01540nam a2200229   4500001000700000090001100007100004100018101000800059200016500067210003300232215002000265300023000285330050300515606003001018606002601048606003501074606001701109606002101126700003401147955001101181995011801192393665  a393665  a20151223d2015       y0pory50      ba  aeng  aAutomatic signature verificationbRecurso eletrónicoebridging the gap between existing pattern recognition methods and forensic sciencefMuhammad Imran Malik  aKaiserslauternc[s.n.]d2015  a1 CD-ROMd12 cm  aTese de Doutoramento em Engenharia apresentada ao "Department of Computer Science", "Technische Universitat Kaiserslautern", tendo como orientadores Andreas Dengel, Marcus Liwicki. Ficheiro de 63,6 MB em formato PDF (248 p.).  a1. Introduction. 2. Automatic Signature Verification. 3. Forensic Signature Comparison: The State-of-the-Art. 4. Approach. 5. Validation. 6. Novel Local Features based Systems for Forensic Signature Verification. 7. Stability Analysis for Signature Verification. 8. Man vs. Machine: A Comparative Analysis for Forensic Signature Verification. 9. Signature Segmentation from Document Images. 10. Hyper-spectral Imaging for Signature Analysis. 11. Associated Research. 12. Conclusion and Future Work.  aEXAME À ESCRITA9343070  aESCRITA MANUAL957532  aANÁLISE DE DOCUMENTO9343080  aTESE9334284  aALEMANHA9333241  aMALIKbMuhammad Imran9357952  xOferta  104PJCD_270_00000000000000052022-06-066PJCD2709426129aPJcPJfPJ1CD270_21kPJCD270n2022-06-06o0rMOxOferta