Engineering Vehicles Detection Based on Modified Faster R-CNN for Power Grid Surveillance
Bibliographic record
Abstract
Engineering vehicles intrusion detection is a key problem for the security of power grid operation, which can warn of the regional invasion and prevent external damage from architectural construction. In this paper, we propose an intelligent surveillance method based on the framework of Faster R-CNN for locating and identifying the invading engineering vehicles. In our detection task, the type of the objects is varied and the monitoring scene is large and complex. In order to solve these challenging problems, we modify the network structure of the object detection model by adjusting the position of the ROI pooling layer. The convolutional layer is added to the feature classification part to improve the accuracy of the detection model. We verify that increasing the depth of the feature classification part is effective for detecting engineering vehicles in realistic transmission lines corridors. We also collect plenty of scene images taken from the monitor site and label the objects to create a fine-tuned dataset. We train the modified deep detection model based on the technology of transfer learning and conduct training and test on the newly labeled dataset. Experimental results show that the proposed intelligent surveillance method can detect engineering vehicles with high accuracy and a low false alarm rate, which can be used for the early warning of power grid surveillance.
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) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 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".