Segmentation of Remotely Sensed Imagery: Moving from Sharp Objects to Fuzzy Regions

Lizarazo, Ivan; Elsner, Paul

doi:10.5772/15421

Cited by 10 publications

(7 citation statements)

References 65 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…9. This shows that the classifications obtained by applying the object-oriented technique produces higher thematic accuracy than traditional per-pixel methods [41,42,43,44].…”

Section: Discussionmentioning

confidence: 88%

Land cover classification at three different levels of detail from optical and radar Sentinel SAR data: a case study in Cundinamarca (Colombia)

Florez¹,

Lizarazo²

2020

DYNA

View full text Add to dashboard Cite

In this paper, the potential of Sentinel-1A and Sentinel-2A satellite images for land cover mapping is evaluated at three levels of spatial detail; exploratory, reconnaissance, and semi-detailed. To do so, two different image classification approaches are compared: (i) a traditional pixel-wise approach; and (ii) an object–oriented approach. In both cases, the classification task was conducted using the “RandomForest” algorithm. The case study was also intended to identify a set of radar channels, optical bands, and indices that are relevant for classification. The thematic accuracy of the classifications displays the best results for the object-oriented approach to exploratory and recognition levels. The results show that the integration of multispectral and radar data as explanatory variables for classification provides better results than the use of a single data source.

show abstract

“…9. This shows that the classifications obtained by applying the object-oriented technique produces higher thematic accuracy than traditional per-pixel methods [41,42,43,44].…”

Section: Discussionmentioning

confidence: 88%

Land cover classification at three different levels of detail from optical and radar Sentinel SAR data: a case study in Cundinamarca (Colombia)

Florez¹,

Lizarazo²

2020

DYNA

View full text Add to dashboard Cite

show abstract

“…Parameter yang digunakan dalam prosedur segmentasi multiresolusi antara lain scale, shape, dan compactness (Anonim, 2014). Parameter yang paling penting adalah scale parameter dimana parameter ini menentukan seberapa banyak jumlah piksel yang menyusun satu buah objek (Lizarazo dan Elsner, 2011;Parsa, 2013;Ureyen dkk., 2014). Menurut Sabins (1986), parameter skala merupakan selisih antara digital number tertinggi dan terendah dalam suatu objek.…”

Section: Pengolahan Dataunclassified

Peningkatan Akurasi Interpretasi Foto Udara Menggunakan Metode Pembobotan Berbasis Objek untuk Pembuatan Peta Skala 1:5000

Lumbantobing

Wikantika

Harto

2018

jrg

View full text Add to dashboard Cite

ABSTRAK Kebutuhan akan adanya pengembangan metode untuk meningkatkan akurasi dari interpretasi objek memerlukan kajian metodologi yang disebut analisis citra berbasis objek. Penelitian ini ditujukan untuk menentukan dan menganalisis akurasi dari interpretasi objek secara otomatis dengan metode berbasis objek dengan memberikan bobot yang berbeda untuk setiap kanal. Data yang digunakan adalah foto hasil pemotretan udara format menengah (medium format) dengan resolusi 16 cm. Ekstrak data menggunakan teknik object based image analysis (OBIA). Data diproses berdasarkan bobot yang yang berbeda untuk setiap kanal. Nilai akurasi ditentukan berdasarkan overall accuracy. Overall accuracy merupakan hasil validasi klasifikasi objek dengan ground truth yang diperoleh dari peta garis skala 1:5000 yang diinterpretasi secara visual. Hasil penelitian menunjukkan terjadi peningkatan nilai akurasi dengan pendekatan OBIA jika setiap kanal diberikan bobot yang berbeda dibandingkan dengan bobot yang sama. Peningkatan akurasi paling tinggi dengan bobot (Red=3, Green=4, Blue=3, IR=4, dan DEM= 3) menghasilkan akurasi 85,88%. Hasil akurasi meningkat sebesar 10,27 % dibandingkan dengan interpretasi tanpa pembobotan. Kata kunci: Interpretasi, Peta 1:5000, Klasifikasi, OBIA, Pembobotan, AkurasiABSTRACT Interpretation of imagery or aerial photo is an attempt to understand or interpret imagery to obtain accurate information and in accordance with the recorded object. The need for developing methods to improve the accuracy of the object interpretation requires assessment methodology which is called as object based image analysis. This study aimed at determining and analyzing the accuracy of the interpretation of the object automatically using object based method by giving different weights to each band. The data used were medium format aerial photos with a resolution of 16 cm. The method of data processing was object based image analysis (OBIA). Data were processed by different weights for each band. Accuracy value is determined based on the overall accuracy. Overall accuracy is the result of the validated object classification with ground truth obtained from the map of 1:5000 which were interpreted visually. The research results showed that the value of the accuracy with OBIA approach increased if each band is given different weights compared with the same weight. The highest accuracy was achieved with weights (Red=3, Green=4, Blue=3 , IR=4, and DEM=3), and resulted overall accuracy 85,88%. Results accuracy increased 10,27% compared with the interpretation without weighting. Keywords: Interpretation, Map 1:5000, Classification, OBIA, Weighting, Accuracy

show abstract

“…This criterion defines the similarity between the pixels and can be chosen to consider the intensity values [Narkhede, 2013] of pixels in a neighborhood of a predefined seed point. In the Multiresolution Segmentation method of Definiens software, for example, the region growing criterion consists of spectral (color) and shape heterogeneity which are combined by a fusion value [Lizarazo et al, 2011]. The shape criterion is defined by the parameters compactness and smoothness.…”

Section: Introductionmentioning

confidence: 99%

SAR Image Segmentation Using Morphological Attribute Profiles

Boldt

Thiele

Schulz

et al. 2014

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

ABSTRACT:In the last years, the spatial resolution of remote sensing sensors and imagery has continuously improved. Focusing on spaceborne Synthetic Aperture Radar (SAR) sensors, the satellites of the current generation (TerraSAR-X, COSMO-SykMed) are able to acquire images with sub-meter resolution. Indeed, high resolution imagery is visually much better interpretable, but most of the established pixel-based analysis methods have become more or less impracticable since, in high resolution images, self-sufficient objects (vehicle, building) are represented by a large number of pixels. Methods dealing with Object-Based Image Analysis (OBIA) provide help. Objects (segments) are groupings of pixels resulting from image segmentation algorithms based on homogeneity criteria. The image set is represented by image segments, which allows the development of rule-based analysis schemes. For example, segments can be described or categorized by their local neighborhood in a context-based manner. In this paper, a novel method for the segmentation of high resolution SAR images is presented. It is based on the calculation of morphological differential attribute profiles (DAP) which are analyzed pixel-wise in a region growing procedure. The method distinguishes between heterogeneous and homogeneous image content and delivers a precise segmentation result.

show abstract

Segmentation of Remotely Sensed Imagery: Moving from Sharp Objects to Fuzzy Regions

Cited by 10 publications

References 65 publications

Land cover classification at three different levels of detail from optical and radar Sentinel SAR data: a case study in Cundinamarca (Colombia)

Land cover classification at three different levels of detail from optical and radar Sentinel SAR data: a case study in Cundinamarca (Colombia)

Peningkatan Akurasi Interpretasi Foto Udara Menggunakan Metode Pembobotan Berbasis Objek untuk Pembuatan Peta Skala 1:5000

SAR Image Segmentation Using Morphological Attribute Profiles

Contact Info

Product

Resources

About