AEFSNN: Adaptive Filtering Spiking Neural Network for Event-Based Sensors
Source abstract
Abstract
Dynamic Vision Sensor (DVS) is an event-based imaging technology inspired by biological photoreceptors, which holds great promise for edge computing. The event streams produced by DVS are often contaminated by Background Activity (BA) noise and hot-pixel noise, which degrade downstream processing. Existing filters typically use fixed parameters, resulting in poor adaptability to changing illumination. In this paper, we propose a lightweight Adaptive Event-based Filtering Spiking Neural Network (AEFSNN) to address these limitations. Inspired by homeostatic plasticity, AEFSNN dynamically adjusts neuronal thresholds by monitoring the input-to-output spike ratio, allowing the network to autonomously converge to an optimal operating point across different lighting conditions. Furthermore, we introduce a novel neuronal wake-up mechanism that inhibits processing neurons until triggered by valid input, which effectively suppresses redundant events generated by neighboring activity. Experiments show that AEFSNN is more robust under varying illumination. Compared with current filters, our method increases the Signal-to-Noise Ratio (SNR) of the output data by 1.42–2.33 dB. Additionally, the filtered data improves classification accuracy on downstream tasks, validating its practical value for neuromorphic vision systems.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Advanced Memory and Neural Computing · CCD and CMOS Imaging Sensors · Ferroelectric and Negative Capacitance Devices
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Institute of Microelectronics, Chinese Academy of Sciences, Beijing 100029, China
School of Integrated Circuits, University of Chinese Academy of Sciences, Beijing 101408, China
Funding
Beijing Major Science and Technology Project
References (57)
Show source-provided references
1. Czech, D., and Orchard, G. (2016, January 26–29). Evaluating noise filtering for event-based asynchronous change detection image sensors. Proceedings of the IEEE International Conference on Biomedical Robotics and Biomechatronics, Singapore.10.1109/BIOROB.2016.7523452 ↗
2. Event-based vision: A survey10.1109/TPAMI.2020.3008413 ↗
3. High speed and high dynamic range video with an event camera10.1109/TPAMI.2019.2963386 ↗
4. Overview of the SpiNNaker system architecture10.1109/TC.2012.142 ↗
5. Parameter optimization and learning in a spiking neural network for UAV obstacle avoidance targeting neuromorphic processors10.1109/TNNLS.2019.2941506 ↗
6. Salvatore, N., Mian, S., Abidi, C., and George, A.D. (2020, January 11–15). A neuro-inspired approach to intelligent collision avoidance and navigation. Proceedings of the AIAA/IEEE 39th Digital Avionics Systems Conference (DASC), San Antonio, TX, USA.10.1109/DASC50938.2020.9256492 ↗
7. Lele, A.S., Fang, Y., Anwar, A., and Raychowdhury, A. (2022). Bio-mimetic high-speed target localization with fused frame and event vision for edge application. Front. Neurosci., 16.10.3389/fnins.2022.1010302 ↗
8. Kang, D., Lee, Y.K., and Jeong, J. (2023). Exploring the potential of event camera imaging for advancing remote pupil-tracking techniques. Appl. Sci., 13.10.3390/app131810357 ↗
9. Zhao, J., Su, L., Wang, X., Li, J., Yang, F., Jiang, N., and Hu, Q. (2023). DTFS-eHarris: A high accuracy asynchronous corner detector for event cameras in complex scenes. Appl. Sci., 13.10.3390/app13095761 ↗
10. Truenorth: Design and tool flow of a 65 mw 1 million neuron programmable neurosynaptic chip10.1109/TCAD.2015.2474396 ↗
11. Loihi: A neuromorphic manycore processor with on-chip learning10.1109/MM.2018.112130359 ↗
12. Fang, W., Yu, Z., Chen, Y., Masquelier, T., Huang, T., and Tian, Y. (2021, January 11–17). Incorporating learnable membrane time constant to enhance learning of spiking neural networks. Proceedings of the IEEE/CVF International Conference on Computer Vision, Montreal, QC, Canada.10.1109/ICCV48922.2021.00266 ↗
13. Cheng, X., Cao, S., Wang, S., Zeng, X., Li, W., and Wang, M. (2024, January 19–22). A 1024-neuron 1M-synapse event-driven SNN accelerator for DVS applications. Proceedings of the 2024 IEEE International Symposium on Circuits and Systems (ISCAS), Singapore.10.1109/ISCAS58744.2024.10558416 ↗
14. Massa, R., Marchisio, A., Martina, M., and Shafique, M. (2020, January 19–24). An efficient spiking neural network for recognizing gestures with a dvs camera on the loihi neuromorphic processor. Proceedings of the 2020 International Joint Conference on Neural Networks (IJCNN), Glasgow, UK.10.1109/IJCNN48605.2020.9207109 ↗
15. Temperature and parasitic photocurrent effects in dynamic vision sensors10.1109/TED.2017.2717848 ↗
16. A 128 × 128 120 dB 15 μs latency asynchronous temporal contrast vision sensor10.1109/JSSC.2007.914337 ↗
17. Delbruck, T. (2008, January 6–7). Frame-Free Dynamic Digital Vision. Proceedings of the International Symposium on Secure-Life Electronics Advanced Electronics for Quality Life and Society, Tokyo, Japan.
18. HashHeat: A hashing-based spatiotemporal filter for dynamic vision sensor10.1016/j.vlsi.2021.04.006 ↗
19. Hast: A hardware-efficient spatio-temporal correlation near-sensor noise filter for dynamic vision sensors10.1109/TCSI.2024.3517133 ↗
20. Chen, Y., Huang, Y., Li, F., Zeng, X., Li, W., and Wang, M. (2023, January 21–25). Denoising method for dynamic vision sensor based on two-dimensional event density. Proceedings of the 2023 IEEE International Symposium on Circuits and Systems (ISCAS), Monterey, CA, USA.10.1109/ISCAS46773.2023.10181865 ↗
21. Baldwin, R., Almatrafi, M., Asari, V., and Hirakawa, K. (2020, January 13–19). Event probability mask (EPM) and event denoising convolutional neural network (EDNCNN) for neuromorphic cameras. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.10.1109/CVPR42600.2020.00177 ↗
22. Zhang, Z., Suo, J., and Dai, Q. (2021, January 10–12). Denoising of event-based sensors with deep neural networks. Proceedings of the Optoelectronic Imaging and Multimedia Technology VIII, Nantong, China.10.1117/12.2602742 ↗
23. Rios-Navarro, A., Guo, S., Gnaneswaran, A., Vijayakumar, K., Linares-Barranco, A., Aarrestad, T., and Delbruck, T. (2023, January 18–22). Within-camera multilayer perceptron DVS denoising. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Vancouver, BC, Canada.10.1109/CVPRW59228.2023.00409 ↗
24. Neuromorphic Camera Denoising Using Graph Neural Network-Driven Transformers10.1109/TNNLS.2022.3201830 ↗
25. Barrios-Avilés, J., Rosado-Muñoz, A., Medus, L.D., Bataller-Mompeán, M., and Guerrero-Martínez, J.F. (2018). Less data same information for event-based sensors: A bioinspired filtering and data reduction algorithm. Sensors, 18.10.3390/s18124122 ↗
26. Padala, V., Basu, A., and Orchard, G. (2018). A noise filtering algorithm for event-based asynchronous change detection image sensors on truenorth and its implementation on truenorth. Front. Neurosci., 12.10.3389/fnins.2018.00118 ↗
27. Xiao, K., Cui, X., Liu, K., Cui, X., and Wang, X.A. (2021, January 18–22). An snn-based and neuromorphic-hardware-implementable noise filter with self-adaptive time window for event-based vision sensor. Proceedings of the 2021 International Joint Conference on Neural Networks (IJCNN), Shenzhen, China.10.1109/IJCNN52387.2021.9534073 ↗
28. Kowalczyk, M., and Kryjak, T. (September, January 31). Hardware architecture for high throughput event visual data filtering with matrix of IIR filters algorithm. Proceedings of the 2022 25th Euromicro Conference on Digital System Design (DSD), Maspalomas, Spain.10.36227/techrxiv.20218304 ↗
29. Yang, J., and Song, T. (2020, January 21–24). A prediction scheme in spiking neural network (SNN) hardware for ultra-low power consumption. Proceedings of the 2020 International SoC Design Conference (ISOCC), Yeosu, Republic of Korea.10.1109/ISOCC50952.2020.9333106 ↗
30. Homeostatic plasticity in neuronal networks: The more things change, the more they stay the same10.1016/S0166-2236(98)01341-1 ↗
31. Integrating Hebbian and homeostatic plasticity: Introduction10.1098/rstb.2016.0413 ↗
32. Smooth pulse recovery based on hybrid wavelet threshold denoising and first derivative adaptive smoothing filter10.1017/hpl.2021.2 ↗
33. Amir, A., Taba, B., Berg, D., Melano, T., McKinstry, J., Di Nolfo, C., and Modha, D. (2017, January 21–26). A low power, fully event-based gesture recognition system. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA.10.1109/CVPR.2017.781 ↗
34. Feng, Y., Lv, H., Liu, H., Zhang, Y., Xiao, Y., and Han, C. (2020). Event density based denoising method for dynamic vision sensor. Appl. Sci., 10.10.3390/app10062024 ↗
35. Siddique, M.A.B., Arif, R.B., and Khan, M.M.R. (2018, January 27–28). Digital image segmentation in matlab: A brief study on otsu’s image thresholding. Proceedings of the 2018 International Conference on Innovation in Engineering and Technology (ICIET), Dhaka, Bangladesh.10.1109/CIET.2018.8660942 ↗
36. Low cost and latency event camera background activity denoising10.1109/TPAMI.2022.3152999 ↗
37. Li, H., Liu, H., Ji, X., Li, G., and Shi, L. (2017). Cifar10-dvs: An event-stream dataset for object classification. Front. Neurosci., 11.10.3389/fnins.2017.00309 ↗
38. Energy-efficient task offloading using dynamic voltage scaling in mobile edge computing10.1109/TNSE.2020.3046014 ↗
39. Ma, Z., Wei, Z., Zhang, W., Lyu, Z., Xu, J., and Zhang, B. (2020, January 17–19). Energy-saving strategy for edge computing by collaborative processing tasks on base stations. Proceedings of the 2020 16th international conference on mobility, sensing and networking (MSN), Tokyo, Japan.10.1109/MSN50589.2020.00044 ↗
40. Lin, Y., and Zhu, C. (September, January 30). Detection of Pedestrian Movement Poses in High-Speed Autonomous Driving Environments Using DVS. Proceedings of the International Conference on Pattern Analysis and Machine Intelligence, Shanghai, China.
41. Wang, W., Zhou, S., Li, J., Li, X., Yuan, J., and Jin, Z. (2021, January 10–15). Temporal pulses driven spiking neural network for time and power efficient object recognition in autonomous driving. Proceedings of the 2020 25th International Conference on Pattern Recognition (ICPR), Milan, Italy.10.1109/ICPR48806.2021.9412302 ↗
42. Wang, S., Yu, S., Liang, T., Shi, Y., Ma, Y., and Ren, P. (2025, January 19–23). Rapid Dynamic Obstacle Avoidance for UAVs Enhanced by DVS and Neuromorphic Computing. Proceedings of the 2025 IEEE International Conference on Robotics and Automation (ICRA), Atlanta, GA, USA.10.1109/ICRA55743.2025.11128605 ↗
43. Bian, S., Schulthess, L., Rutishauser, G., Di Mauro, A., Benini, L., and Magno, M. (2023, January 8–9). Colibriuav: An ultra-fast, energy-efficient neuromorphic edge processing uav-platform with event-based and frame-based cameras. Proceedings of the 2023 9th International Workshop on Advances in Sensors and Interfaces (IWASI), Monopoli, Italy.10.1109/IWASI58316.2023.10164354 ↗
44. Stroobants, S., De Wagter, C., and De Croon, G.C.H.E. (2023, January 1–3). Neuromorphic Control using Input-Weighted Threshold Adaptation. Proceedings of the 2023 International Conference on Neuromorphic Systems, Santa Fe, NM, USA.10.1145/3589737.3605963 ↗
45. Singh, A.P., and Leite, A.C. (2025). Neuromorphic Control of Robotic Systems with Numerical Simulations. Robotics, 14.10.3390/robotics14110166 ↗
46. Stagsted, R.K., Vitale, A., Binz, J., Renner, A., Larsen, L.B., and Sandamirskaya, Y. (2020, January 12–16). Towards neuromorphic control: A spiking neural network based PID controller for UAV. Proceedings of the Robotics: Science and Systems (RSS), Corvalis, OR, USA.10.15607/RSS.2020.XVI.074 ↗
47. Marrero, D., Kern, J., and Urrea, C. (2024). A Novel Robotic Controller Using Neural Engineering Framework-Based Spiking Neural Networks. Sensors, 24.10.3390/s24020491 ↗
48. Shrestha, S.B., Timcheck, J., Frady, P., Campos-Macias, L., and Davies, M. (2024, January 14–19). Efficient Video and Audio Processing with Loihi 2. Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Seoul, Republic of Korea.10.1109/ICASSP48485.2024.10448003 ↗
49. Advancing Neuromorphic Computing With Loihi: A Survey of Results and Outlook10.1109/JPROC.2021.3067593 ↗
50. Kryjak, T. (2024, January 28–30). Event-based vision on FPGAs—A survey. Proceedings of the 2024 27th Euromicro Conference on Digital System Design (DSD), Paris, France.10.1109/DSD64264.2024.00078 ↗
51. Hardware implementation of FPGA-based spiking attention neural network accelerator10.7717/peerj-cs.3077 ↗
52. An FPGA-Based Event-Driven SNN Accelerator for DVS Applications With Structured Sparsity and Early-Stop10.1109/TCSI.2025.3560666 ↗
53. Bartolomei, L., Poggi, M., Conti, A., and Mattoccia, S. (October, January 29). LiDAR-Event Stereo Fusion with Hallucinations. Proceedings of the European Conference on Computer Vision (ECCV), Milan, Italy.
54. Sakhai, M., Sithu, K., Oke, M.K.S., and Wielgosz, M. (2025). Cyberattack Resilience of Autonomous Vehicle Sensor Systems: Evaluating RGB vs. Dynamic Vision Sensors in CARLA. Appl. Sci., 15.10.20944/preprints202505.2262.v1 ↗
55. Cheng, H., Liu, C., Gu, W., Wu, Y., Zhao, M., Liu, W., and Wang, N. (2025). LGMMFusion: A LiDAR-guided multi-modal fusion framework for enhanced 3D object detection. PLoS ONE, 20.10.1371/journal.pone.0331195 ↗
56. Wang, M., Li, D., Casas, J.R., and Ruiz-Hidalgo, J. (2025). Adaptive Fusion of LiDAR Features for 3D Object Detection in Autonomous Driving. Sensors, 25.10.3390/s25133865 ↗
57. Liu, A., Zhang, Y., Shi, H., and Chen, J. (2025). Robust 3D Object Detection in Complex Traffic via Unified Feature Alignment in Bird’s Eye View. World Electr. Veh. J., 16.10.3390/wevj16100567 ↗