Computer vision and scene understanding
Researches computer vision, scene understanding and medical image analysis, including scene-text detection, railway inspection, uncertainty estimation and neural networks for segmentation and 3D reconstruction.
Published research, with a short guide to the question and method.
8 selected publications
IEEE Access, 13, pp. 51645–51659
Studies conformal prediction for quantifying uncertainty in HER2 status classification from breast-cancer immunohistochemistry images.
SN Computer Science, 5, 82
Evaluates transfer learning with AlexNet and ResNet50 to locate and classify parasite eggs in images from a low-cost USB microscope.
Scientific Reports, 13, 455
Develops FracReconNet to reconstruct fractured proximal femurs in 3D from two radiographic views, using fracture representations and training-data augmentation.
Scientific Reports, 13, 20098
In a medical-imaging collaboration, introduces MANet: a U-Net-based network with residual learning and channel and spatial attention for liver-tumour segmentation in CT images.
Electronics, 9(1), 117
Combines text segmentation with polygon offsets and border information to distinguish nearby and irregularly shaped text in natural images.
IEEE ICSIPA 2017, pp. 419–424
Investigates image-based detection of missing fasteners for railway inspection.
IEEE ICSIPA 2015, pp. 220–225
Uses a convolutional neural network to locate Thai text with different orientations in natural-scene images.
Real-Time Imaging, 11(3), pp. 172–185
Develops a codebook model for separating foreground objects from changing backgrounds in video.
Try another keyword, researcher or year.
Publication details follow the linked sources, checked 8 September 2026. The roster follows the 2022 directory; individual publications do not establish current center project ownership or a PI appointment.
← All researcher portfolios