ANALISA PERBANDINGAN OBJECT COUNTING DENGAN ECOGNITION DAN PICTERRA

Yulianandha Mabrur, Adkha and Arafah, Fenny (2021) ANALISA PERBANDINGAN OBJECT COUNTING DENGAN ECOGNITION DAN PICTERRA. ANALISA PERBANDINGAN OBJECT COUNTING DENGAN ECOGNITION DAN PICTERRA, 2 (1). pp. 1-7. ISSN 2745-3723

	Text Analisa Perbandingan Object Counting Dengan Ecognition Dan Picterra_Jurnal ENMAP.pdf Download (1MB)
	Text Analisa Perbandingan Object Counting Dengan Ecognition Dan Picterra_Jurnal ENMAP (4).pdf Download (4MB)

Abstract

Object Counting is the process of counting objects based on their connectivity to surrounding pixels, based on 4 connection pixels or 8 connection pixels. Object counting is used to quickly see the number of objects based on the result of feature extraction automatically. This research was conducted on the object of oil palm trees using UAV photo data. This research was conducted in two areas, namely an area of 5 hectares and 15 hectares. The algorithm used is Template Matching, this algorithm allows us to find a certain part of the input image that matches the template created and the threshold value used in the template matching process to produce the number of oil palm tree calculations based on the results of the two software. The results of the template maching method on eCognition and Picterra will be analyzed based on the accuracy of the number of trees extracted by the software, then validated to determine which results are compatible or support the actual data. Based on these results, it will be known the ability of the two software in calculating the number of objects automatically, quickly and efficiently, so that it can facilitate a job such as calculating the number of oil palm, traffic lights, street lighting, and several objects that have same characteristic. Based on the results of using two different pieces of software, it can be seen that there are some areas that are less accurate. This can be obtained from making train detectors that are used automatically regarding objects that will stop. The value of the proportion of errors in the comparison taken automatically and manually obtained in an area of 5 Ha is 1.64%. Keywords: Object Counting, Citra UAV, eCognition, Picterra

Item Type:	Article
Uncontrolled Keywords:	Keywords: Object Counting, Citra UAV, eCognition, Picterra
Subjects:	Engineering > Geodesy Engineering
Divisions:	Fakultas teknik Sipil dan Perencanaan > Teknik Geodesi S1
Depositing User:	haning haning
Date Deposited:	21 Aug 2023 04:58
Last Modified:	21 Aug 2023 05:12
URI:	http://eprints.itn.ac.id/id/eprint/12648

Actions (login required)

View Item