umum Massal, Citra Quickbird Multispektral, Citra Quickbird Pan-Sharpened, Klasifikasi Berbasis Objek, Interpretasi Visual, dan Pemodelan Spasial