Abstract
Machine Learning has unwaveringly found its way into many modern-day applications and it only seems to be spreading widely soon. Most of the applications make use of machine learning algorithms for classification, object detection purposes, and are dependent on GPU’s to perform complex computational tasks. This posed a serious limitation for the development of such applications for mobile platforms due to the lower processing capabilities of mobile chips and storage issues. For a long time in the past, developers were reliant on a cloud-based approach for machine learning capabilities to their applications with the help of REST API services. This is where “On-device Machine Learning” comes into the picture, using a mobile computing platform that utilizes the internal CPU and GPU, present in every mobile device. However, due to certain limitations in the computing power of mobile processing chips, it is very easy to throttle this solution. In this paper, we present a multi-threaded approach to sequentially classify a large number of images on the mobile CPU. This approach aids the On-Device Machine Learning approach by providing multiple threads for the application to process and handle data without throttling the hardware.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Mather T, Kumaraswamy S, Latif S (2009) Cloud security and privacy: an enterprise perspective on risks and compliance. Published by O’Reilly Media, Inc.
Li L-H, Chu Y-S, Chu J-Y, Guo S-H (2019) A machine learning approach for detection plant disease: taking orchid as example. In: Proceedings of the 3rd international conference on vision, image and signal processing. Association for computing machinery, New York, NY, USA, Article 43, pp 1–6. https://doi.org/10.1145/3387168.3387238
Example on-device model personalization with TensorFlow Lite. https://blog.tensorflow.org/2019/12/example-on-device-model-personalization.html. Accessed December 2020
Processes and threads overview. https://developer.android.com/guide/components/processes-and-threads. Accessed December 2020
Interpreter TensorFlow Lite. https://www.tensorflow.org/lite/api_docs/java/org/tensorflow/lite/Interpreter. Accessed December 2020
Better performance through threading Android Developers. https://developer.android.com/topic/performance/threads. Accessed 06 December 2020
Keeping your app responsive. https://developer.android.com/training/articles/perf-anr. Accessed December 2020
Producer Consumer Design Pattern with Blocking Queue Example in Java (nd). https://javarevisited.blogspot.com/2012/02/producer-consumer-design-pattern-with.html. Accessed 04 December 2020
Interface BlockingQueue. https://docs.oracle.com/en/java/javase/11/docs/api/java.base/java/util/concurrent /BlockingQueue.html. Accessed December 2020
Content provider basics Android Developers . https://developer.android.com/guide/topics/providers/content-provider-basics. Accessed December 2020
Cursor Android Developers. https://developer.android.com/reference/android/database/Cursor. Accessed December 2020
TensorFlow Lite guide. https://www.tensorflow.org/lite/guide. Accessed December 2020
TensorFlow Lite inference. https://www.tensorflow.org/lite/guide/inference. Accessed December 2020
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Kothari, S., Crasta, R., Biju, A., Lotlikar, T., Rai, H. (2022). On-Device ML: An Efficient Approach to Classify Large Number of Images Using Multi-threading in Android Devices. In: Mathur, G., Bundele, M., Lalwani, M., Paprzycki, M. (eds) Proceedings of 2nd International Conference on Artificial Intelligence: Advances and Applications. Algorithms for Intelligent Systems. Springer, Singapore. https://doi.org/10.1007/978-981-16-6332-1_64
Download citation
DOI: https://doi.org/10.1007/978-981-16-6332-1_64
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-16-6331-4
Online ISBN: 978-981-16-6332-1
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)