Improving Accuracy of Object Detection in Autonomous Drones with Convolutional Neural Networks
Bibliographic record
Abstract
Increasing use of autonomous drones in such areas as agriculture, disaster response and surveillance means that an effective and precise method of object recognition is becoming more important. In this study, we use Convolutional Neural Networks (CNNs) to bring additional precision to autonomous drone object detecting systems. While conventional machine learning approaches fail miserably in having complicated contexts and real time processing, CNN have sucessfully handles visual identification task. Based on state of the art methods being data augmentation and transfer learning, plus being real time with data augmentation of the raw data on the edge processor, this research introduces a convolutional neural network (CNN) model which has been fine-tuned and can be used for drones. The model is trained and tested on a big dataset of aerial photos acquired by drones. The item classification on this dataset varies from lighting conditions, to weather variables, etc. A considerable increase in the object detection accuracy is found from experimental findings in reducing false positives and improving resilience in dynamic contexts. Furthermore, the suggested model allows for detection with little latency, making it a good model for drones that must be deployed in real-time. Not only does this work contribute to bolstering rapidly growing autonomous drone navigation, it has great promise in environmental monitoring, SAR, and precision agriculture, and other applications. The components of the models are being researched on how scalable they are and how do they scale up on hardware that has limited resources.
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) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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".