A noise filtering algorithm for event-based asynchronous change detection image sensors on TrueNorth and its implementation on TrueNorth
Asynchronous event-based sensors, or “silicon retinae,” are a new class of vision sensors inspired by biological vision systems. The output of these sensors often contains a significant number of noise events along with the signal. Filtering these noise events is a common preprocessing step before u...
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2018
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/84744 http://hdl.handle.net/10220/45090 |
| _version_ | 1811686579302825984 |
|---|---|
| author | Padala, Vandana Basu, Arindam Orchard, Garrick |
| author2 | School of Electrical and Electronic Engineering |
| author_facet | School of Electrical and Electronic Engineering Padala, Vandana Basu, Arindam Orchard, Garrick |
| author_sort | Padala, Vandana |
| collection | NTU |
| description | Asynchronous event-based sensors, or “silicon retinae,” are a new class of vision sensors inspired by biological vision systems. The output of these sensors often contains a significant number of noise events along with the signal. Filtering these noise events is a common preprocessing step before using the data for tasks such as tracking and classification. This paper presents a novel spiking neural network-based approach to filtering noise events from data captured by an Asynchronous Time-based Image Sensor on a neuromorphic processor, the IBM TrueNorth Neurosynaptic System. The significant contribution of this work is that it demonstrates our proposed filtering algorithm outperforms the traditional nearest neighbor noise filter in achieving higher signal to noise ratio (~10 dB higher) and retaining the events related to signal (~3X more). In addition, for our envisioned application of object tracking and classification under some parameter settings, it can also generate some of the missing events in the spatial neighborhood of the signal for all classes of moving objects in the data which are unattainable using the nearest neighbor filter. |
| first_indexed | 2024-10-01T05:02:40Z |
| format | Journal Article |
| id | ntu-10356/84744 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T05:02:40Z |
| publishDate | 2018 |
| record_format | dspace |
| spelling | ntu-10356/847442020-03-07T13:57:29Z A noise filtering algorithm for event-based asynchronous change detection image sensors on TrueNorth and its implementation on TrueNorth Padala, Vandana Basu, Arindam Orchard, Garrick School of Electrical and Electronic Engineering TrueNorth Neuromorphic Vision Asynchronous event-based sensors, or “silicon retinae,” are a new class of vision sensors inspired by biological vision systems. The output of these sensors often contains a significant number of noise events along with the signal. Filtering these noise events is a common preprocessing step before using the data for tasks such as tracking and classification. This paper presents a novel spiking neural network-based approach to filtering noise events from data captured by an Asynchronous Time-based Image Sensor on a neuromorphic processor, the IBM TrueNorth Neurosynaptic System. The significant contribution of this work is that it demonstrates our proposed filtering algorithm outperforms the traditional nearest neighbor noise filter in achieving higher signal to noise ratio (~10 dB higher) and retaining the events related to signal (~3X more). In addition, for our envisioned application of object tracking and classification under some parameter settings, it can also generate some of the missing events in the spatial neighborhood of the signal for all classes of moving objects in the data which are unattainable using the nearest neighbor filter. Published version 2018-07-17T02:16:57Z 2019-12-06T15:50:40Z 2018-07-17T02:16:57Z 2019-12-06T15:50:40Z 2018 Journal Article Padala, V., Basu, A., & Orchard, G. (2018). A Noise Filtering Algorithm for Event-Based Asynchronous Change Detection Image Sensors on TrueNorth and Its Implementation on TrueNorth. Frontiers in Neuroscience, 12, 118-. 1662-4548 https://hdl.handle.net/10356/84744 http://hdl.handle.net/10220/45090 10.3389/fnins.2018.00118 en Frontiers in Neuroscience © 2018 Padala, Basu and Orchard. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms 14 p. application/pdf |
| spellingShingle | TrueNorth Neuromorphic Vision Padala, Vandana Basu, Arindam Orchard, Garrick A noise filtering algorithm for event-based asynchronous change detection image sensors on TrueNorth and its implementation on TrueNorth |
| title | A noise filtering algorithm for event-based asynchronous change detection image sensors on TrueNorth and its implementation on TrueNorth |
| title_full | A noise filtering algorithm for event-based asynchronous change detection image sensors on TrueNorth and its implementation on TrueNorth |
| title_fullStr | A noise filtering algorithm for event-based asynchronous change detection image sensors on TrueNorth and its implementation on TrueNorth |
| title_full_unstemmed | A noise filtering algorithm for event-based asynchronous change detection image sensors on TrueNorth and its implementation on TrueNorth |
| title_short | A noise filtering algorithm for event-based asynchronous change detection image sensors on TrueNorth and its implementation on TrueNorth |
| title_sort | noise filtering algorithm for event based asynchronous change detection image sensors on truenorth and its implementation on truenorth |
| topic | TrueNorth Neuromorphic Vision |
| url | https://hdl.handle.net/10356/84744 http://hdl.handle.net/10220/45090 |
| work_keys_str_mv | AT padalavandana anoisefilteringalgorithmforeventbasedasynchronouschangedetectionimagesensorsontruenorthanditsimplementationontruenorth AT basuarindam anoisefilteringalgorithmforeventbasedasynchronouschangedetectionimagesensorsontruenorthanditsimplementationontruenorth AT orchardgarrick anoisefilteringalgorithmforeventbasedasynchronouschangedetectionimagesensorsontruenorthanditsimplementationontruenorth AT padalavandana noisefilteringalgorithmforeventbasedasynchronouschangedetectionimagesensorsontruenorthanditsimplementationontruenorth AT basuarindam noisefilteringalgorithmforeventbasedasynchronouschangedetectionimagesensorsontruenorthanditsimplementationontruenorth AT orchardgarrick noisefilteringalgorithmforeventbasedasynchronouschangedetectionimagesensorsontruenorthanditsimplementationontruenorth |