AgMIC: Agricultural Masked Image Consistency for Cross-Domain Segmentation
Bibliographic record
Abstract
Canola crop segmentation is critical for monitoring crop coverage and health, and identifying growth variability. It allows farmers to implement variable rate applications based on localized crop conditions. Recent advances in deep learning (DL) have shown promising results for crop segmentation using high-resolution imagery. Although DL-based models perform well on their source domain (ground vehicle imagery), they exhibit significant performance drops when applied to the target domain (unmanned aerial vehicles). This degradation occurs due to substantial differences in viewing perspective, image resolution, object appearance scale, and spatial context between the two domains. To bridge this domain gap, this paper presents agricultural masked image consistency (AgMIC), an enhanced unsupervised domain adaptation (UDA) framework for cross-domain canola crop segmentation. Unlike existing MIC method that employs random masking strategies, AgMIC introduces agricultural pattern-aware masking, which leverages crop density patterns to preserve structural field information for enhanced cross-domain contextual understanding. Additionally, it incorporates confidence-aware pseudo-label enhancement with adaptive thresholding to filter unreliable predictions. Experimental results demonstrate that AgMIC achieves superior performance with a mean intersection over union of 0.7603, outperforming state-of-the-art UDA methods, including MIC (0.7436), RUDA (0.7326), and HRDA (0.7248).
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 | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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".