COMPARATIVE ANALYSIS OF SOFTWARE SOLUTIONS FOR AUTOMATED RECOGNITION OF INTEGRATED CIRCUIT TOPOLOGY

Alexey Malakhanov; Kristina Azarova; Kirill Moiseyev

doi:10.30987/2658-6436-2026-2-10-20

UDC 621.3.049.771

COMPARATIVE ANALYSIS OF SOFTWARE SOLUTIONS FOR AUTOMATED RECOGNITION OF INTEGRATED CIRCUIT TOPOLOGY

Published in Automation and modeling in design and management · Volume 2026, Issue 2 · Pages 10–20 · Rubric: Automation and control of technological processes and production, automated design systems

DOI: https://doi.org/10.30987/2658-6436-2026-2-10-20

Received: 18.03.2026 Accepted: 14.04.2026 Published: 20.04.2026

Authors

Alexey Malakhanov

ORCID

Kristina Azarova

ORCID

Kirill Moiseyev

ORCID

Alexey Malakhanov

Academic degree: candidate of technical sciences

Science rang: docent

Bryansk State Technical University Russian Federation · Bryansk

ORCID

Kristina Azarova

Bryansk State Technical University Russian Federation

ORCID

Kirill Moiseyev

Bryansk State Technical University Russian Federation

ORCID

ABSTRACT

The aim of this study is a comparative analysis of commercial and proprietary software tools designed for the automated recognition of integrated circuit (IC) topology elements in reverse engineering tasks. The objective is to determine the specific error profiles of each tool under consideration, namely the commercial Pix2Net software package and an author-developed Python/OpenCV code and to formulate recommendations for their improvement. The research methods include experimental recognition of IC elements fabricated using the BiCDMOS process; error analysis based on the Pareto principle with classification into three categories, namely element omission, false positive detection, and geometric error. The novelty of the work lies in comparing neural network-based and template-matching approaches to recognizing IC topology elements, revealing a trade-off between detection completeness and accuracy. The research results show that Pix2Net achieves an accuracy of 98% but generates 78.9% false positives, whereas the proprietary approach demonstrates an accuracy of 72% with omissions dominating at 60% while maintaining a low level of false detections. The conclusions indicate that each approach possesses a specific error profile. Pareto analysis allows for targeted identification of priority areas for improvement. Furthermore, the proprietary tool proves promising for educational and research purposes due to its independence from GPU hardware and commercial licenses.

KEYWORDS

reverse engineering, integrated circuits, pattern recognition, Pix2Net, OpenCV, SEM images, netlist, GDSII, Pareto analysis, BiCDMOS

1. Kimura A.G., Elliott A.S., Perkins D.A. Recovery of a Hierarchical Functional Representation of an Integrated Circuit. Patent US No. 11651126B2; 2023.

2. Dizon-Paradis O, Wilson R, Koblah D, et al. Hands-On Introduction to AI in Hardware Security: IC Reverse Engineering Using Image Processing, Computer Vision, and Machine Learning. Florida Institute for National Security; 2025.

3. Luchinin V.V., Gasnikov A.O. Analysis of the Possible Methods for Search and Identification of the Unit Crystal Radio Channel on the Integrated Circuits. Proceedings of Saint Petersburg Electrotechnical University. 2012;8:14-19.

4. Lippmann B., Junghanns A. System and Method for Integrated Circuit Planar Netlist Interpretation. Patent US No. 7937678B2; 2011.

5. MITRE. Pix2Net™ [Internet]. Available from: https://www.mitre.org/our-impact/intellectual-property/mitre-pix2net.

6. Bachelot D. Degate: Multi-Platform Software for Semi-Automatic VLSI Reverse Engineering of Digital Logic in Chips [Internet]. Available from: https://www.degate.org

7. TechInsights. Reverse Engineering Solutions [Internet]. Available from: https://www.techinsights.com/capabilities/reverse-engineering

8. Rothaug L, Klix J, et al. Towards Unsupervised SEM Image Segmentation for IC Layout Extraction. Proc. Conf. on Hardware Security; 2023.

9. iST. How to Use SEM Full-Vision Imaging Technology to Reverse-Perceive a Nano-Scale IC Circuit [Internet]. 2018. Available from: https://www.istgroup.com/en/tech_20181120/

10. Nahum T, et al. Automatic Defect Segmentation by Unsupervised Anomaly Learning. Proc. of 29th IEEE ICIP: 2022. p. 306-310.

11. Doudkin A.A. Fuzzy Neural Network for Object Identification in Integrated Circuit Layouts. Systems Analysis and Applied Informatics Science. 2015;4:11-15.

12. Azimi R, Kong Y, et al. SEMU-Net: A Segmentation-Based Corrector for Fabrication Process Variations of Nanophotonics with Microscopic Images. Proc. IEEE/CVF WACV; 2025.

13. Almira J.M., Phelippeau H., Martinez-Sanchez A. Fast Normalized Cross-Correlation for Template Matching with Rotations. J. Applied Math. and Computing. 2024. doi:https://doi.org/10.1007/s12190-024-02157-6.

14. OpenCV Documentation. Template Matching [Internet]. Available from: https://docs.opencv.org/4.x/d4/dc6/tutorial_py_template_matching.html

15. Rajarathnam RS, Lin Y, Jin Y, et al. ReGDS: A Reverse Engineering Framework from GDSII to Gate-Level Netlist. Proc. IEEE HOST; 2020.

16. LayoutEditor Documentation. GDSII: The Standard for IC Layout Data [Internet]. Available from: https://layouteditor.org/layout/file-formats/gdsii

17. MicroNet Solutions. Pix2Net Suite: Reverse Engineering Software [Internet]. Available from: http://micronetsol.net

18. Zhang Y, et al. A Rotation Invariant Template Matching Algorithm Based on Sub-NCC. Math. Biosci. Eng. 2022;19(9):9505-9519.

19. NVIDIA. CUDA Toolkit 8.0 Release Notes [Internet]. 2016. Available from: https://developer.nvidia.com/cuda-toolkit-archive

20. Visual Studio Code: The Open Source AI Code Editor. Computer Program [Internet]. Microsoft Corporation. Available from: https://code.visualstudio.com

21. Belous A.I., Efimenko S.A., Turtsevich A.S. Semiconductor Power Electronics. Moscow: Tekhnosfera [Internet]. 2013 [cited 2026 Mar 13]. Available from: https://e.lanbook.com/book/73530

22. Pareto Charts in Quality Control: A Snapshot. Quality Magazine; 2023.

GOST

Malakhanov A. A., Azarova K. M., Moiseyev K. A. COMPARATIVE ANALYSIS OF SOFTWARE SOLUTIONS FOR AUTOMATED RECOGNITION OF INTEGRATED CIRCUIT TOPOLOGY // Automation and modeling in design and management. 2026. no. 2. pp. 10-20. DOI: https://doi.org/10.30987/2658-6436-2026-2-10-20. (Date of access 10.10.2026).

MLA

Alexey Malakhanov, Azarova K. M., Moiseyev K. A.. "COMPARATIVE ANALYSIS OF SOFTWARE SOLUTIONS FOR AUTOMATED RECOGNITION OF INTEGRATED CIRCUIT TOPOLOGY." Automation and modeling in design and management 2026.2 (2026): 10-20. DOI: https://doi.org/10.30987/2658-6436-2026-2-10-20.

APA

Malakhanov A. A., Azarova K. M., Moiseyev K. A. (2026). COMPARATIVE ANALYSIS OF SOFTWARE SOLUTIONS FOR AUTOMATED RECOGNITION OF INTEGRATED CIRCUIT TOPOLOGY. Automation and modeling in design and management, 2026(2), 10-20. https://doi.org/10.30987/2658-6436-2026-2-10-20

Chicago

Alexey Malakhanov, Azarova K. M., Moiseyev K. A.. "COMPARATIVE ANALYSIS OF SOFTWARE SOLUTIONS FOR AUTOMATED RECOGNITION OF INTEGRATED CIRCUIT TOPOLOGY." Automation and modeling in design and management 2026, no. 2 (2026): 10-20. https://doi.org/10.30987/2658-6436-2026-2-10-20.

WIKI

{{cite journal | author = Malakhanov A. A., Azarova K. M., Moiseyev K. A. | title = COMPARATIVE ANALYSIS OF SOFTWARE SOLUTIONS FOR AUTOMATED RECOGNITION OF INTEGRATED CIRCUIT TOPOLOGY | journal = Automation and modeling in design and management | year = 2026 | volume = 2026 | issue = 2 | pages = 10-20 | doi = 10.30987/2658-6436-2026-2-10-20 }}

Copy

Export: BibTeX EndNote RefMan RusMARC

COMPARATIVE ANALYSIS OF SOFTWARE SOLUTIONS FOR AUTOMATED RECOGNITION OF INTEGRATED CIRCUIT TOPOLOGY

Confirmation