Results 191 to 200 of about 119,156 (290)
A Foot-Arch Parameter Measurement System Using a RGB-D Camera. [PDF]
Chun S, Kong S, Mun KR, Kim J.
europepmc +1 more source
Accurately estimating forest age is key to understanding how forests recover and evaluating restoration success. We developed a two‐step deep learning approach using historical greyscale aerial photographs to map forest age at fine spatial scales. By combining a pre‐trained model with localized fine‐tuning, our U‐Net + ResNet50 architecture achieved ...
Ying Ki Law +10 more
wiley +1 more source
Indoor Scene Point Cloud Registration Algorithm Based on RGB-D Camera Calibration. [PDF]
Tsai CY, Huang CH.
europepmc +1 more source
The single‐scan approach to terrestrial laser scanning (TLS) and the self‐terrain‐normalized form of drone‐based digital aerial photogrammetry (DAP) offer practical options for rapid assessment of the vegetation structure in tropical landscapes.
Magnus Onyiriagwu +7 more
wiley +1 more source
Real-Time External Respiratory Motion Measuring Technique Using an RGB-D Camera and Principal Component Analysis. [PDF]
Wijenayake U, Park SY.
europepmc +1 more source
Big Bird: A global dataset of birds in drone imagery annotated to species level
Drones are a valuable tool for surveying birds, but manually detecting and identifying birds in drone images is costly. We assembled a diverse dataset of 23 865 images of birds captured with 21 different drones across 11 countries. We labelled 4824 of these images, detailing the location, species, posture category, age category, and sex of 49 990 birds
Joshua P. Wilson +19 more
wiley +1 more source
Objective Arm Swing Analysis in Early-Stage Parkinson's Disease Using an RGB-D Camera (Kinect®). [PDF]
Ospina BM +5 more
europepmc +1 more source
This study presents a semi‐automated, rule‐based image analysis pipeline to detect ice seals in aerial surveys of the Western Antarctic Peninsula during an unusually low sea ice year. By using simple hierarchical clustering instead of deep learning, the method substantially reduced human annotation effort while achieving 82% recall, identifying 758 ...
Claire McGinnity +8 more
wiley +1 more source
Detecting Target Objects by Natural Language Instructions Using an RGB-D Camera. [PDF]
Bao J, Jia Y, Cheng Y, Tang H, Xi N.
europepmc +1 more source
This study presents a UAV‐based framework that integrates deep learning‐based super‐resolution reconstruction and an enhanced YOLO detector to improve centimetre‐scale benthic organism monitoring. Using hermit crabs in Lake Hamana, a coastal lagoon in Japan, as a case study, the method substantially enhanced small‐object detection performance ...
Fan Zhao +10 more
wiley +1 more source

