%PDF-1.4 % 1 0 obj << /Type /Pages /Count 1 /Kids [ 2 0 R ] >> endobj 2 0 obj << /Type /Page /Parent 1 0 R /Resources << /ProcSet [ /PDF /Text /ImageB /ImageC /ImageI ] /Font << /QuickPDFF291673dd 74 0 R /QuickPDFFe394841f 115 0 R >> >> /Contents [ 5 0 R ] /MediaBox [ 0 0 595.2756 841.8898 ] /CropBox [ 0 0 595.2756 841.8898 ] >> endobj 3 0 obj << /Type /Catalog /Pages 1 0 R /Metadata 116 0 R >> endobj 4 0 obj << /Producer (Debenu Quick PDF Library 11.15 \(www.debenu.com\)) /Creator (Debenu Quick PDF Library 11.15 \(www.debenu.com\)) /CreationDate (D:20210628124621+02'00') /ModDate (D:20210628124621+02'00') >> endobj 5 0 obj << /Length 3201 /LC /iSQP >> stream 0 Tr /QuickPDFFe394841f 14 Tf 0 0 0 rg 100 Tz 0 Tw 0 Tc 0 Ts BT 1 0 0 1 77.9218 770.8329 Tm (Developing property and tax registration using AI algorithms: automated)Tj 1 0 0 1 154.3058 752.1709 Tm (mapping the Municipality of Kendal, Indonesia,)Tj ET /QuickPDFFe394841f 11 Tf 0 0 0 rg BT 1 0 0 1 128.0453 717.2179 Tm (Christopoulos Konstantinos \(Netherlands\) and Jati Pratomo \(Indonesia\))Tj ET /QuickPDFFe394841f 11 Tf 0 0 0 rg BT 1 0 0 1 56.6929 675.188 Tm 14.663 TL (Key words: )' ET /QuickPDFF291673dd 11 Tf 0 0 0 rg BT 1 0 0 1 141.7323 675.188 Tm 14.454 TL (Cadastre; Capacity building; Digital cadastre; Property taxes)' ET /QuickPDFFe394841f 11 Tf 0 0 0 rg BT 1 0 0 1 56.6929 646.8416 Tm 14.663 TL (SUMMARY)' ET /QuickPDFF291673dd 11 Tf 0 0 0 rg BT 1 0 0 1 56.6929 618.4951 Tm 14.454 TL (Amid the COVID-19 global pandemic, tax \(i.e., property tax\) remains crucial as a source of funding for)' (public services. Updated and reliable data of property tax enormously influences the efficiency of property)' (tax collection over time. However, many developing countries, including Indonesia, show inefficient and)' (inaccurate property tax collection due to unavailability, obsolesce, and low accuracy of the tax object data.)' (The development of Artificial Intelligence allows improving property tax collection by giving more reliable)' (source of tax object. We are helping the municipality of Kendal a fast-growing municipality in Central)' (Java, by providing an updated tax object information by utilizing AI. To serve this purpose, we utilized)' (Generative Adversarial Networks \(GANs\). LeCun \(2014\) declared that GANs is the most interesting idea in)' (the last ten years in Machine Learning. GANs is developed upon a deep learning model with a significant)' (model variation to obtain data distribution through unsupervised learning and generate more actual data. In)' (this case study, we examined the use of GANs for a binary training and classification of building and)' (non-building objects. For the creation of the training data, we used high-resolution \(0.30m\) imagery from)' (WorldView-3 and we manually digitized the 2 training sample classes. The workflow included also some)' (pre- and post-processing steps. The model was trained with 50 epochs and the results gave a satisfactory)' (82% of accuracy. Since the project is still ongoing, no solid conclusions have been drawn, but trying using)' (more training data and epochs would possibly achieve more accurate results for the tax object information)' (\(building footprints\). Furthermore, the produced data will be implemented in tax registration processes. )' ET /QuickPDFF291673dd 10 Tf 0 0 0 rg BT 1 0 0 1 56.6929 163.8497 Tm 13.14 TL (__________________________________________________________________________________________)' ()' (Developing property and tax registration using AI algorithms: automated mapping the Municipality of Kendal,)' (Indonesia, \(11272\))' (Christopoulos Konstantinos \(Netherlands\) and Jati Pratomo \(Indonesia\))' ()' (FIG e-Working Week 2021)' (Smart Surveyors for Land and Water Management - Challenges in a New Reality)' (Virtually in the Netherlands, 2125 June 2021)' ET endstream endobj 6 0 obj << /Length 316 /Filter /FlateDecode >> stream xMRKnCAۿSpk`y^Uu[jIɒT%C/:b}Rx\%=
c%d:WMQ&1֎PP/3k v'[-A:;NcC瑊Sd3G.#Oss
endstream
endobj
7 0 obj
<<
/Length 839
/Filter /FlateDecode
>>
stream
x=UM9)||Lfv$:<$HU.Uي'{mgZᙉ0c>_C.K.+r˄C,ZMdw:每G^*xCPEYOHb0ww.s%/a2ӯ4gHr^1-- }q]'U"sL|@
_tn6)Glȉv;A=cY:K?c ڠ|iH~eM`"G Jil$'~q|=|NuP@ ;ཁ"Z #p,NsDg@&86#]): {eH j9¶(um4bIהUbٗBV9b4/_d?p yBI EQ_(y0%Yq 4&_JF(
2iawq`;YP>8P LyP:iϼ{1xM