Name
Abstract | ||
|---|---|---|
This paper presents compressive image sensor techniques based on sparse measurement matrices. Existing compressive sensing (CS) CMOS image sensors use dense random measurement matrices, which face the challenges of excessive hardware overhead and large signal swing requirement. The sparse measurement matrices proposed in this paper dramatically simplify the circuit implementation and relax the signal swing requirement. The validity of the proposed sparse measurement matrices is justified and their performances are compared with results reported in literature as well as results obtained using dense random measurement matrices in our own study. Circuit techniques to implement CS image sensors based on sparse measurement matrices are discussed and a 250×250 Pixel image sensor with a compression rate of 250/62 is designed using a 0.13 μm CMOS technology. Circuit simulation shows that the image sensor can achieve a peak signal to noise ratio of 31.7dB. |
Year | DOI | Venue |
|---|---|---|
2016 | 10.1109/SOCC.2016.7905472 | 2016 29th IEEE International System-on-Chip Conference (SOCC) |
Keywords | Field | DocType |
image sensor,compressive sensing | Peak signal-to-noise ratio,Data compression ratio,Image sensor,Computer science,Matrix (mathematics),Electronic engineering,Pixel,Image resolution,Sparse matrix,Compressed sensing | Conference |
ISBN | Citations | PageRank |
978-1-5090-1368-5 | 0 | 0.34 |
References | Authors | |
5 | 3 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Stefan Leitner | 1 | 0 | 1.69 |
| Haibo Wang | 2 | 66 | 13.83 |
| Spyros Tragoudas | 3 | 625 | 88.87 |