Toward in-sensor imaging classification enabled by on-chip all-optical modulation and photonic neural networks
Bibliographic record
Abstract
Image sensors play a crucial role in autonomous driving and security. However, the separation of sensors and processors results in large amounts of redundant data being exchanged between sensory terminals and computing units. Here, leveraging silicon photonic integrated technology, we propose an in-sensor imaging classification solution that combines sensing components with photonic neural networks. The sensing components consist of a microring array which can convert visible light signals from free space into near-infrared light signals propagating through silicon waveguides, which are subsequently processed by photonic neural networks. The proposed all-optical modulator is based on a pn-doped microring resonator, enabling light-to-light modulation through two mechanisms: the thermo-optic and plasma-dispersion effects. We demonstrate that these effects can be controlled by shorting the pn-junctions, achieving a modulation depth of 15 dB. Two cascaded microrings—used for converting visible light signals and applying weight signals—demonstrate the ability to perform dot product operations with an effective resolution of near 8 bits, sufficient for data processing. Furthermore, we validated the in-sensor scheme using the Modified National Institute of Standards and Technology (MNIST) dataset, achieving a recognition accuracy of 96.82%. These results pave the way for advanced ‘in-sensor’ imaging classification technologies.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".