Estimation of dislocation density on GaAs plates using machine vision

The paper presents the results of the application of artificial intelligence (AI) and machine vision technologies for the detection of surface defects. Plates with a crystallographic surface of {100} monocrystalline GaAs grown by the Czochralski method with liquid encapsulation of the melt were analyzed. Based on the YOLOv8 open architecture, a neural network was trained to recognize pits formed as a result of selective etching of single-crystal GaAs plates, and a solution was proposed to automate the calculation of dislocation density based on the identified pits of selective etching. Monochrome images were used for processing by the neural network, the data array at the training stage amounted to about 40,000 objects. It was found that the average density of etching pits (detection objects) is (3 – 7) × 10⁴ cm^–2. When trained on a sufficient amount of data, AI and machine vision algorithms are able to recognize target objects with high confidence, including overlapping ones. Continuous counting (over the entire surface of the plate) and further software processing of the results made it possible to obtain a map of the density distribution of dislocation etching pits, as well as lines of density levels (isolines) with reference to its absolute value. It is shown that the use of the continuous counting method using AI and machine vision technologies in comparison with traditional averaging methods for analyzing the structural uniformity of single-crystal GaAs is justified and appropriate. The obtained results can be used for technological control of dislocation density and identification of patterns in the change of the dislocation structure of single-crystal GaAs depending on the growth modes and post-processing of ingots.

1. Milvidsky M. G., Osvensky V. B. Structural defects in single crystals of semiconductors. — Moscow: Metallurgiya, 1984. — 256 p. [in Russian].

2. Malygin G. A. Dislocations as linear topological defects / Fiz. Tv. Tela. 2001. Vol. 43. No. 5. P. 822 – 826 [in Russian].

3. Khludkov S. S., Tolbanov O. P., Vilisova M. D. Prudayev I. A. Semiconductor devices based on gallium arsenide with deep impurity centers. — Tomsk: TGU, 2016. — 258 p. [in Russian].

4. Knyazev S. N., Kudrya A. V., Komarovsky N. Yu., et al. Methods of studying the dislocation structure of semiconductor single crystals of the AIIIBV group / Mater. Élektron. Teknn. 2022. Vol. 25. No. 4. P. 323 – 336 [in Russian]. DOI: 10.17073/1609-3577-2022-4-323-336

5. Verbitsky R. A., Polzikova K. S., Syrov Yu. V. Features of detecting dislocation outputs on the surface of GaAs plates / The 3rd International Conf. on Technologies, Properties and Applications of Rare Metals and Related Materials (RedMet-2024): coll. of works. — Moscow: Giredmet, 2024 [in Russian].

6. Levchenko D. S., Teplova T. B., Yugova T. G. Investigation of the dislocation structure of gallium arsenide single crystals used to create high-speed microelectronics devices / II International Scientific and Practical Conf. «Economics and practical management in Russia and abroad»: coll. of works. — Kolomna: Kolomensky institut (filial) MAMI, 2015. P. 135 – 137 [in Russian].

7. Oksana A. P., Kogdas M. G., Androsiuk M. S. Methods for improving the structural perfection of semi-insulating GaAs with a diameter of 100 mm / Radioélektr. Inform. 2014. No. 14. P. 15 – 19 [in Russian].

8. Kulchitsky N. A., Naumov A. V., Startsev V. V. Photonics — the new GaAs driver / Fotonika. 2020. Vol. 14. No. 2. P. 138 – 149 [in Russian]. DOI: 10.22184/1993-7296.FRos.2020.14.2.138.149

9. Orlov A. I. Artificial intelligence, neural networks, big data, and mathematical research methods / Industr. Lab. Mater. Diagn. 2023. Vol. 89. No. 7. P. 5 – 7 [in Russian]. DOI: 10.26896/1028-6861-2023-89-7-5-7

10. Bagaev I. I. Analysis of the concepts of neural network and convolutional neural network, training of convolutional neural network using TensorFlow module / Matem. Progr. Obesp. Sist. Prom. Sots. Sferakh. 2020. Vol. 8. No. 1. P. 15 – 22 [in Russian]. DOI: 10.18503/2306-2053-2020-8-1-15-22

11. Ponomarev A. N., Kazantseva L. V. Machine vision: relevance and application / Nauch.-Tekhn. Vestn. Povolzh. 2021. No. 4. P. 100 – 102 [in Russian].

12. Mikhalev O. N., Yanushkin A. S. Machine vision and object recognition using neural networks / Robototekhn. Tekhn. Kibern. 2022. Vol. 10. No. 2. P. 113 – 120 [in Russian]. DOI: 10.31776/rtcj.10204

13. Vlasov S. O., Gladyshev A. I., Boguslavsky A. A., Sokolov S. M. Solving the problem of object detection using neural network technologies / Preprinty IPM im. M. V. Keldysha. 2023. No. 16. P. 1 – 27 [in Russian]. DOI: 10.20948/prepr-2023-16

14. Martinez-García S. E., Fernándezy-Fernández C. A., Ramos-Pérez E. G. Deep Learning for Non-functional Requirements: A Convolutional Neural Network Approach / Proceedings of the Institute for System Programming of the RAS. 2024. Vol. 36. No. 1. P. 131 – 142. DOI: 10.15514/ispras-2024-36(1)-8

15. Agareva O. Yu., Skripova A. E. Convolutional neural network learning algorithm for graphic object recognition based on neural network with a hierarchical classifier / Collection of scientific papers of the Department of Applied Mathematics and Programming based on the results of the ongoing seminar «Theory of Systems». — Moscow: RGU im. A. N. Kosygina, 2022. P. 38 – 47 [in Russian].

16. Verbitsky R. A., Syrov Yu. V., Knyazev S. N., Latonov V. D. Application of machine vision for detection and calculation of dislocation outputs on the surface of gallium arsenide plates / Proceedings of the National Scientific and Practical Conf. «Fundamental, exploratory, applied research and innovative projects». — Moscow: Association of Graduates and Employees of the Professor N. E. Zhukovsky VVIA, 2023. P. 139 – 143 [in Russian].

17. Verbitsky R. A., Latonov V. D., Syrov Yu. V. Technical implementation of the automated «Kolibri» dislocation counting system / The 3rd International Conf. on Technologies, Properties and Applications of Rare Metals and Related Materials (RedMet-2024): coll. of works. — Moscow: Giredmet, 2024 [in Russian].

18. Yaryshev S. N., Ryzhova V. A. Deep learning and neural network technologies in video analysis tasks. — St. Petersburg: ITMO, 2022. — 82 p. [in Russian].

19. Wang M., Yang W., Wang L., et al. Fe-YOLOv5: Feature Enhancement Network Based on YOLOv5 for Small Object Detection / Social Science Research Network. 2022. DOI: 10.2139/ssrn.4170610

20. Kutyrev A. I., Smirnov I. G., Andrianov N. A. Comparative analysis of neural network models for recognizing apple fruits on the crown of a tree / Sadovod. Vinogradar. 2023. No. 5. P. 56 – 63 [in Russian]. DOI: 10.31676/0235-2591-2023-5-56-63

21. Tikhanovsky G. S., Kochkin V. D. Investigation of the possibilities of using deep learning algorithms and computer vision in the task of classifying mushrooms and berries / Vestn. VGU. 2023. No. 4(22). P. 41 – 46 [in Russian].