Comparison of the Efficiency of a Neural Network for Image Recognition on Microcontrollers
The paper is devoted to comparing two popular models of 32-bit microcontrollers for working with neural networks for object recognition. The target devices were the ESP32 and STM32 microcontrollers, on which an artificial neural network was deployed, written using the Python programming language and...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Igor Sikorsky Kyiv Polytechnic Institute
2024-07-01
|
| Series: | Mìkrosistemi, Elektronìka ta Akustika |
| Subjects: | |
| Online Access: | https://elc.kpi.ua/article/view/300851 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846173669295915008 |
|---|---|
| author | Rostyslav Dmytovych Sharuiev Pavlo Vasyliovych Popovych |
| author_facet | Rostyslav Dmytovych Sharuiev Pavlo Vasyliovych Popovych |
| author_sort | Rostyslav Dmytovych Sharuiev |
| collection | DOAJ |
| description | The paper is devoted to comparing two popular models of 32-bit microcontrollers for working with neural networks for object recognition. The target devices were the ESP32 and STM32 microcontrollers, on which an artificial neural network was deployed, written using the Python programming language and the TensorFlow library. Micropython was chosen as the operating system for the microcontrollers. The paper compares the performance of the ESP32 and STM32 microcontrollers for object detection using a neural network and their classification. The image recognition time and the percentage of correctly classified objects were compared depending on the number of neuron layers and the number of training epochs within these networks. The article shows that the number of layers and training epochs directly affects the accuracy of object classification in the image. The obtained results show that increasing the number of layers of the neural network increases the overall accuracy of object recognition using the studied neural network, increasing the number of training epochs logarithmically increases the accuracy of recognition and classification within the neural network, but at the same time, increasing the number of neuron layers leads to an increase in the total recognition time. The difference in the obtained results for the accuracy of image recognition of microcontrollers differs within 5%.
|
| format | Article |
| id | doaj-art-6afa36e1d21840ea9f615b78e035395d |
| institution | Kabale University |
| issn | 2523-4447 2523-4455 |
| language | English |
| publishDate | 2024-07-01 |
| publisher | Igor Sikorsky Kyiv Polytechnic Institute |
| record_format | Article |
| series | Mìkrosistemi, Elektronìka ta Akustika |
| spelling | doaj-art-6afa36e1d21840ea9f615b78e035395d2024-11-08T13:16:18ZengIgor Sikorsky Kyiv Polytechnic InstituteMìkrosistemi, Elektronìka ta Akustika2523-44472523-44552024-07-0129210.20535/2523-4455.mea.300851Comparison of the Efficiency of a Neural Network for Image Recognition on MicrocontrollersRostyslav Dmytovych Sharuiev0Pavlo Vasyliovych Popovych1National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute”National Technical University of Ukraine "Ihor Sikorsky Kyiv Polytechnic Institute"The paper is devoted to comparing two popular models of 32-bit microcontrollers for working with neural networks for object recognition. The target devices were the ESP32 and STM32 microcontrollers, on which an artificial neural network was deployed, written using the Python programming language and the TensorFlow library. Micropython was chosen as the operating system for the microcontrollers. The paper compares the performance of the ESP32 and STM32 microcontrollers for object detection using a neural network and their classification. The image recognition time and the percentage of correctly classified objects were compared depending on the number of neuron layers and the number of training epochs within these networks. The article shows that the number of layers and training epochs directly affects the accuracy of object classification in the image. The obtained results show that increasing the number of layers of the neural network increases the overall accuracy of object recognition using the studied neural network, increasing the number of training epochs logarithmically increases the accuracy of recognition and classification within the neural network, but at the same time, increasing the number of neuron layers leads to an increase in the total recognition time. The difference in the obtained results for the accuracy of image recognition of microcontrollers differs within 5%. https://elc.kpi.ua/article/view/300851microcontrollerneural networkepochtrainingclassification |
| spellingShingle | Rostyslav Dmytovych Sharuiev Pavlo Vasyliovych Popovych Comparison of the Efficiency of a Neural Network for Image Recognition on Microcontrollers Mìkrosistemi, Elektronìka ta Akustika microcontroller neural network epoch training classification |
| title | Comparison of the Efficiency of a Neural Network for Image Recognition on Microcontrollers |
| title_full | Comparison of the Efficiency of a Neural Network for Image Recognition on Microcontrollers |
| title_fullStr | Comparison of the Efficiency of a Neural Network for Image Recognition on Microcontrollers |
| title_full_unstemmed | Comparison of the Efficiency of a Neural Network for Image Recognition on Microcontrollers |
| title_short | Comparison of the Efficiency of a Neural Network for Image Recognition on Microcontrollers |
| title_sort | comparison of the efficiency of a neural network for image recognition on microcontrollers |
| topic | microcontroller neural network epoch training classification |
| url | https://elc.kpi.ua/article/view/300851 |
| work_keys_str_mv | AT rostyslavdmytovychsharuiev comparisonoftheefficiencyofaneuralnetworkforimagerecognitiononmicrocontrollers AT pavlovasyliovychpopovych comparisonoftheefficiencyofaneuralnetworkforimagerecognitiononmicrocontrollers |