Abstract | ||
---|---|---|
This live demonstration shows a biology-inspired highly efficient approach to high-speed video data acquisition based on the ATIS image sensor technology. Gray-level image information is acquired pixel-individually and event-driven at a temporal resolution equivalent to thousands of frames-per-second while the data rate is kept at a fraction of the rate encountered with conventional high-speed cameras by means of near ideal pixel-level redundancy suppression. The presented ATIS camera combines three consecutive exposure estimation / measurement steps that apply different trade-offs between measurement speed, accuracy and noise. Early estimates of pixel values deliver between 10 and 100 times faster updates than the eventual full-swing integrating exposure measurement that reaches 1 kFPS equivalent temporal resolution at sunny outdoor conditions. In addition, a miniaturized ATIS camera is demonstrated that constitutes one of the first (if not the first) neuromorphic device designed specifically for application in a commercial medical product. |
Year | DOI | Venue |
---|---|---|
2014 | 10.1109/BioCAS.2014.6981677 | BioCAS |
Keywords | Field | DocType |
bio-optics,cameras,data acquisition,high-speed optical techniques,ATIS image sensor technology,event driven high speed imaging,gray level image information,high speed video data acquisition,live demonstration,pixel level redundancy suppression,temporal resolution | Computer vision,Image sensor,Computer graphics (images),Computer science,Electronic engineering,Artificial intelligence | Conference |
ISSN | Citations | PageRank |
2163-4025 | 0 | 0.34 |
References | Authors | |
2 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Matolin, D. | 1 | 0 | 0.34 |
Orchard, G. | 2 | 24 | 3.77 |
Floderer, J.-B. | 3 | 0 | 0.34 |
Posch, C. | 4 | 0 | 0.34 |