fktd 0.1.0__tar.gz

fktd-0.1.0/.gitignore

@@ -0,0 +1,28 @@
+# Python
+__pycache__/
+*.py[cod]
+*.egg-info/
+.eggs/
+
+# Build artifacts
+build/
+dist/
+
+# Virtual env
+.venv/
+venv/
+
+# Test / lint caches
+.pytest_cache/
+.ruff_cache/
+.mypy_cache/
+
+# Editor
+.vscode/
+.idea/
+
+# OCR model caches
+~/.EasyOCR/
+
+# Sample KTP images (privacy: may contain personal data) — excluded by default
+image-test-sample/

fktd-0.1.0/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 rezayw
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

fktd-0.1.0/PKG-INFO

@@ -0,0 +1,161 @@
+Metadata-Version: 2.4
+Name: fktd
+Version: 0.1.0
+Summary: FKTD — Fraud KYC KTP Detector: OCR + strict cross-validation for Indonesian ID cards (KTP).
+Project-URL: Homepage, https://github.com/rezayw4/Fraud-KYC-KTP-Detector
+Project-URL: Issues, https://github.com/rezayw4/Fraud-KYC-KTP-Detector/issues
+Author: rezayw4
+License: MIT
+License-File: LICENSE
+Keywords: fraud-detection,indonesia,ktp,kyc,ocr,tesseract
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python :: 3
+Classifier: Topic :: Scientific/Engineering :: Image Recognition
+Requires-Python: >=3.9
+Requires-Dist: easyocr>=1.7
+Requires-Dist: numpy>=1.26
+Requires-Dist: opencv-python>=4.10
+Requires-Dist: pillow>=10.4
+Requires-Dist: pydantic>=2.8
+Requires-Dist: python-dateutil>=2.9
+Provides-Extra: dev
+Requires-Dist: pytest>=8.0; extra == 'dev'
+Requires-Dist: ruff>=0.6; extra == 'dev'
+Provides-Extra: paddle
+Requires-Dist: paddleocr>=2.8; extra == 'paddle'
+Requires-Dist: paddlepaddle>=2.6; extra == 'paddle'
+Description-Content-Type: text/markdown
+
+# FKTD — Fraud KYC KTP Detector
+
+OCR multi-engine + validasi silang **ketat** untuk KTP Indonesia. Bukan sekadar
+cek field kosong — FKTD memanfaatkan struktur NIK untuk mendeteksi data palsu.
+
+## Flow
+
+```
+upload gambar ──▶ OCR (EasyOCR / PaddleOCR) ──▶ ekstraksi field ──▶ deteksi fraud ──▶ FraudResult
+```
+
+## Instalasi
+
+```bash
+pip install fktd            # langsung lengkap: EasyOCR + deteksi fraud
+```
+
+Cukup `pip install fktd` — EasyOCR sudah termasuk, jalan di mana saja (termasuk
+macOS / Apple Silicon), tanpa perlu install binary sistem.
+
+Mau engine PaddleOCR (akurasi ekstra, ideal di server Linux)? Tambahkan opsional:
+
+```bash
+pip install 'fktd[paddle]'  # opsional; paddlepaddle bisa rewel di Apple Silicon
+```
+
+| Engine | Status | Kapan dipakai |
+|---|---|---|
+| **EasyOCR** | Sudah termasuk | Default. Mulus di Mac, native bahasa Indonesia. |
+| **PaddleOCR** | Opsional `[paddle]` | Akurasi & kecepatan terbaik untuk dokumen; paling pas di Linux. |
+
+## Pemakaian
+
+```python
+from fktd import FKTD
+
+detector = FKTD(engine="easyocr")  # atau engine="paddle"
+result = detector.process("ktp.jpg")
+
+print(result.is_passed)           # True/False
+print(result.risk_score)          # 0 (aman) .. 100 (mencurigakan)
+print(result.summary())           # daftar check yang gagal
+print(result.data.model_dump())   # semua field hasil OCR
+```
+
+## Field yang diekstrak (20)
+
+Provinsi · Kota/Kabupaten · NIK · Nama · Tempat Lahir · Tgl Lahir · Jenis Kelamin ·
+Gol. Darah · Alamat · RT/RW · Kel/Desa · Kecamatan · Agama · Status Perkawinan ·
+Pekerjaan · Kewarganegaraan · Berlaku Hingga · Tempat & Tgl Dikeluarkan ·
+Foto (bool) · Tanda Tangan (bool)
+
+## Aturan deteksi fraud
+
+**Kelengkapan** — semua field wajib harus terisi (kosong = CRITICAL).
+
+**Cross-check NIK** (NIK ter-encode):
+
+| Cek | Severity |
+|---|---|
+| NIK tepat 16 digit | CRITICAL |
+| Tgl lahir di NIK == field Tgl Lahir | CRITICAL |
+| Gender di NIK (DD>40 = perempuan) == field Jenis Kelamin | CRITICAL |
+
+**Validasi lokasi (via API wilayah)** — kode di NIK dicocokkan ke data wilayah
+resmi (API emsifa + snapshot offline 514 kab/kota):
+
+| Cek | Severity |
+|---|---|
+| Kode provinsi (2 digit) valid & cocok field Provinsi | CRITICAL |
+| Kode kab/kota (4 digit) valid & cocok field Kota/Kabupaten | CRITICAL |
+
+> Contoh fraud: NIK `3171...` = Jakarta Selatan, tapi kartu tertulis "JAKARTA
+> BARAT" (`3174`) → terdeteksi tidak cocok.
+
+**Enum (exact match) — nilai resmi saja:**
+
+| Field | Nilai resmi |
+|---|---|
+| Jenis Kelamin | LAKI-LAKI, PEREMPUAN |
+| Agama | ISLAM, KRISTEN, KRISTEN PROTESTAN, KATOLIK, HINDU, BUDHA, KONGHUCU (6 yang diakui) |
+| Status Perkawinan | BELUM KAWIN, KAWIN, CERAI HIDUP, CERAI MATI |
+| Kewarganegaraan | WNI, WNA |
+| Gol. Darah | A, B, AB, O, - |
+
+**Format teks KTP asli:**
+- Semua nilai field (setelah `:`) harus **HURUF KAPITAL**.
+- **Nama tidak boleh mengandung angka**.
+- RT/RW = `NNN/NNN`; tanggal lahir valid & wajar (umur 0–120, tidak di masa depan).
+- **Berlaku Hingga harus `SEUMUR HIDUP`** (e-KTP sejak 2016 berlaku seumur hidup;
+  tanggal kedaluwarsa = kartu lama/tidak valid).
+
+**OCR confidence** — confidence rendah pada NIK/Nama/Tgl Lahir menaikkan risiko.
+
+**Vision** — keberadaan foto (deteksi wajah) & tanda tangan.
+
+Hasil **PASS** bila: semua field wajib terisi, tidak ada kegagalan CRITICAL, dan
+`risk_score < risk_threshold` (default 25). Mode data wilayah diatur lewat
+`FKTD(wilayah_provider=...)`: `"auto"` (default, snapshot + API), `"api"` (live),
+`"local"` (offline).
+
+## ⚠️ Disclaimer
+
+**Khusus KTP elektronik (KTP-el / e-KTP).** FKTD dirancang untuk format **e-KTP**
+— standar KTP Indonesia sejak 2011 dengan masa berlaku **SEUMUR HIDUP**. Aturan
+seperti "Berlaku Hingga harus SEUMUR HIDUP" **tidak berlaku** untuk KTP non-
+elektronik / KTP lama (sebelum 2011) yang masih memuat tanggal kedaluwarsa —
+kartu semacam itu akan ditandai gagal oleh FKTD karena memang sudah tidak berlaku.
+
+**Validasi bersifat struktural & heuristik, BUKAN verifikasi resmi:**
+
+- Pengecekan NIK & wilayah hanya memvalidasi **struktur dan konsistensi** data di
+  kartu. FKTD **tidak** mengakses database kependudukan resmi (Dukcapil), jadi
+  **tidak dapat memastikan** sebuah NIK benar-benar terdaftar atau milik seseorang.
+- Hasil `PASS` **tidak menjamin** KTP asli; hasil `FAIL` **tidak selalu berarti**
+  pemalsuan (bisa karena OCR meleset, kartu lama, atau foto berkualitas rendah).
+- `risk_score` adalah indikator bantu, bukan keputusan hukum.
+
+**Untuk KYC produksi**, hasil FKTD harus dikombinasikan dengan **verifikasi resmi**
+ke Dukcapil/penyedia berlisensi (mis. layanan verifikasi NIK pemerintah) dan,
+bila perlu, peninjauan manual.
+
+**Privasi.** KTP memuat data pribadi (NIK, nama, alamat). Pastikan pemrosesan
+mematuhi **UU PDP (UU No. 27/2022)** dan regulasi terkait: punya dasar hukum/
+persetujuan, batasi penyimpanan, dan amankan datanya. Library ini tidak mengirim
+gambar KTP ke mana pun; hanya pemanggilan API **data wilayah publik** (kode
+provinsi/kab-kota) yang dilakukan ke jaringan — dan itu pun bisa dimatikan dengan
+`wilayah_provider="local"`.
+
+## Lisensi
+
+MIT

fktd-0.1.0/PYPI_UPLOAD_TUTORIAL.pdf

@@ -0,0 +1,124 @@
+%PDF-1.4
+%���� ReportLab Generated PDF document (opensource)
+1 0 obj
+<<
+/F1 2 0 R /F2 3 0 R /F3 4 0 R /F4 5 0 R
+>>
+endobj
+2 0 obj
+<<
+/BaseFont /Helvetica /Encoding /WinAnsiEncoding /Name /F1 /Subtype /Type1 /Type /Font
+>>
+endobj
+3 0 obj
+<<
+/BaseFont /Helvetica-Bold /Encoding /WinAnsiEncoding /Name /F2 /Subtype /Type1 /Type /Font
+>>
+endobj
+4 0 obj
+<<
+/BaseFont /Symbol /Name /F3 /Subtype /Type1 /Type /Font
+>>
+endobj
+5 0 obj
+<<
+/BaseFont /Courier /Encoding /WinAnsiEncoding /Name /F4 /Subtype /Type1 /Type /Font
+>>
+endobj
+6 0 obj
+<<
+/Contents 12 0 R /MediaBox [ 0 0 595.2756 841.8898 ] /Parent 11 0 R /Resources <<
+/Font 1 0 R /ProcSet [ /PDF /Text /ImageB /ImageC /ImageI ]
+>> /Rotate 0 /Trans <<
+
+>> 
+  /Type /Page
+>>
+endobj
+7 0 obj
+<<
+/Contents 13 0 R /MediaBox [ 0 0 595.2756 841.8898 ] /Parent 11 0 R /Resources <<
+/Font 1 0 R /ProcSet [ /PDF /Text /ImageB /ImageC /ImageI ]
+>> /Rotate 0 /Trans <<
+
+>> 
+  /Type /Page
+>>
+endobj
+8 0 obj
+<<
+/Contents 14 0 R /MediaBox [ 0 0 595.2756 841.8898 ] /Parent 11 0 R /Resources <<
+/Font 1 0 R /ProcSet [ /PDF /Text /ImageB /ImageC /ImageI ]
+>> /Rotate 0 /Trans <<
+
+>> 
+  /Type /Page
+>>
+endobj
+9 0 obj
+<<
+/PageMode /UseNone /Pages 11 0 R /Type /Catalog
+>>
+endobj
+10 0 obj
+<<
+/Author (\(anonymous\)) /CreationDate (D:20260604012807+07'00') /Creator (\(unspecified\)) /Keywords () /ModDate (D:20260604012807+07'00') /Producer (ReportLab PDF Library - \(opensource\)) 
+  /Subject (\(unspecified\)) /Title (Tutorial Upload PyPI) /Trapped /False
+>>
+endobj
+11 0 obj
+<<
+/Count 3 /Kids [ 6 0 R 7 0 R 8 0 R ] /Type /Pages
+>>
+endobj
+12 0 obj
+<<
+/Filter [ /ASCII85Decode /FlateDecode ] /Length 2286
+>>
+stream
+GauHLD/\/e&H88.EF4G]>tmuAH[pQ9f4F9kL2;!`%J<OO(+lk=SO>'Q0^Q:;8A.EKm)3ks-9rPL,k(N`_rpGuJq:\6rjto+?7]N)@=h'Z\H1/+2D!b&pWJab:Pe!)5.,s/Sni["&/lBti'J<PA@U8">iJkG%8Pgml8^KL%`.V""XSrKdhCLNIrVnT6,>Cb$/-<5;7#r70o81JQ\',o`%ucOQ7:$1hCN$NMGas;8lg`:k4e?&c12KREllog02H$P$'nBA^!n3QAKW,`e:R'*`"28Q=Ot!??jXJ!P+75FH#rdrYUT,3pm)"MrM9ODi\!t+^]&cHmp3BKT'mLSUu9l`U!@$oJTiIIM<hbh&DODj9(<QZ9Xi((5-K;Xe0gRLg9,T\&oq2`O`d6J9)]@5\R')])33,.G;(#rG]u(@OX2Pi)46%6^L4kS7:'c\_h^A'F-.mFo3l%S!*^=<!QV`q1?tom3AtknC3oZ&)=E$HX^`,dM"rG+oC4U'kAF(p2tGJ;+0W_RfUd"cjuuRfp?N:6Ic;4uYCqiH,0`DqWjjJ\aigXAA&>N"Am5Od._78X/GJpQp-.o9@/,@A9Kb6bje)*&;"!aa(<kWJZB!8(i<OmVDYA#?!O-*G`8qjNU&^<UOsLD_%S\;(n@HmJ:kQ=aSBnE1E#SlqY&VbQqlc.9+pH^[$)P@i?dt@oZbMEB3&mLhVP@W@kSP6AIUMCT.l#',8R4N-3nJBWed_TQXt:<*=\#HAG`AG4HS=&AZ4^8EB+;k=$gZ*&'m8=*dG<>l<AOsEQ`i[:0gB[`Ob?Q8R*0!*YBbN>ZMKc8:r&5q;I/U`[`G0KY?EFba6`:30?$E!X:bZm@Kc1,GN#.dSqV;b",g-;:`@FVHl03g;/A&q>aDJ&5s!/D/NZ..l.5F>m(J=LF0nOZSH2cgkNrk2GF+^ZC1Ipkd3H'TgEG^M?'q/nW(stFZX2DlCM%2j2V/&1ZB__oc5)=5E#3?(cKe/5Wf'7kH+>m7TRfpe8Us9VoU,^E#WVnl<\ljd+Y'p+SY#L68XrQt"W%+32:s`>Eth_SCfE`m!(JWeG@>X=5)CdY-jAbu'G>kS'$?)"$N[J8DL!SJRUI4W^hq,tp8_$4[N7f8AY(8L,P>[i;Ke+Y*hQmVhP,<X='gfD:Z`]"'Qo8_T&?*]P&KnHAQ&*g'#dQ[Ae^.qPY!qjY[T>H^rJK]N#<DXNCqp_X/["&2OiuV2Cqtmo=B`M(@jAmh7@u07;E-/n(qPSS"aZ(8d\#1T.t->2jI\+B-MdA-efS$1:GaJ&P`,U)/!fgjYgpX=]^t;719o<,K^QDq%ZO.I"t;*a_R:<0o&Muh"M7&6DSil[XdX?nPGl!Sjdn@Pc%TM3mS9qRc>%LMc!lF+GqrSK`.Q]Jeem:+8Z1Z2sbdqFB`FI*U^M-]>)X?NR>cP2#UCW:$ofcN<`\4Ia06ETIbl!diRjn#!@Z7"g^RcJDRQ26e,KK'fYXk&:Z-H_q-K[Q$g/Q\76k;eY#<@SgMoDX'$o%10pe3S]A%'Y$dFC*S@,%W`>\32s[$X:l'5r-*XB$ZL!;hXg4n>_S^+*R_,[)H8W;"KTOaK1Je4ug6lZ;QW&\4Y_3Un]4VAIQPJ-TH9=qD/-7K9Z@/^pp&iNc6h@p:%!u?>TuB9SON#k87RCB_`?),VX6M`!6^.lC$))5TDB4J1cBpptG85nKK/]9Kq(+K^F$pdkJLoL'/@<TD[gV"Z3,bfFPZ'#"(0/EC1T=CN%^)8#m]jK\iGE(L8-j(d;s[/3E+!9"PM94-3nCoc?LTa!GU_0SO'.&s(($k7-f:[5Q5nUNpB_01L5@:V+AHq#6M=))XE[nSWScNrq$grQ-V>)EXA09,M$@=@k6!M1nGF<$db\7t!L(R3Jn^D/<MH!Zpspm6UV_gB_Ma(K807npl$O3]%Eog.UA*s'h"V%@UO`kdrF"5Y*H5s93;_urLWd,MR%6:G*Mr!Q./tl@.p&ltFYrE$RYTf>#)5mfW+`OZWK%]7VGZD9:N2C1Xi;GZWgV0"/'4PZ2:F=:]<>K:&7Qkf99E2'oIjK<(@o6''efU(CC8DbEPNZ=rm?72E/0)B_$;jga'`55];]fVl`#pe28C:u%Nb@)n@:G2s/l79\jB2aUIKoZ>hu[5g&@M-m[e&uV.uIuA)=n[8eG9m,-"/]*Ql:,#de=/\2r[S#$#J>=i!_P2nBtp%8Rhh[-?'i`<l;TY.'oK#I)]6<3,(BP,7l$P*RQ@b7/-o)Af"Y*^t.4D=Tm!`Jr[8)9c$hD598&iT,JE3hI8~>endstream
+endobj
+13 0 obj
+<<
+/Filter [ /ASCII85Decode /FlateDecode ] /Length 2586
+>>
+stream
+GauHLH#FW\(B<9aik`UbEY))hEK1aF5T7[WR^T9Jao[LO_Or`M7F,*HViL?4pQ77p8%2qTCV^nuO7s]Fn"pO0IVeU=blE24_;O#OgO[`\Cl`fpq?>`9cMfpN((EXO"(DML(YD!CaKd$4jP<t!RtJd0e>455L_;).5'W=M]Bb(00RIeo7.Go1%1<DWbHPqh"upA,Z;@#mNpAb<g(k!r0S!")^$?3(.i:?C!8!Ca0mrcj)u)**_A-."WX="XUO*ikBPH]uKUP1pr+upM@$*.s?neUCMI7l\/CT3i0O,%DGDMC^O7,shqV-krM,\-kJ%W\jY=*Pr?UuaZ18(4/*829&iT:OlnO)1;i`U0[3;CFImB:Ih,k[%O',&pg4l;'D-e,f#OD0gd,lc((%te#ZicR(Y'J)jiYX.1pQElJ2i,n\jNXB-1$<jk=N<jY.*t8CO4(Kr4LoY7C`a&-b+qj*\=QUjs1m^h1q1:n5kL2JG=MV"/@kPU%*F*s74hXrsC=&!D..poKM76\S)],sbrbX(ZXH4;uMT8(2)&faJ"<K&>p*'m,\#h&HXYe*PBd#AuEV*D@7%mW`)QDPkUE^Z!d0%`dpFUuCl'DdeT,0NVZ[>%[jsfN/n).X"Gm`Tg7-&UB9!Aq#<jDR?I(>FN;,!.bc`mEC*E8Ibd]&+-:ANr4IC_`i>=!8UGLtBt*#tN5RWD!]e<I`fT"pF6>eplAn?gL!@t&XBk/VW(q[p**WtT]L6IZ+s$I_#--=E)0i[?ME7J:UDG)-G-[_4NW:g<T$pA:pLCKX_3RpK6<]<O;BI<5#RC1YQ2@An6I-_Rh(9XmbuKpp82`[W8IoYc\=(,NqfHhcXgl`4CM_I"i/nbA9;J@?PBcM'H?qZK+>$b34c>5RWA=)&s.mM!IC<fuBG.D9VnFMBJS=]>H4pAu%$=>!3cU+O`b)pjnUC1h)C7%u$LomU>*8r`=r]M]S5/*+k6,>+o_OsNfJa8D.C-GZW\[*"GTa`/NFD&*&78cWMSU6s*njU%W@3oO.ZX+5m#5V>LYSh*<hbo-Ac9OY)'=hP^^1t4p'9aZ/)W'j[`HiW2H-`R2bjI_ZI419(plm=<j^Kd^_O[FU*kI@=Lat3>h8h"d<mcbO.Hi"VGPpl4Xg[XD$0m.hFR=$U<#Y5m8+lUeJHROf^:+EL\STi6iFCk87e,(#X`1rl7kap<t1oiKR[&KU(7ii->?_2S4:O6]=:G+Dk'U<-RXUkt=Ck4fnkh<$k[d:5_`BDe7VKgfI51$]YR#$VCYpOsi>gB"Do2AJ///!@ePe4*0;>]aF`XRQc1sbiIOMJ&)MH!RE0C=Yd,JGJ3_Fj8V/1u',Juisu_rJnjB&7:F=_UKbQmJ+co/lJFj@qPW0n?KFSXS13`^Hgf7c2%XX5-s7kISS7b[3]^7&#P!797KQ7kBrrlWP!+;I`$]'j:Bto$qmHo?15%g&-8Qnt/t"s/prq)a<6LE3@A[qqMmTV@nU^5KjYm,Vh6J@qZtBLkAVCIYFf`nM8V%]M9@@L#/>g%'.'E]\<'eR6b3*9N9\GD*-`p"O=8>rQW=Ke2Dqj!m4nKM_tDk9_Lor/u')[4t!T/ibVGO)O-TF]AMkH(SOK3LeH,=5%[p]Ih+IAAf.dS$K!-q&5a75H$WZCTe?gj%qY>7'%fC%5'Cgco:j%%_T5;1lWh&-73Y)78E<Y?"FU&H&:2NJ@C3I[+8K<*$qI6plQkaDljh.Lakkg\XC$X&[#9PX;Dr33Y>;_]9=,,NO3UL[XUR:f.J28:r#U!k=*Z]HYckl\VSd0cM^@13X&^Lj.pqcPRiJmP$;j22HDp15S]T_L'/j6;?-%kYZ#YSV1"Ek\n`*%j3QSV=le9U,<$[2=]#:81(+"a3Q5GpDmYS3cY:D5HQ>sFJ<I3osXf3,\mj>eCGg>@\X0>e\42#54AuI.Ol4Rme@_n61rXOMRAEYSS(+BKS9$ZE"n4sm<WoLlXR:B/+@%M&<CT0Ug@utIU<<BqC\b2]A65R42>>M"IS(QJ+JO?qDn$%*Ka>o]10\X.c`R>!H._edc;d)<Q(`c.U;4/NHknt-3`\Q+sUruDJ;MDP4XCOapi#U"G41roqmXXL4.dg4]C9JYH@S3o,'IcRb)XGXpd62G2RItO`)qmt_.p/5sdZ?"DDu?VVl-nFl*5V]^kYjp\dW[AU,/lHLN\q?XYDO=#TEZ-:jW939T^s.\3B+'#.'s)B@f\DBc5Z'Z"Aq]m#,UR-9Z(=WM_m<qRKQPkN@QV/27D?*.d2h<Ys;(u1IN$"Zdc@@GO<+GqAK>%&)DQ,TF\(sSip[HrMm'4F*#XP`;6^g=kHl_LJUnSF=5>KEn[>dcE):u:'PK5K1DYXK-ouK`-gC/_RF-op4>c?Rcs$J.I-C!$FIn%PnEsE4dU!tGb$<):6h3[Js!,blk3].:TtA!$Oe$8hOF8;>XMonh\#0I40YS^C:_/M_db_#`L,'h]'F*0grnhX!_;93dB34FhmRrh*=\KGn4opSl3Y+p.a/b+hXmH/YMO@P\2RmMK["W>.2u/iU,-3JRA(X9kGbuNWLc#i[Lp`j)GT:@-tG*NfY7G.SG`tB$Fg<~>endstream
+endobj
+14 0 obj
+<<
+/Filter [ /ASCII85Decode /FlateDecode ] /Length 826
+>>
+stream
+GasIda_oie&A@rk"pbRG#"3MIW?.PY?.+@pWeUYhd$?4=elGC0i-]rCMPAg=P)95gOu%%5]GWJt37op'$\7t+/6<5sLNj%t&de)0p^kc#cNmOjE!_3^:hg2f8!HFcphX!*Z]7AHKT_uP7CZAA84@JN"S`[SM]%-+5jKQr#=aI[HjkP'Cuh<-Tj36j%&63hUEM6t^:K?.i"tPp<o/!+>lVHilZ`k?Bq'"pb.iZ*(?3;heL+f%!"fP;N?C))-@-:ImL[@j*.HC_AJ8/ff:Pi(M8o1A!n,CW_A,pI"?.DWs,"%sGXbO:"Xkf+=!pKuI%R)XTa:l68O^QXY.1]TdcE@;;\3i&OJ*'fg6laNh1faH@d2?gHb<SN%W-6h`e6Sl;>LU7-_&l-Q68qM^1"i*-U!47.O'c9&/teV:LGVB,'kdu[**J90`g+n6&4DTR+KC40!ft6\#l^R!4OSZ]1q^^r=HG6I<:nk\?*-h!?*M+DYN'0RK7HqYW?pl>N?0]eF.87EcOYHH/d7G`*Y20,apJUmI/EabNgJC+UX<c&^HBO9jcbD#dEM,T8*.*WV/_p,K8]tpa`?eWY]f&/?JllkG!9Rk;TQf%^s3JTc1(M;'VUe]DkNT742PTnn.:<?0)IK[V<%UhY,1'QC'@i&]NmC1c!^nkJRpb4n+VBqEEa.oS.gcf<.Ri9l'>?JtT/#Dj(knS]_[[mp)8E17\4C!l'B+2=5Y'+DI,FY2!$@;139T<=(t]?o"8`:pAcRgV\QG>WfR^*hI<^pp@1BLCa7fD;)@FgPm6SDundsHerhLH2bF+g/gjiB\]$:0`(qP$VZc~>endstream
+endobj
+xref
+0 15
+0000000000 65535 f 
+0000000061 00000 n 
+0000000122 00000 n 
+0000000229 00000 n 
+0000000341 00000 n 
+0000000418 00000 n 
+0000000523 00000 n 
+0000000728 00000 n 
+0000000933 00000 n 
+0000001138 00000 n 
+0000001207 00000 n 
+0000001495 00000 n 
+0000001567 00000 n 
+0000003945 00000 n 
+0000006623 00000 n 
+trailer
+<<
+/ID 
+[<8f4ee96732a5a80fb288346d97b09bc5><8f4ee96732a5a80fb288346d97b09bc5>]
+% ReportLab generated PDF document -- digest (opensource)
+
+/Info 10 0 R
+/Root 9 0 R
+/Size 15
+>>
+startxref
+7540
+%%EOF

fktd-0.1.0/README.md

@@ -0,0 +1,132 @@
+# FKTD — Fraud KYC KTP Detector
+
+OCR multi-engine + validasi silang **ketat** untuk KTP Indonesia. Bukan sekadar
+cek field kosong — FKTD memanfaatkan struktur NIK untuk mendeteksi data palsu.
+
+## Flow
+
+```
+upload gambar ──▶ OCR (EasyOCR / PaddleOCR) ──▶ ekstraksi field ──▶ deteksi fraud ──▶ FraudResult
+```
+
+## Instalasi
+
+```bash
+pip install fktd            # langsung lengkap: EasyOCR + deteksi fraud
+```
+
+Cukup `pip install fktd` — EasyOCR sudah termasuk, jalan di mana saja (termasuk
+macOS / Apple Silicon), tanpa perlu install binary sistem.
+
+Mau engine PaddleOCR (akurasi ekstra, ideal di server Linux)? Tambahkan opsional:
+
+```bash
+pip install 'fktd[paddle]'  # opsional; paddlepaddle bisa rewel di Apple Silicon
+```
+
+| Engine | Status | Kapan dipakai |
+|---|---|---|
+| **EasyOCR** | Sudah termasuk | Default. Mulus di Mac, native bahasa Indonesia. |
+| **PaddleOCR** | Opsional `[paddle]` | Akurasi & kecepatan terbaik untuk dokumen; paling pas di Linux. |
+
+## Pemakaian
+
+```python
+from fktd import FKTD
+
+detector = FKTD(engine="easyocr")  # atau engine="paddle"
+result = detector.process("ktp.jpg")
+
+print(result.is_passed)           # True/False
+print(result.risk_score)          # 0 (aman) .. 100 (mencurigakan)
+print(result.summary())           # daftar check yang gagal
+print(result.data.model_dump())   # semua field hasil OCR
+```
+
+## Field yang diekstrak (20)
+
+Provinsi · Kota/Kabupaten · NIK · Nama · Tempat Lahir · Tgl Lahir · Jenis Kelamin ·
+Gol. Darah · Alamat · RT/RW · Kel/Desa · Kecamatan · Agama · Status Perkawinan ·
+Pekerjaan · Kewarganegaraan · Berlaku Hingga · Tempat & Tgl Dikeluarkan ·
+Foto (bool) · Tanda Tangan (bool)
+
+## Aturan deteksi fraud
+
+**Kelengkapan** — semua field wajib harus terisi (kosong = CRITICAL).
+
+**Cross-check NIK** (NIK ter-encode):
+
+| Cek | Severity |
+|---|---|
+| NIK tepat 16 digit | CRITICAL |
+| Tgl lahir di NIK == field Tgl Lahir | CRITICAL |
+| Gender di NIK (DD>40 = perempuan) == field Jenis Kelamin | CRITICAL |
+
+**Validasi lokasi (via API wilayah)** — kode di NIK dicocokkan ke data wilayah
+resmi (API emsifa + snapshot offline 514 kab/kota):
+
+| Cek | Severity |
+|---|---|
+| Kode provinsi (2 digit) valid & cocok field Provinsi | CRITICAL |
+| Kode kab/kota (4 digit) valid & cocok field Kota/Kabupaten | CRITICAL |
+
+> Contoh fraud: NIK `3171...` = Jakarta Selatan, tapi kartu tertulis "JAKARTA
+> BARAT" (`3174`) → terdeteksi tidak cocok.
+
+**Enum (exact match) — nilai resmi saja:**
+
+| Field | Nilai resmi |
+|---|---|
+| Jenis Kelamin | LAKI-LAKI, PEREMPUAN |
+| Agama | ISLAM, KRISTEN, KRISTEN PROTESTAN, KATOLIK, HINDU, BUDHA, KONGHUCU (6 yang diakui) |
+| Status Perkawinan | BELUM KAWIN, KAWIN, CERAI HIDUP, CERAI MATI |
+| Kewarganegaraan | WNI, WNA |
+| Gol. Darah | A, B, AB, O, - |
+
+**Format teks KTP asli:**
+- Semua nilai field (setelah `:`) harus **HURUF KAPITAL**.
+- **Nama tidak boleh mengandung angka**.
+- RT/RW = `NNN/NNN`; tanggal lahir valid & wajar (umur 0–120, tidak di masa depan).
+- **Berlaku Hingga harus `SEUMUR HIDUP`** (e-KTP sejak 2016 berlaku seumur hidup;
+  tanggal kedaluwarsa = kartu lama/tidak valid).
+
+**OCR confidence** — confidence rendah pada NIK/Nama/Tgl Lahir menaikkan risiko.
+
+**Vision** — keberadaan foto (deteksi wajah) & tanda tangan.
+
+Hasil **PASS** bila: semua field wajib terisi, tidak ada kegagalan CRITICAL, dan
+`risk_score < risk_threshold` (default 25). Mode data wilayah diatur lewat
+`FKTD(wilayah_provider=...)`: `"auto"` (default, snapshot + API), `"api"` (live),
+`"local"` (offline).
+
+## ⚠️ Disclaimer
+
+**Khusus KTP elektronik (KTP-el / e-KTP).** FKTD dirancang untuk format **e-KTP**
+— standar KTP Indonesia sejak 2011 dengan masa berlaku **SEUMUR HIDUP**. Aturan
+seperti "Berlaku Hingga harus SEUMUR HIDUP" **tidak berlaku** untuk KTP non-
+elektronik / KTP lama (sebelum 2011) yang masih memuat tanggal kedaluwarsa —
+kartu semacam itu akan ditandai gagal oleh FKTD karena memang sudah tidak berlaku.
+
+**Validasi bersifat struktural & heuristik, BUKAN verifikasi resmi:**
+
+- Pengecekan NIK & wilayah hanya memvalidasi **struktur dan konsistensi** data di
+  kartu. FKTD **tidak** mengakses database kependudukan resmi (Dukcapil), jadi
+  **tidak dapat memastikan** sebuah NIK benar-benar terdaftar atau milik seseorang.
+- Hasil `PASS` **tidak menjamin** KTP asli; hasil `FAIL` **tidak selalu berarti**
+  pemalsuan (bisa karena OCR meleset, kartu lama, atau foto berkualitas rendah).
+- `risk_score` adalah indikator bantu, bukan keputusan hukum.
+
+**Untuk KYC produksi**, hasil FKTD harus dikombinasikan dengan **verifikasi resmi**
+ke Dukcapil/penyedia berlisensi (mis. layanan verifikasi NIK pemerintah) dan,
+bila perlu, peninjauan manual.
+
+**Privasi.** KTP memuat data pribadi (NIK, nama, alamat). Pastikan pemrosesan
+mematuhi **UU PDP (UU No. 27/2022)** dan regulasi terkait: punya dasar hukum/
+persetujuan, batasi penyimpanan, dan amankan datanya. Library ini tidak mengirim
+gambar KTP ke mana pun; hanya pemanggilan API **data wilayah publik** (kode
+provinsi/kab-kota) yang dilakukan ke jaringan — dan itu pun bisa dimatikan dengan
+`wilayah_provider="local"`.
+
+## Lisensi
+
+MIT

fktd-0.1.0/TESTING.pdf

@@ -0,0 +1,124 @@
+%PDF-1.4
+%���� ReportLab Generated PDF document (opensource)
+1 0 obj
+<<
+/F1 2 0 R /F2 3 0 R /F3 4 0 R /F4 6 0 R
+>>
+endobj
+2 0 obj
+<<
+/BaseFont /Helvetica /Encoding /WinAnsiEncoding /Name /F1 /Subtype /Type1 /Type /Font
+>>
+endobj
+3 0 obj
+<<
+/BaseFont /Helvetica-Bold /Encoding /WinAnsiEncoding /Name /F2 /Subtype /Type1 /Type /Font
+>>
+endobj
+4 0 obj
+<<
+/BaseFont /Courier /Encoding /WinAnsiEncoding /Name /F3 /Subtype /Type1 /Type /Font
+>>
+endobj
+5 0 obj
+<<
+/Contents 12 0 R /MediaBox [ 0 0 595.2756 841.8898 ] /Parent 11 0 R /Resources <<
+/Font 1 0 R /ProcSet [ /PDF /Text /ImageB /ImageC /ImageI ]
+>> /Rotate 0 /Trans <<
+
+>> 
+  /Type /Page
+>>
+endobj
+6 0 obj
+<<
+/BaseFont /Symbol /Name /F4 /Subtype /Type1 /Type /Font
+>>
+endobj
+7 0 obj
+<<
+/Contents 13 0 R /MediaBox [ 0 0 595.2756 841.8898 ] /Parent 11 0 R /Resources <<
+/Font 1 0 R /ProcSet [ /PDF /Text /ImageB /ImageC /ImageI ]
+>> /Rotate 0 /Trans <<
+
+>> 
+  /Type /Page
+>>
+endobj
+8 0 obj
+<<
+/Contents 14 0 R /MediaBox [ 0 0 595.2756 841.8898 ] /Parent 11 0 R /Resources <<
+/Font 1 0 R /ProcSet [ /PDF /Text /ImageB /ImageC /ImageI ]
+>> /Rotate 0 /Trans <<
+
+>> 
+  /Type /Page
+>>
+endobj
+9 0 obj
+<<
+/PageMode /UseNone /Pages 11 0 R /Type /Catalog
+>>
+endobj
+10 0 obj
+<<
+/Author (\(anonymous\)) /CreationDate (D:20260604012807+07'00') /Creator (\(unspecified\)) /Keywords () /ModDate (D:20260604012807+07'00') /Producer (ReportLab PDF Library - \(opensource\)) 
+  /Subject (\(unspecified\)) /Title (Dokumen Pengujian) /Trapped /False
+>>
+endobj
+11 0 obj
+<<
+/Count 3 /Kids [ 5 0 R 7 0 R 8 0 R ] /Type /Pages
+>>
+endobj
+12 0 obj
+<<
+/Filter [ /ASCII85Decode /FlateDecode ] /Length 2264
+>>
+stream
+Gb!;e>>s99'Ro4HS8"l_cimk"Z=Wd4Ksk4LQF(27Za9#bLdVA)^8/Jraj0n\4d*'i6L4iQ02%$:q+0\OF$9m;(4N4X*^p'IL2d'N!bF'u!Qk_(iCA,Sh#"fU/B@?!#==-f"U2bjMu#'%PJ%eL$f<7jm;R#lLG7\s^`A#:gP/\Dc-&7\MBA@*6nk:"&]*l=i*1N2MaN??CmT]C/Ku`A^bCB=GNZkB\7/.I]2TPU3</"AcA)K]<Oj%08U:BSD_QQacV#59B`UD;+%tOD&[oRV1mo,AMIE1;8b"BoEJ#DHBY*R2dGN0g-Pr*/_jC+2H`aH1Z961e&:q_Z@8>i_#a]o%%bGc=+Pa8N'Mg\PW*B.p9B01]o8(XsmM_WD6%3$CP>Mr5Vs5A7\3%Emk2U#Ke/U#p3+@TX0/AcSN.En<I)]RGMo<;2J^JS,E%sqqF.jc)!j77@@,kM!N-X,26.%P2'hBF,fD$Ahp4k-@K6VEZ:oSTE]Z=cs?;Do6oQ9Une=J!lGhf05HId@[!jFE>OG:F2.a8%!#R4taM(S#?>.sSB/?bcI@jS+V<R[uE>1<20K)MC%-PIB3_M)Vk*!!=4$Xe)3s0mB+p3I;TfdeKBk7RYJQ[+/CO%!/](`DT[,PrJ@o+7E%F"NmV&(,@T%d0@)D^WdR!q'L4R:kPFDk.[CY-=AmT:2%p1"#\hZmgPZQuc2o/T(GEaVZD?,u-(t9rN:W.O)5BbC"DFWgjSR/<NOZ:cO1\:M$[U_'+cu[$\u1&B;4#DIjT:0WVZQ(gNso?,/gR7`i.G@>nn\`DjK!PBAj`XP!OaO-<[WJV7g(?6d&j/l',g1uHGlJU^rGi8T!5]]V$Y`]*uq,\dK]L%33?T0Abe6C:G(Bo[RI%mIcL:0GZm6bjCn0RpAV335R*,\\MY7n@,sk:.JWBn/\3Q^Oh*E:4dj/ANg]9lAc#CX;CC$JN-L6B>PhQl\S7Et\b_d]+PX8).oGRHS#MB8T;GAf.L<:1XD=)-tF-a^(YecBDJ\RKTcfK^^!$,:s2Y-N/$e;]?[O=X'g1^_-6Z\C);?4=kMgB+a#T#3'^Eni_B3(7):ASM.E\0>(WH*H*Ir[bm"TRCRKL[:cKL\Uk_<3tkuL1([0UnZXS_nZ.*,kmiVVYSN_1s#QXKBkt3r(,;t]H%,jeE2s-/70TgA1C`fM_/";E8%aUK0#NsVdr!)7_Pl3tKu\L[29f^=ik>.tBu/)+=YI@4(:^HT>M`"h[$qNW5sbgb!,hOAqHL9oU5R/tf\rX[P27C"K:p?ogL1`$H[G%%3R%rI:8k#n>4mQ.Z%Gk@JiMf;ekJ,;0-qI9ScTIS7d;Tt[jY1'7EcP$:"UU@@]to7L;2WWTGt2ICNR/OTqI`*1leWq>t)&b?<^'$)j=Hnl#,0"iijVpL=i/eECIf(fHB.O,+(jO+IR4<`:hb)iM]ep?a87e=&Ij*j.`ZS#7jj_DM+>3F_Z$'IY,hQK-+<j?UitK7_aY.Ja:Us/Js;soEcAUh[0AjV[X,P_@cXu@5Iq,,drPZ)`Cp[hi>W5Y]%Ek9AJ9afEUb\fPK5"56QfL^T+)*L.^W#1rI4'Hfe0Ac[;%bS!:,=d!cBBJXL<HXJQf#>?K7AHR))tQbZkCZ#.7IbsDT#5o=!:4Tq0B0(21hPZtg_h-cB&-JLq$#8oGI2ih5K<(OZ7PM8G&h@eQ=kE!FTpZM3$eDCGi'1!J4eR1+*da9XAV2SNI$.Tu_?d)Jsno_9%]`i@AmB2AX>ql(j0pZf#N&I`OK+Omhkh6EXU%kN0)D!f4hO?`]'@M#.g];<=9G'<(b?Pd-d!kPcNS$N4Oua8T0AO8mU2-"qg*!GsZn6mW;jb;&;c+=c8GF(.=O+5,,R9*FR*+aTBp-Zq&APu3I^_cCrW]I]?rWl;-IKK)aU%j66ladRmdnO"i!A:$AbaL^U)I"VBg"_GLQ*Bs5-P?`aAg$[.fLb^),I_&,s8+MR]<Gb?X@'.2f<?Q)ta)-I:&KkR'<QY-!?<Wl/L6*R?"hJFc=N\K2e+;d^^8,ck-/Z+XT1G2<DQkn:?DKi[[Xcb;)BO4^p?n<V[0"k<K?d.:`q[+O(93Wd#"<5nI\>$q$E(WQK4qN>6WIZ&0-5Nn(eVAhJuY*(;R5l\PPCM-3`P,?;H?^(0.43Qct7f"6_!_5rrEUu3-d':JmnXJ21Ec`nh4=V\,6Kgf2#5/%Z/NrDQ+=&rh\YP+/_(*p-nY7?lee5rj#QS3J2o?ni^4-L7#e;VV;gC<NJ#>1rQV#~>endstream
+endobj
+13 0 obj
+<<
+/Filter [ /ASCII85Decode /FlateDecode ] /Length 3355
+>>
+stream
+GauHOlZ:cN'*%C7U8tHjMX2Xp7W3--&UPN3HbJ8QDeeUJU';HBN#nK1nTk02lPY88#Y;7c@$_>N8H"qT9i#u676nh!,6(GeJ.G",^EPp"(s[[LM?1V!piUiTh_R\Y*ChnN+gtt7";K*\M/,Da3",4fgU._o2[qHH;pfF1!lJ",=P"^*0`UqWM;g0p/QlaP<67J]0Yk#*,E#,-<RWN/*24pklI[f_Tm(W8]S)EEfl.\RF9X[R?,4_S*\LU4VF_W<J`.7NJ)8Nb8iVERX'"8"0PjJ2p?5k<X41b_gp:CN5&hJT#V%a4Y+))?s-ek<L96msI/\r5kMkdJb;!!!HnZ!sdEKHV%j'#83\8>"5qu?p7J;Ui+2l=bUe9X:M<64t@2YAqH,5\U3`2cJIibl0l-&\oI$Mpu1QbU49uXL6#[C;hqD82SE6fV>m//2W_q%9X9[1!W,WK$OMgr;^B93ZJ[T#WAKd+S1L#-=T'1c5q^k>-2T@3q=#Qbm5gO%n,mdYhD5@K:)NdMo;38pp_FZGptL'jRKF@A3a/IoIt/:qO&d20@s5Feu1EBV(Y$`DK95cPV*F<*#%'J']*bT"6@$OI9n'\5*C/<FHD:Gh(IJd-/&Zdt&f'$i*WaOs7>,[J7*5UjtCiHZ_E@9tJ8=D>ne1#I!`R;rnM@ci#l!nT9k*luGDMTHn3E5#C1:?0`kUV^,H5he/=5[6,Z7P&f?_Bj*QMG5W8h@WgDpD:ER8ru/_S`38YTX]`U\dj0mWpLZ!SS6IVoMrLLIEcaMd+9hH/UbNR6=4fgdROOG:S5.3M7u`R;((L'\]XQ(,?!%=f8dpPU@,OnKK"tGXGi\p#-&@8kffuG61*J'hA00prh,`IJ%<j^)rkgmCPo`0jGqk(O:HSE\^(4pF<t_2@bEZE4-@9F+""\A;)Z?N'qCD(D#QbF$C;_q24nr3;>_h("+?tT3dN/sdPDs=a.OZe,ita4%o=V1:aQXq;,(<Q'%atV,;-FGnVg1^%iYb$2>4#i)VCmo17+.=g^S/:UFZ==r;-ZK;>?C]0n0Q=L7;Y*>N3N;$^NolrB$ZL"5@Z];TNHm);01b_4oRq?=V15a4nXi.U1%GgJ*4L^m]E,?X_8@',DTuW"ZKa$a,+o^IoWPAeFXd.>B:H?6V]SlG:4DFn>mUZ05I?>%=>q`&WBO2-4Rm\mB=))WGldrCT4:)m>FtGYR4c`V$*EYDpVL<Zs*935L!]_gH#af@pS_pu_.0Am0mR[larjrJ-0;&Fk!&Ni+&g@qr^1F4!clFA16=,2R!@p,_beL<$]d]Su$DH*d%2Ej(>]4/d6I(r8"ViY06q*9&##l34\-j8*VLJ?/\-(5Z:D*Xi8irS*2t@0_=R#,ctOVC(\BbA70/C>sBu(I!3^:<#JYI@<irj(ZFF^.g2[@F/2[@D;Q_\D4)7/8pa4QC"Oe$_&_S)PCZBkUEa1l`Jd)eGq1DjI>'RMkP141dFY<V*7L`*J*d<"[%MYk#%`U-NM@chbudp(bOq%D?-&=i_A?o"lE8P)Si^\l$a<1Ii$Fo'Dq<Ufd_@N]F`\.]c*Gf`70sR_bA"q_+\4UU;E>f0R*b=\u9cc-M74j:;.*V7QhVcA'PQj.#J,__OGiO'f8,YKIg3Pg.o$0!(0f3L-\MPW!(M$7@<qLj/f=Fp/?6S*9Y:r?L"F#%aWRG"'furpHp87>\=fF8\2dop#4NCh]>#"p#,E2rWa[C+T!U\@C115jETgq"%G0@3nFut_^O9()Q#R_[1P#:1GuGaV(?l^j*-Vjg/HX,C[Z"0K9B^rrAn0S+KK5'"_O+2WB@G"_Ia6NhEeE\i$C8m0;T7_i`Vl13dPJL]X:[!I^aqO,RmA0DHFRpoDbMdVr8iXJ$U#Aol0/L5s3Hk^8Pu#US*O>(.cYdX?YuYnEVOmiUQKMh`_I)R@ah",qS%>XiH-B;Qk"!*H9E!\nT\oS8*&K@a?l7d"!]]6tq_f"OB(\!^7*hnT/YsDKBc=dIsFg#hmh(R$3!0)k;Dgli1'o;3:YXQn[6[LId'/B/T2`eaA[@3=+(+RRL-n2g@+.>,aTDG2?a7F@K],7d1mnCs'H"7Rq)VK,JOqi6aMq!NUWNd%oIM[=0ERJS1Q*a"eug79n2*1&45kB;4L"Q3I_iieb])b"sIE$sH1"%B>_b@Xari0kObPH'j8gW,B)pk6fm-JuScB4HC29?`R*>OFDT8rsA^sW]CIVro^r#b27?/LdTQR(eF(K!O[#Q/R2Dfj9VH\0<7&\:ZTEfTrp)*L\[%u\1/!2'AHsC<`KtNL\[%uV()tt'D\lkQ3Qk5+-Ko-`Bebn+!r9s`Sktgl[LAmAgd.XF,t9`+s/^8KZ.@`V!SiHMM.N`\k\Bh\Mk%2E>nJt/Egt;Q$lBt]/4B2dY54,_9UoU<jPdT`q+fa?AP1)"0(_^?p"A.^9cNO0C":XpBr%V5V%":*qfblBYDW;JL_X17GI>ZdM#b+L3W##iJ;r^Dl2#a<REblI_(Tm4H+ZVi9k]JFNtTNdVK#P*l5gtiJ6\ZWj*q4-CUO.X[CmGD:t*,;tb[5g!@EiWrDX_4AmLAa<lHU@,S%7RU4:;%Oe3Jc1UPT-#+q!pr$SP,]"'L!it[ZrCHknVVIoRi$&Smn9D3_-^:bMKIWl5:1i#"Op0<P+TT3&^1O[_GKk_]M-8^((S>nh^5R`.P:mFID@]<,97'(uS%%.q>]c3?DE6`Qe>%Sn3miW#TYk4t02jka$J_k=@jLV3b?Zm:Dmq66bHI8q(BkrbHP+[+ZAe>675Vq/CeZFQT"!ttbh/G_Xh'1AoWP^VOrV37Y1^,9aEb/GOrU3%F9sfFP:jt@Y?*[_I:1H[*j8b'eRaU3!DK>Q6hW^^@2:e10]Tbl`a/k>j)L5Jcl_MC[/(F:a&a!]A$2]-%Qa,49_oVddtCb$9h_e@p*oqi=hc3(ds=l,TIsIjVPo5sfq@=jfZ#CHpf*OlEg)kJ7FE4+^?';aT8:UECnoi7oR5^NKso=^<]!FsN"4iio]A;94E&?h/YBYf06DaP@5\k\WlE78&CH_<pE+/od[Gk`4FAd<m&7lnHR=`fd@0ts18I@R"#;.=hF#L6FMSOu=W&R.EjQ.D;S_,#WpHk?>qE0<EXj+W8D?qgM#-cS]*"Bh`ii(+X5$%m)E8<5=ea6^#]Hu(G`3F9Xag66q8N$i_8C6=^VauB]=s5IZKEbUJE;+OAC:@hYU=\>Vh]Y2o1V%M4Gl.W4OGRH/RE`_T.L_?iCboc,/oLV03L)h\Jdp^`Jf^k"tV@DQ/K4O?=ZWL9-Q^b<1E?QEQ??U.PKEqm&i=:<]QTSmEG#3m99qq\a?3FG_`Jm6N;K7?aBFdp'_@Brs,2C$I&~>endstream
+endobj
+14 0 obj
+<<
+/Filter [ /ASCII85Decode /FlateDecode ] /Length 2059
+>>
+stream
+GauHL>?BQ=&:WeDbZdG=KJ0WIpLTD"BM6fP$L[?@D"^/OPskD5q>uHNofKGm`<cZFEjg`*7^j8[I%Nka1;UaJZhZ[X!DYs@ra:=l5r\n/&dmS63"6CI*Y?6Z4Gs7q>t/*5LufSQc_-XHA@U:b)0'JJK&\r2EZGA6+kX5?-j34M+.uI._:aeT$V[<m@4`1$YIW;D@hQ\bR0B0r&D-DV#ZKr8(kd;S>*]O<$=64V\:+s@n,iH\'&tGp%O&K9nF],61;ZrsI)6q>TWU_b%JfrbY#CJ!Kug.aNT=54+E>eNgB"&J>!@D7CH+L_IRMr`G&>aj3<k0O*(l(LFLi*p3^@mCO4cds@Jd!p;#H\SqrF^sCF,9:Cs"/Ap,nVjX<*LDSfS^-?[dKK4Gg+Ph%NtDi=O&c&FbN(4QRI5SMf::Cq[nqW-&\Ha)W&\"G_tK\D=Wn).]IP,un+-'=VP71?HS"#'2m3_E6tPVK6'0,]W9,(Ia&SiV[Kl_qr&IDe2sAGQTZB*QNHELPIj1S<*8$`a-Rc,$o9!P6o"e1F-"h8mE3<S:O4(3T4mb_NSNKa"`erT9lms)BP'<ECR9c:V;nugknEnA;pQ'>I2"Y4gU"2qMl#D=WQh#3scs)jq?9V%NLm#EAe>XRcJ_Gs6)<!2e%@5e1l=]^W^X[\m/qG4%O;d+.^tOXuY_3Vm(e9.%=Fn]17&K[mg31D*1@HZ7*`2'e$#)+q!u/<kg\3_5]Xt+D(P$8,q8B__b%r,8>L'(a&j?#CHXk(@%Ws(,SIg%1fr5[?uTo_PR,Z7c*/S.&qTo-JS3UJbEDVI'8R]+5;>JC-Qk#[:Jl[jLCX#REm\I`U[lph(F-_remT`^4cT8%L0`qI]lM]"CZ9K4QJ.r\(#L4>R'0YJ1pTRjb7[Z"-?*&"7Ho0N?#]`IsBNT'W3>D8pu<A%TrFfA1s[/ofNu_d"#Z"c]V]Iq.k(s)C/%34k/jk4ACWsY2;eW\d'>D*C@Y:a^?"@V<c4l;C2Mo5;%QWoGl_["m;M->j`.<rc_AC^p"rt!d-ia`!#[OQ*U;p$a,+P[!fsOS#;EX>^X$)J^RO"Jh<8fArYeF^Tl@,XQFY;4>*L&T/o>6HT?m2;J-8>bCLuNl3oQ;G#7AR8u$9;NEFug.MZcSF[H[Db\47^_n)OsD3q^+W+"La%"I0q^T3PCJ%)4p%9'i+e8uTor<?#`^]<\r@YEueM4+!!c_f'UZ)IH!QJ9aFE\EHQmM:0&<['^sn=t.K=D81<fA0>8YfF-Kgo)g[[^ImWQCuC;7?+S%MkrjI<X"Ph(k`S:q\b?W`MG+o`!]]hiMattbq;%79!>?Hk<.`Xaf?CZWMU\sl@PqD1:u^(9Vb3'PmC,tWdIXnaEa?`?!?kQ75j9G=!c+I@pP51"0pS*37!LRB"SaHr/P8:rWXQD`o<$Fi-XVd%c9'l0Ag+gat6o1.;+Wc>gL6`TM)n*cbZ&h<Sfk=K0X)C"eXCOJQLmK30+4;2M/nqL5"SW,^$-c5HbZs[QN&$=9th9)I43,)d3a+e>rnE1d==D5cFbF$L:5`4H$_&=3WA)L.Xp+NX-.n`fjK9J4Zm#!@G5J!qQJe;j@AeSuMQk"-SNUrILQ>7o(0QXT%Jt8tcT>FB\\QY.oCtDUY%UE*HdQE-`CLddWf5KkFagrsI?>iUNK%Pu*@CX=qQ?4>17lDoF>h`oq-Ld*^nj4P`l@iG"d+l]*kF3r2XD+G8$)1f_4r<(kMO:jV?g3j(.P,nGD%(.P^KB_@[rHm6D*-_Ohe@m64cSt/9p:j-33j+PTT'Q(4h4:V-o:HK(4gYGGLf!0;2bK*VK;(9t>SZKN1D,X0BNp=[6.,3JHo[LLi-rJJM0H&B.XIp2;"\G7-4X5&SI3V$iQ'`?"Tm'OM2g+`E<7Ha)N;#(@d,`'";(RSDNjLES49Gag&V$f)r?/c"9TF2TCPf-H,Z""%D=^;TnD]GG'0]?g,YqCd&8Ee)T=U,&"Fcg\Pqh,3G[&9L-P.KXj40\%Fn8jmdUZTa)"=n5KMr2tGhX(`\U%<N>(dCDJ?(06M)ZLQ#B,_^K)~>endstream
+endobj
+xref
+0 15
+0000000000 65535 f 
+0000000061 00000 n 
+0000000122 00000 n 
+0000000229 00000 n 
+0000000341 00000 n 
+0000000446 00000 n 
+0000000651 00000 n 
+0000000728 00000 n 
+0000000933 00000 n 
+0000001138 00000 n 
+0000001207 00000 n 
+0000001492 00000 n 
+0000001564 00000 n 
+0000003920 00000 n 
+0000007367 00000 n 
+trailer
+<<
+/ID 
+[<7242f470c8f00ae737cffeb3fee2725d><7242f470c8f00ae737cffeb3fee2725d>]
+% ReportLab generated PDF document -- digest (opensource)
+
+/Info 10 0 R
+/Root 9 0 R
+/Size 15
+>>
+startxref
+9518
+%%EOF

fktd-0.1.0/fktd/__init__.py

@@ -0,0 +1,21 @@
+"""FKTD — Fraud KYC KTP Detector.
+
+OCR multi-engine (EasyOCR / PaddleOCR) + validasi silang ketat untuk KTP Indonesia.
+"""
+
+from .detector import FKTD
+from .fraud import detect
+from .models import FraudCheck, FraudResult, KTPData, Severity
+from .nik import NIKInfo, decode_nik
+
+__version__ = "0.1.0"
+__all__ = [
+    "FKTD",
+    "detect",
+    "decode_nik",
+    "NIKInfo",
+    "KTPData",
+    "FraudResult",
+    "FraudCheck",
+    "Severity",
+]