Face recognition using traditional machine learning algorithms and deep neural networks with application to face verific...
Electrical Engineering | Circuits and Systems
Flowers, leaves or both? How to obtain suitable images for automated plant identification
Published in Plant Methods
BackgroundDeep learning algorithms for automated plant identification need large quantities of precisely labelled images in order to produce reliable classification results. Here, we explore what kind of perspectives and their combinations contain more characteristic information and therefore allow for higher identification accuracy.ResultsWe devel...
Visualization of the Rolling Contact Fatigue Cracks in Rail Tracks with a Magnetooptical Sensor
Published in Journal of Nondestructive Evaluation
The rolling contact fatigue cracks (RCF), produced at the surface of the rail tracks, can be detected using a magnetooptical (MO) sensor. Rail tracks are carbon steels with pearlite microstructure. This microstructure has a lamellar texture composed of alternating layers of ferrite and cementite. Both phases are soft ferromagnetic materials at room...
First-person Video Analysis for Evaluating Skill Level in the Humanitude Tender-Care Technique
In this paper, we describe a wearable first-person video (FPV) analysis system for evaluating the skill levels of caregivers. This is a part of our project that aims to quantize and analyze the tender-care technique known as Humanitude by using wearable sensing and AI technology devices. Using our system, caregivers can evaluate and elevate their c...
Vision-based detection of container lock holes using a modified local sliding window method
Published in EURASIP Journal on Image and Video Processing
Container yards have been facing the increase of freight volume. In order to improve the efficiency of container handling, automatic stations have been established in many terminals. However, current container handling still needs a manual operation to locate container lock holes. Hence, it is inefficient and potential to risk workers’ health under...
Deep learning investigation for chess player attention prediction using eye-tracking and game data
This article reports on an investigation of the use of convolutional neural networks to predict the visual attention of chess players. The visual attention model described in this article has been created to generate saliency maps that capture hierarchical and spatial features of chessboard, in order to predict the probability fixation for individu...
Cell extraction automation in single cell surgery using the displacement method
Published in Biomedical Microdevices
Micromanipulation is the precise in vitro handling and study of individual biological cells, where the smallest error can be disastrous. One such example is the extraction of cellular material from multicellular organisms, such as cells from early stage embryos. In this paper, we propose automation methods for the extraction and retrieval of indivi...
Learning segmentation from multiple datasets with different label sets
The thesis deals with the task of object instance segmentation, which aims to learn a per-pixel mask of every object instance of defined classes within an image. Stateof-the-art convolutional neural networks require large quantities of annotated training data. Segmentation masks are costly to annotate.The work focuses on training instance segmentat...
Learning human body and human action representations from visual data
The focus of visual content is often people. Automatic analysis of people from visual data is therefore of great importance for numerous applications in content search, autonomous driving, surveillance, health care, and entertainment.The goal of this thesis is to learn visual representations for human understanding. Particular emphasis is given to ...
Multi-modal representation learning towards visual reasoning
In this thesis, we are interested in Visual Question Answering (VQA), which consists in building models that answer any natural language question about any image. Because of its nature and complexity, VQA is often considered as a proxy for visual reasoning. Classically, VQA architectures are designed as trainable systems that are provided with imag...
