OBJECT-BASED ANALYSIS FOR URBAN LAND COVER MAPPING USING THE INTERIMAGE AND THE SIPINA FREE SOFTWARE PACKAGES

Antunes, Rodrigo Rodrigues; Bias, Edilson de Sousa; Costa, Gilson Alexandre Ostwald Pedro da; Brites, Ricardo Seixas

doi:10.1590/S1982-21702018000100001

Acessibilidade / Reportar erro

Brasil

Boletim de Ciências Geodésicas

Español English

Brasil

Español English

sumário « anterior atual seguinte »

Sumário

ARTICLE • Bol. Ciênc. Geod. 24 (1) • Mar 2018 • https://doi.org/10.1590/S1982-21702018000100001 copy

OBJECT-BASED ANALYSIS FOR URBAN LAND COVER MAPPING USING THE INTERIMAGE AND THE SIPINA FREE SOFTWARE PACKAGES

Análise baseada em objeto para mapeamento do uso do solo urbano utilizando os pacotes de software InterIMAGE e SIPINA

Authorship SCIMAGO INSTITUTIONS RANKINGS

Abstract:

In this work we introduce an object-based method, applied to urban land cover mapping. The method is implemented with two open-source tools: SIPINA, a data mining software package; and InterIMAGE, an object-based image analysis system. Initially, segmentation, feature extraction and sample selection procedures are performed with InterIMAGE. In order to reduce the time and subjectivity involved to develop the decision rules in InterIMAGE, a data mining step is then carried out with SIPINA. In sequence, the decision trees delivered by SIPINA are analysed and encoded into InterIMAGE decision rules for the final classification step. Experiments were conducted using a subset of a GeoEye image, acquired in January 01, 2013, covering the urban portion of the municipality of Goianésia, Brazil. Five decision tree induction algorithms, available in SIPINA, were tested: ID3, C45, GID3, Assistant86 and CHAID. The TAU and Kappa coefficients were used to evaluate the results. The TAU values obtained were in the range of 0.66 and 0.70, while those for Kappa varied from 0.65 to 0.69.

Keywords:
Object-Based Image Analysis; Data Mining; InterIMAGE; SIPINA

Classes	Input Band	Relative Band Weight	Compactness	Color	Scale

Metal roofing	0,1,2,3	1,1,1,1	0.8	0.4	90
Asbestos roofing	0,1,2,3	1,1,1,1	0.7	0.5	80
Clear ceramics roofing	0,1,2,3	1,1,1,1	0.5	0.5	80
Dark ceramics roofing	0,1,2,3	1,1,1,1	0.5	0.5	70
Swimming pools	0,1,2,3	1,1,1,1	0.8	0.4	60
Bare soil	0,1,2,3	1,1,1,1	0.5	0.5	60
Concrete pavement	0,1,2,3	1,1,1,1	0.5	0.5	60

Classes	Samples	Best rule	Confidence (Leaf)	Rejected rules (-50% confidence)
Asbestos roof	30	IF div4_3 < 2.06 and ratio3 >=0.20 and div4_1 >=2.00 and brightness < 993.93	100%	NÃO
Asbestos roof_1	30	IF div4_3 >=2.06 and ratio2 >=0.26 and mean1 >=266.58 and ratio3 < 0.15 and minpixel1 >=226.00	100%	NÃO
Ceramic_clear	30	IF div4_3 >=2.06 and ratio2 < 0.26 and mean3 >=317.63 and ratio2 < 0.23	100%	SIM (40%)
Ceramic_dark	30	IF div4_3 >=2.06 and ratio2 >=0.26 and mean1 < 266.58	100%	NÃO
Metallic	30	IF div4_3 < 2.06 and ratio3 >=0.20 and div4_1 < 2.00	100%	NÃO
Metallic_1	30	IF div4_3 < 2.06 and ratio3 < 0.20 and ratio1 >=0.22 and compacity < 0.07	100%	SIM (17%)
Metalica_2	4	IF div4_3 < 2.06 and ratio3 < 0.20 and ratio1 < 0.22 and brightness >=916.20	100%	SIM (33%)
Metalica_3	30	IF div4_3 < 2.06 and ratio3 < 0.20 and ratio1 < 0.22 and brightness < 916.20 and ratio3 < 0.17	50%	SIM (17% e 20%)
Pool	2	IF div4_3 < 2.06 and ratio3 < 0.20 and ratio1 >=0.22 and compacity >=0.07	33%	SIM (33%)
Bare soil	30	IF div4_3 >=2.06 and ratio2 < 0.26 and mean3 >=317.63 and ratio2 >=0.23	60%	SIM (40%)
Concrete pavement	25	IF div4_3 >=2.06 and ratio2 >=0.26 and mean1 >=266.58 and ratio3 < 0.15 and minpixel1 < 226.00	100%	SIM (40%)

Algorithms	Nodes	Leaf	Depth	Reject rules (<50% confidence)
ID3	31	16	6	8
C45	27	14	7	7
GID3	31	16	7	5
Assistent 86	21	11	6	12
CHAID	39	20	8	4

Universidade Federal do Paraná Centro Politécnico, Jardim das Américas, 81531-990 Curitiba - Paraná - Brasil, Tel./Fax: (55 41) 3361-3637 - Curitiba - PR - Brazil
E-mail: bcg_editor@ufpr.br

Acompanhe os números deste periódico no seu leitor de RSS

Algorithm	Global accuracy (%)	TAU	Kappa
ID3	70	0.66	0.65
C45	72	0.68	0.67
GID3	73	0.70	0.69
Assistent 86	73	0.70	0.69
CHAID	73	0.70	0.69