Addressing lighting and bounding box accuracy for the Embedded Automated Generator of Labeled Images (EAGL-I) system
Bibliographic record
Abstract
The Embedded Automated Generator of Labeled Images (EAGL-I) system is a tool for generating labeled images, particularly for data-driven methods, such as deep learning models. The system has already generated hundreds of thousands of images of weeds and crops. We present modifications made to the original system that are based on the experiences gathered from generating such large-scale datasets. The improvements relate to lighting conditions, ease of use, refined image segmentation, and pathfinding for camera-movements. To address lighting conditions, we made three major changes to the hardware. First, the blue keying fabric was replaced by solid black panels, mitigating reflections and achieving reliable color accuracy; second, sunlight entering the room through a window is diffused and partially blocked by a screen, achieving consistent and uniform lighting of the imaging environment; third, dimmable LED lights are installed allowing us to image with lower ISO and to reduce noise in the resulting images. A YOLO machine learning model was trained to replace the previous methods of estimating bounding boxes around the plants. This new way of creating bounding boxes adapts to different plant architectures, such as grasses or different kind of dicots. Finally, we implemented a version of the A* pathfinding algorithm to define save zones through which the camera will not be moved. Overall, these modifications improved system performance and image quality significantly, while making EAGL-I easier to use. We have extended potential applications of EAGL-I, particularly for plant phenotyping research and in fine-tuning machine learning models for image analysis.
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 | 0.004 | 0.021 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.003 | 0.003 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.007 | 0.002 |
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".