Projekt inovuje oblast forenzního zkoumání materiálů využitím možností umělé inteligence(AI) pro exaktní a bezchybné vyhodnocení big dat, která produkují moderní analytické systémy a které začínají být nad možnostmi operátora, např. pro zjištění nepatrných rozdílů mezi originálem a spornou stopou, nebo úmyslnou modifikací. Aplikace schopností na míru vytvořeného SW AI pro možnost zpracování objemů dat ze systému pro nedestruktivní analýzu materiálů, zahrnujícího nedestruktivní metody (spektrální rentgen, počítačovou tomografii, ultrazvuk, rentgenovou fluorescenci, rentgenovou difrakci a multispektrální zobrazení). Zpracování datových souborů v reálném čase a vyhledávání anomálií pomocí SW AI v podobě zcela nových technologických postupů, které budou unikátní jak v rámci ČR, tak i ve světě.

Description in English
The project fundamentally innovates the field of forensic analysis by using Artificial Intelligence (AI) for exact and error-free evaluation of big data produced by modern analytical systems (e.g. developed prototype system for robotic multimodal non-destructive analysis, based on X-ray imaging technology, multispectral imaging, XRF, XRD, laser ranging, VNIR, SWIR, UV and others - VB01000046). Exact data evaluation, anomaly detection and match design is becoming beyond the capabilities of the operator/expert (e.g. to detect tiny differences between the original and the suspect trace). The application of customised AI software for real-time processing of datasets and anomaly finding in the form of new technological approaches is unique not only in the Czech Republic but in the world.

Keywords
zpracování obrazu, umělá inteligence

Key words in English
artificial intelligence, computer vision, defectoscopy, forensic science

VK01010153

Czech

Burget Radim, prof. Ing., Ph.D. - principal person responsible

Department of Telecommunications
- responsible department (16.5.2022 - not assigned)
Department of Telecommunications
- beneficiary (16.5.2022 - not assigned)

NANDI, T.; GUPTA, S.; KAUSHAL, A.; DUTTA, M.; BURGET, R.; JEŽEK, Š. Semantic Fusion of Text and Images: A Novel Multimodal-RAG Framework for Document Analysis. In ICUMT 2024; 16th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops. International Congress on Ultra Modern Telecommunications and Workshops. Meloneras: 2024. p. 106-110. ISBN: 978-3-8007-6544-7.
Detail

BURGET, R.; JEŽEK, Š.; JONÁK, M.; MYŠKA, V.; MEZINA, A.; DORAZIL, J.; SMÉKAL, Z.; KOTRLÝ, M.; ČEJKA, P.; TURKOVÁ, I.; JUNEK, M.: AI pro analýzu struktur a charakteristik 2D objektů. URL: https://www.utko.fekt.vut.cz/ai-pro-analyzu-struktur-charakteristik-2d-objektu. (Software)
Detail

MEZINA, A.; SCHILLER, V.; BURGET, R. ForgAnoNet: A Neural Network for Anomaly Detection in Artworks Using X-ray and Visible Spectrum Imaging. JOURNAL OF CULTURAL HERITAGE, 2025, vol. Volume, no. 76, p. 29-38.
Detail

MEZINA, A.; BURGET, R.; KOTRLÝ, M. A deep learning approach for anomaly detection in X-ray images of paintings. Heritage Science, 2025, vol. 13, no. 5, p. 1-11. ISSN: 2050-7445.
Detail

SCHILLER, V.; BURGET, R.; MEZINA, A.; GENZOR, S.; MIZERA, J. xU-NetFullSharp: The Novel Deep Learning Architecture for Chest X-ray Bone Shadow Suppression. Biomedical Signal Processing and Control, 2025, vol. 100, no. B, p. 1-20. ISSN: 1746-8094.
Detail

BURGET, R.; CHAUHAN, R.; KARNATI, M.; DUTTA, M. Plant Disease Identification Using a Dual Self-Attention Modified Residual-Inception Network. In 2023 15th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT). Ghent: 2023. p. 170-175. ISBN: 979-8-3503-9328-6.
Detail

SINHA,H..; KARNATI, AGGARWAL, G.;M.; DUTTA, M.K.; MEZINA, A., BURGET, R. DMRBNet: Dilated Multi-scale Residual Block-based Deep Network for Detection of Breast Cancer from MRI Images. In 15th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT). Ghent: 2023. p. 38-43. ISBN: 979-8-3503-9328-6.
Detail

GUPTA, S.; NANDI, T.; KAUSHAL, A.; DUTTA, M.; MEZINA, A.; BURGET, R. DeepMedFuseX: Explainable DeepFake Cancer CT Scan Classification with Multi-Scale Attention and Transfer Learnin. In ICUMT 2024; 16th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops. Meloneras: 2024. p. 1-6. ISBN: 978-3-8007-6544-7.
Detail

MYŠKA, V.; MEZINA, A.; VANĚK, P.; BURGET, R.; GENZOR, S.; MIZERA, J.; ŠTÝBNAR, M.; KIAC, M.; FROLKA, J. CovidStopHospital: e-Health Service for X-Ray-Based COVID-19 Classification and Radiologist-Assisted Dataset Creation. In 15th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT). Ghent: 2023. p. 62-67. ISBN: 979-8-3503-9328-6.
Detail

MEZINA, A.; BURGET, R.; KOTRLY, M. Defect Detection in Battery Cells Using U-Net-Based Neural Networks. In 2025 17th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT). Italy: IEEE, 2025. p. 164-169. ISBN: 979-8-3315-7675-2.
Detail

Responsibility: Burget Radim, prof. Ing., Ph.D.