Uniform Sampling Method with Optimized VLSI Circuit for Data Augmentation in Pixel Detector
Keywords:
Uniform Sampling Method, VLSI Circuit; Data Augmentation, Pixel Detector, Fish Swarm Optimization.Abstract
The Uniform Sampling Method combined with an optimized VLSI circuit offers an efficient approach to data augmentation in pixel detectors. This method ensures uniform coverage of pixel data by systematically selecting representative samples, reducing redundancy and improving the quality of augmented datasets. The optimized VLSI circuit enhances processing speed and energy efficiency, enabling real-time augmentation of high-resolution detector data. The proposed Uniform Sampling Fish Swarm Optimization for Data Circuit (SFWO-DC) introduces an innovative approach to data augmentation in pixel detectors by combining uniform sampling with the intelligent optimization capabilities of fish swarm algorithms. The proposed Uniform Sampling Fish Swarm Optimization for Data Circuit (SFWO-DC) introduces a transformative approach to data augmentation in pixel detectors by synergizing uniform sampling techniques with the optimization prowess of fish swarm algorithms. The fish swarm optimization dynamically mimics the intelligent foraging behavior of fish, enabling it to explore and exploit the pixel data space effectively. This ensures that the selected pixel samples are not only uniformly distributed but also represent the most diverse and informative regions of the detector data, addressing redundancy while maintaining high data quality. SFWO-DC ensures balanced pixel selection while dynamically adapting to data patterns, optimizing both the coverage and diversity of augmented datasets. Integrated into a VLSI circuit, this method enhances processing efficiency, reducing latency and energy consumption. For instance, in a 1024 × 1024 pixel detector, SFWO-DC achieves a 35% reduction in computational overhead compared to traditional methods, with an augmentation accuracy improvement of up to 20%. This approach is ideal for real-time imaging systems and machine learning-driven pixel analysis.
References
[1] A. Haldorai, R.B. Lincy, M. Suriya, M. Balakrishnan and P. Mithun, “The Performance Enhancement of Image Segmentation Technique based on VLSI Technology,” In 2024 5th International Conference on Innovative Trends in Information Technology (ICITIIT), Kottayam, India, pp. 1-6, 2024.
[2] A. Deeb, A. Ibrahim, M. Salem, J. Pichler, S. Tkachov, A. Karaj et al., “A robust automated analog circuits classification involving a graph neural network and a novel data augmentation strategy,” Sensors, vol.23, no.6, pp.2989, 2023.
[3] X. Cui, H. Zhang, J. Xu, X. Fang, W. Ning et al., “A Data Augmentation Method for Side-Channel Attacks on Cryptographic Integrated Circuits,” Electronics, vol.13, no.7, pp.1348, 2024.
[4] C. Xu, W. Liu, Y. Zheng, S. Wang and C.H.Chang, “An Imperceptible Data Augmentation Based Blackbox Clean-Label Backdoor Attack on Deep Neural Networks,” IEEE Transactions on Circuits and Systems I: Regular Papers, 2024.
[5] Y. Feng, T. Ma, Y. Zhu and X. Zhang, “BlissCam: Boosting Eye Tracking Efficiency with Learned In-Sensor Sparse Sampling,” arXiv preprint arXiv:2404.15733.
[6] K. Miyauchi, M. Uno, T. Isozaki, H. Fukuhara, K. Mori et al., “A 3.96-$mu $ m, 124-dB DynamicRange, Digital-Pixel Sensor With Triple-and Single-Quantization Operations for Monochrome and Near-Infrared Dual-Channel Global Shutter Operation,” IEEE Journal of Solid-State Circuits, 2024.
[7] Y. Y. Bao, E. Li, H.Q.Yang and B.B. Jia, “Wafer Map Defect Classification Using AutoencoderBased Data Augmentation and Convolutional Neural Network,” arXiv preprint arXiv:2411.11029,2024.
[8] H.J. Kim, “Design of A Prototype 128× 128 ROIC Array for 2.6 μm-wavelength SWIR Image Sensor Applications,” Integration, pp.102232, 2024.
[9] M. Wang, H. Qiu, T. Zhang, M. Qiu and B. Thuraisingham, “Mitigating Query-based Neural Network Fingerprinting via Data Augmentation,” ACM Transactions on Sensor Networks, 2023.
[10] J. A. Chen, W. Niu, B. Ren, Y. Wang and X. Shen, “Survey: Exploiting data redundancy for optimization of deep learning,” ACM Computing Surveys, vol.55, no.10, pp.1-38, 2023.
[11] B. Prasetiyo, A. Alamsyah, M. Hakim, J. Jumanto and M.H. Adi, “Differential augmentation data for vehicle classification using convolutional neural network,” In AIP Conference Proceedings, vol. 2614, no. 1, 2023.
[12] Y. Liang, H. Huang, Y. He, C. Hu, S. Yang and H.Chen, “Pixel-wise exposure control for singleshot HDR imaging: A joint optimization approach,” APL Photonics, vol.9, no.6, 2024.
[13] Q. Chen, K. Nie, J. Gao, X. Zhang and J. Xu, “A 1920$times $1080 Array 2-D/3-D Image Sensor With 3-$mu $ s Row-Time Single-Slope ADC and 100-MHz Demodulated PPD Locked-In Pixel,” IEEE Journal of Solid-State Circuits, 2024.
[14] G. B. Rozendo, B. L. D. O. Garcia, V. A. T. Borgue, A. Lumini, T. A. A. Tosta et al., “Data Augmentation in Histopathological Classification: An Analysis Exploring GANs with XAI and Vision Transformers,” Applied Sciences, vol.14, no.18, pp.8125, 2024.
[15] J. Qiao, C. Sun, X. Cheng, J. Yang and N. Chen, “Stainless steel cylindrical pot outer surface defect detection method based on cascade neural network,” Measurement Science and Technology, vol.35, no.3, pp.036201, 2023.
[16] A. Sridhar, C.C. Chang, J. Zhang and Y. Chen, “Improving Routability Prediction via NAS Using a Smooth One-shot Augmented Predictor,” arXiv preprint arXiv:2411.14296, 2024.
[17] S. Kasarapu, S. Shukla and S.M.P. Dinakarrao, “Resource-and workload-aware model parallelisminspired novel malware detection for iot devices,” IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, vol.42, no.12, pp.4618-4628, 2023.
[18] J. Song, A. Swope, R. Kirby, R. Roy, S. Godil et al., “CircuitVAE: Efficient and Scalable Latent Circuit Optimization,” In Proceedings of the 61st ACM/IEEE Design Automation Conference, pp. 1-6, 2024.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Journal of Electronics and Power Engineering (JEPE)
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Fringe Global Scientific Press publishes all the papers under a Creative Commons Attribution-Non-Commercial 4.0 International (CC BY-NC 4.0) (https://creativecommons.org/licenses/by-nc/4.0/) license. Authors have the liberty to replicate and distribute their work. Authors have the ability to use either the whole or a portion of their piece in compilations or other publications that include their own work. Please see the licensing terms for more information on reusing the work.
