Image Data Labeler and Dataset Synthesizer for Black Box Detection Project
Image
PolygonI led the creation and expansion of a black box object detection dataset using image fusion and annotation techniques. Images from drone aerial surveys were meticulously annotated with the Labelme software, generating diverse and representative labels under varied conditions. Synthetic augmentation broadened dataset diversity and specific evaluation metrics ensured quality and realism. • Labeled black box objects in UAV images using polygons • Augmented image data with lighting, weather, and seasonal variability • Evaluated synthetic image fidelity with dedicated metrics • Utilized Labelme and image processing pipelines for annotation tasks.
2023 - Present