S.A. Nenashev1, T.A. Pisklenov2, A.A. Zalishchuk3, V.A. Nenashev4

1-4 St. Petersburg State University of Aerospace Instrument Engineering (St. Petersburg, Russia)

1 nenashev_sergey178@mail.ru; 2 tim.kirp@mail.ru; 3 sacha1501@yandex.ru, 4 nenashev.va@yandex.ru 

Problem. The problem of recognizing objects on the Earth's surface using hardware and software systems based on single-board computers and neural networks is of particular relevance in the context of the development of autonomous aviation systems based on small aircraft. Currently, there is a need to create compact energy-efficient hardware and software solutions capable of analyzing optical video frames in real time. However, the use of single-board computers on board of small aircraft with the realization of neural networks often faces the problem of limited computing power, requirements to their power consumption and mass-size characteristics for their work in real time. With the increase in the volume and detail of video frames of the Earth's surface, received from onboard sensors, there is an opportunity to improve the quality of object recognition through the use of modern neural networks. However, it is difficult to directly implement neural networks in systems based on single-board small-size computers because of the complexity of their architectures, which require more computational resource than is available in such computational on-board facilities. Therefore, it is necessary to adapt neural networks to the resources of used single-board small-size computers.

Purpose. To analyze single-board computers and neural network models for the realization of aviation hardware-software vision system, controlling agricultural objects of the earth surface on the basis of small aircraft.

Results. The research of efficiency of neural networks realization on the basis of small-sized single-board computer and capable to be based on a small aircraft has been carried out. It was found that the implementation of neural network “YOLOv8n” on the basis
of single-board computer “Raspberry Pi 4 Model B” demonstrates the best efficiency over other neural network architectures when recognizing haystacks, providing the required speed of video frames processing and a high proportion of correctly recognized objects of interest.

Practical relevance. The presented results provide automation of the processes of agricultural land control in real time. The proposed system based on the neural network “YOLOv8n” and single-board computer “Raspberry Pi 4 Model B” allows to quickly recognize haystacks, which is important for planning harvesting operations, controlling the quality of forage and reducing yield losses.

Nenashev S.A., Pisklenov T.A., Zalishchuk A.A., Nenashev V.A. Development and research of a hardware and software system based on a single-board computer using neural network models for the task of recognizing objects of interest on the Earth's surface. Radiotekhnika. 2025. V. 89. № 8. P. 15−27. DOI: https://doi.org/10.18127/j00338486-202508-02 (In Russian)

1. Nenashev V.A., Sentsov A.A. (2022) [Spatially distributed systems of radar and optical monitoring]. St. Petersburg: SUAI, 191 p. (in Russ.).
2. Antohin E.A., Zalishhuk A.A., Nenashev V.A., Nenashev S.A. Sravnitel'nyj analiz nejrosetevyh modelej dlja klassifikacii zon zemnoj poverhnosti po opticheskim izobrazhenijam. Radiotehnika. 2024. T. 88. № 8. pp. 35-44. DOI: 10.18127/j00338486-202408-04 (in
   Russian).
3. Mal'cev N.D. Issledovanie oblasti primenenija Raspberry Pi-mikrokomp'jutera s vysokoj proizvoditel'nost'ju. Sb. trudov HIII Mezhdunar. konf. «Sovremennye informacionnye tehnologii v obrazovanii, nauke i promyshlennosti», HI Mezhdunar. konkursa nauchnyh i nauchno-metodicheskih rabot, V konkursa «Nauchnoe shkol'noe soobshhestvo» (Moskva, 26–27 aprelja 2019 g.). M.: OOO «Izdatel'stvo «Sputnik+». 2019. pp. 60-62 (in Russian).
4. Bryzhinskij K.A., Nikulin V.V. Issledovanie vozmozhnostej mikrokomp'jutera RASPBERRY PI. Materialy nauch. konf. «XLV Ogarjovskie chtenija». V 3-h chastjah (g. Saransk, 08–13 dekabrja 2016 g.). Otv. red. P.V. Senin. Ch. 1. Saransk: Nacional'nyj issledovatel'skij Mordovskij gos. un-t im. N.P. Ogarjova. 2017. pp. 644-648 (in Russian).
5. Sentsov A. A., Nenashev V. A., Ivanov S. A., Turnetskaya E. L. Combination of generated radar images with a digital terrain map in on-board systems for operational monitoring of the earth's surface // Proceedings of MAI. 2021. No. 117. DOI 10.34759/trd-2021-117-08. (in Russian).
6. Assuncao E., Gaspar P.D., Alibabaei K., Simoes M.P., Proença H., Soares V.N.G.J., Caldeira J.M.L.P. Real-Time image detection for edge devices: a peach fruit detection application. Future Internet. 2022. № 14. Р. 323. https://doi.org/10.3390/fi14110323.
7. Alibabaei K., Gaspar P.D., Lima T.M., Campos R.M., Girao I., Monteiro J., Lopes C.M. A Review of the challenges of using deep learning algorithms to support decision-making in agricultural activities. Remote Sens. 2022. № 14. Р. 638.
8. Zalishhuk A.A., Nenashev V.A. Razrabotka sistemy dlja raspoznavanija ljudej po izobrazhenijam zemnoj poverhnosti. Materialy 34-j Mezhdunar. konf. po komp'juternoj grafike i mashinnomu zreniju «GraphiCon 2024» (g. Omsk, 17–19 sentjabrja 2024 g.). Omsk: Omskij gos. tehnich. un-t. 2024. pp. 544-550. DOI: 10.25206/978-5-8149-3873-2-2024-544-550 (in Russian).
9. Jacob B., Kligys S., Chen B., Zhu M., Tang M., Howard A., Adam H., Kalenichenko D. Quantization and training of neural networks for efficient integer-arithmetic-only inference. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Salt Lake City. UT. USA. 18–23 June 2018. pp. 2704–2713.
10. Peleshenko V.A. Kvantovoe mashinnoe obuchenie. Mjagkie izmerenija i vychislenija. 2022. T. 60. № 11-1. pp. 82-107. DOI: 10.36871/2618-9976.2022.11.008 (in Russian).
11. Malik S., Haider M.U., Iqbal O., Taj M. Neural network pruning through constrained reinforcement learning. 2022 26th International Conference on Pattern Recognition (ICPR). Montreal. QC. Canada. 2022. Р. 3027-3033. DOI: 10.1109/ICPR56361.2022.9956050.
12. Bowman R.W., Vodenicharski B., Collins J.T., Stirling Ju. Flat-field and colour correction for the Raspberry Pi Camera module. Journal of Open Hardware. 2020. V. 4. № 1. DOI: 10.5334/joh.20.
13. Nenashev V.A., Voronov R.M., Berezin A.V. i dr. Raspoznavanie nazemnyh ob’ektov v aviacionnyh sistemah potokovoj obrabotki opticheskih kadrov. Informacionno-izmeritel'nye i upravljajushhie sistemy. 2024. T. 22. № 4. pp. 85-90. DOI: 10.18127/j20700814-202404-10 (in Russian).
14. Pavlenko D.A., Snezhko Je.V., Kovalev V.A. Issledovanie jeffektivnosti bjudzhetnogo tenzornogo processora Google Coral dlja reshenija prakticheskih zadach klassifikacii izobrazhenij. Sb. dokladov Mezhdunar. nauch. konf. «Prikladnoj iskusstvennyj intellekt: perspektivy i riski» (Sankt-Peterburg, 17 oktjabrja 2024 g.). SPb: Sankt-Peterburgskij gos. un-t ajerokosmicheskogo priborostroenija. 2024. pp. 82-90 (in Russian).
15. Lebedev M.S., Beleckij P.N. Realizacija iskusstvennyh nejronnyh setej na PLIS s pomoshh'ju otkrytyh instrumentov. Trudy Instituta sistemnogo programmirovanija RAN. 2021. T. 33. № 6. S.175-192. DOI: 10.15514/ISPRAS-2021-33(6)-12 (in Russian).