An Event-Driven Ultra-Low-Power Smart Visual Sensor

@article{Rusci2016AnEU,
  title={An Event-Driven Ultra-Low-Power Smart Visual Sensor},
  author={Manuele Rusci and Davide Rossi and Michela Lecca and Massimo Gottardi and Elisabetta Farella and Luca Benini},
  journal={IEEE Sensors Journal},
  year={2016},
  volume={16},
  pages={5344-5353}
}
In this paper, we present an ultra-low-power smart visual sensor architecture. A 10.6-μW low-resolution contrast-based imager featuring internal analog preprocessing is coupled with an energy-efficient quad-core cluster processor that exploits near-threshold computing within a few milliwatt power envelope. We demonstrate the capability of the smart camera on a moving object detection framework. The computational load is distributed among mixed-signal pixel and digital parallel processing. Such… Expand
A Low-Power Vision System With Adaptive Background Subtraction and Image Segmentation for Unusual Event Detection
TLDR
A sensor embeds a low-level image processing technique that autonomously detects unusual events occurring in the scene, relying on adaptive background subtraction, resulting in an extremely low-power system suitable for embedded applications. Expand
An ultra-low power Address-Event sensor interface for energy-proportional time-to-information extraction
TLDR
An FPGA-based methodology that tags the AER spikes with timestamps to make them carriable by standard interfaces and uses a recursively divided clock generated on-chip by a pausable ring-oscillator to reduce power while keeping accuracy above 97% on timestamp. Expand
Energy-efficient design of an always-on smart visual trigger
TLDR
The optimized power management is enabled by defining an always-on camera interface for the System-on-Chip (SoC) processor, which is able to individually activate both the sensor and the processor while running at minimal power consumption. Expand
A Sub-mW IoT-Endnode for Always-On Visual Monitoring and Smart Triggering
TLDR
A fully programmable Internet of Things visual sensing node that targets sub-mW power consumption in always-on monitoring scenarios and achieves lower power consumption compared to MCU-based cameras with significantly lower on-board computing capabilities is presented. Expand
A low power smart camera for video surveillance and forensic applications
  • M. Lecca, Y. Zou, M. Gottardi
  • Computer Science
  • 2017 International Conference on Engineering, Technology and Innovation (ICE/ITMC)
  • 2017
TLDR
A custom, low-power surveillance camera for the control of isolated, peripheral areas, that are often chosen for illegal activities, like traffic of drugs, weapons and cigarettes is described, which expresses a high level of interest in such a system and a positive vote about its event detection performance. Expand
Enabling energy efficient machine learning on a Ultra-Low-Power vision sensor for IoT
TLDR
This paper presents the development, analysis, and embedded implementation of a realtime detection, classification and tracking pipeline able to exploit the full potential of background filtering Smart Vision Sensors (SVS). Expand
An Energy Optimized JPEG Encoder for Parallel Ultra-Low-Power Processing-Platforms
TLDR
This paper proposes an optimized algorithm implementation tailored for PULP (Parallel Ultra Low Power) processors, that permits to shrink the image size and the data to transmit and achieves the best trade-off between energy consumption and image distortion. Expand
A 142MOPS/mW integrated programmable array accelerator for smart visual processing
TLDR
Experimental results show that the proposed accelerator delivers 507 MOPS and 142 MOPS/mW on the target application, surpassing a low-power processor optimized for DSP applications by 6x in performance and by 10x in energy efficiency. Expand
EFCam: Configuration-Adaptive Fog-Assisted Wireless Cameras with Reinforcement Learning
TLDR
Deep reinforcement learning is applied to learn the optimal adaptation policy of an industrial wireless camera system, namely, EFCam, which uses low-power wireless communications and edge-fog computing to achieve cordless and energy-efficient visual sensing. Expand
People/Car Classification using an Ultra-Low-Power Smart Vision Sensor
TLDR
This paper evaluates the use of a smart sensor, the FORENSOR sensor, which embeds motion detection in hardware, in a classification scenario, and achieves 87% accuracy, and demonstrates the advantages of the sensor w.r.t frame-difference based ones. Expand
...
1
2
3
4
...

References

SHOWING 1-10 OF 46 REFERENCES
Low power high-performance smart camera system based on SCAMP vision sensor
TLDR
This work describes a stand-alone vision system incorporating a SCAMP-3 vision chip, an FPGA and an ARM Cortex-M3 microcontroller, and shows an algorithm developed for loiterer detection at 8fps has been shown to consume an average power of 5mW. Expand
Wi-FLIP: A wireless smart camera based on a focal-plane low-power image processor
TLDR
This paper presents Wi-FLIP, a vision-enabled WSN node resulting from the integration of FLIP-Q, a prototype vision chip, and Imotel, a commercial WSN platform, and reports experimental results for different algorithms, image resolutions and clock frequencies. Expand
Energy-efficient feedback tracking on embedded smart cameras by hardware-level optimization
TLDR
This paper introduces two methodologies to increase the energy-efficiency and battery-life of an embedded smart camera by hardware-level operations when performing object detection and tracking by delegating image down-sampling and cropping tasks to the micro-controller on the image sensor. Expand
A Bio-Inspired Vision Sensor With Dual Operation and Readout Modes
This paper presents a novel event-based vision sensor with two operation modes: intensity mode and spatial contrast detection. They can be combined with two different readout approaches: pulseExpand
A 3.4µW CMOS image sensor with embedded feature-extraction algorithm for motion-triggered object-of-interest imaging
TLDR
An event-based imaging system that generates signals only when an event happens, by monitoring temporal changes, has extraneous redundancy because the sensor may also respond to environmental conditions, such as change of illumination or background movement in addition to actual target objects. Expand
A 100 $\mu$ W 128 $\times$ 64 Pixels Contrast-Based Asynchronous Binary Vision Sensor for Sensor Networks Applications
TLDR
An ultra-low power 128 times 64 pixels vision sensor is here presented, featuring pixel-level spatial contrast extraction and binarization, and the pixel-embedded binary frame buffer allows the sensor to directly process visual information, such as motion and background subtraction, which are the most useful filters in machine vision applications. Expand
A 128 128 120 dB 15 s Latency Asynchronous Temporal Contrast Vision Sensor
This paper describes a 128 128 pixel CMOS vision sensor. Each pixel independently and in continuous time quantizes local relative intensity changes to generate spike events. These events appear atExpand
A 3.4-$\mu$ W Object-Adaptive CMOS Image Sensor With Embedded Feature Extraction Algorithm for Motion-Triggered Object-of-Interest Imaging
TLDR
A low-power object-adaptive CMOS imager, which suppresses spatial temporal bandwidth and saves a temporal bandwidth from the motion-triggered wakeup and object adaptive imaging, is reported. Expand
A low-bandwidth camera sensor platform with applications in smart camera networks
TLDR
The efforts to build a low-bandwidth wireless camera network platform, called CITRIC, and its applications in smart camera networks are provided, and concrete research results will be demonstrated in two areas, namely, distributed coverage hole identification and distributed object recognition. Expand
A 64$\,\times\,$64 CMOS Image Sensor With On-Chip Moving Object Detection and Localization
This paper presents a 64×64 CMOS image sensor with on-chip moving object detection and localization capability. Pixel-level storage elements (capacitors) enable the sensor to simultaneously outputExpand
...
1
2
3
4
5
...