Klaas Dijkstra

Klaas Dijkstra
  • Doctor of Philosophy
  • Professor at Stenden University of Applied Sciences

About

29
Publications
16,813
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
316
Citations
Introduction
Klaas Dijkstra is a professor of applied sciences in computer vision & data science at the NHL Stenden University of Applied Sciences. His main research interest is in computer vision, deep learning and hyperspectral imaging. He obtained his PhD degree from the University of Groningen.
Current institution
Stenden University of Applied Sciences
Current position
  • Professor

Publications

Publications (29)
Chapter
Object detection using CNNs requires a large amount of data to achieve decent performance in real-world scenarios. The creation of traditional datasets involves acquiring numerous images and manually annotating them. In this paper, we introduce a method for simulating apple orchards utilizing the Unity 3D engine. We created a tool that uses this si...
Chapter
Full-text available
Despite the notable achievements of deep object detection models, a major challenge remains to be the need for vast amounts of training data. The process of acquiring such real-world data is laborious, prompting the exploration of new research directions such as synthetic data generation. In this study, we assess the capability of two distinct synt...
Preprint
Full-text available
Deep object detection models have achieved notable successes in recent years, but one major obstacle remains: the requirement for a large amount of training data. Obtaining such data is a tedious process and is mainly time consuming, leading to the exploration of new research avenues like synthetic data generation techniques. In this study, we expl...
Preprint
Full-text available
Despite the notable accomplishments of deep object detection models, a major challenge that persists is the requirement for extensive amounts of training data. The process of procuring such real-world data is a laborious undertaking, which has prompted researchers to explore new avenues of research, such as synthetic data generation techniques. Thi...
Article
Full-text available
Laser speckle contrast imaging (LSCI) is so sensitive to motion that it can measure the movement of red blood cells. However, this extreme sensitivity to motion is also its pitfall as the clinical translation of LSCI is slowed down due to the inability to deal with motion artefacts. In this paper we study the effectiveness of a real-time, multi-spe...
Preprint
Full-text available
Given the hyper-spectral imaging unique potentials in grasping the polymer characteristics of different materials, it is commonly used in sorting procedures. In a practical plastic sorting scenario, multiple plastic flakes may overlap which depending on their characteristics, the overlap can be reflected in their spectral signature. In this work, w...
Preprint
Full-text available
The importance of plastic waste recycling is undeniable. In this respect, computer vision and deep learning enable solutions through the automated analysis of short-wave-infrared hyper-spectral images of plastics. In this paper, we offer an exhaustive empirical study to show the importance of efficient model selection for resolving the task of hype...
Article
Full-text available
Context. At GeV energies, the sky is dominated by the interstellar emission from the Galaxy. With limited statistics and spatial resolution, accurately separating point sources is therefore challenging. Aims. Here we present the first application of deep learning based algorithms to automatically detect and classify point sources from gamma-ray dat...
Preprint
Full-text available
At GeV energies, the sky is dominated by the interstellar emission from the Galaxy. With limited statistics and spatial resolution, accurate separation of point sources is therefore challenging. Here we present the first application of deep learning based algorithms to automatically detect and classify point sources from gamma-ray data. To detect p...
Article
This paper presents CentroidNetV2, a novel hybrid Convolutional Neural Network (CNN) that has been specifically designed to segment and count many small and connected object instances. This complete redesign of the original CentroidNet uses a CNN backbone to regress a field of centroid-voting vectors and border-voting vectors. The segmentation mask...
Thesis
Recently there has been much interest in hyperspectral imaging research and applications. Hyperspectral cameras collect image data from across the electromagnetic spectrum. These cameras aim to capture a spectrogram for each pixel to form a hyperspectral cube. The application area for these types of cameras is broad and varies from vegetation inspe...
Article
Full-text available
Anastomotic leakage is a worldwide problem in gastrointestinal surgery which seems to be related to the state of microcirculation. Laser speckle contrast imaging (LSCI) could give surgeons insight in the state of microcirculation to attune the site of anastomosis. This work studies the feasibility of LSCI as a tool for this purpose. An experimental...
Article
Full-text available
Precision agriculture using unmanned aerial vehicles (UAVs) is gaining popularity. These UAVs provide a unique aerial perspective suitable for inspecting agricultural fields. With the use of hyperspectral cameras, complex inspection tasks are being automated. Payload constraints of UAVs require low weight and small hyperspectral cameras; however, s...
Chapter
In precision agriculture, counting and precise localization of crops is important for optimizing crop yield. In this paper CentroidNet is introduced which is a Fully Convolutional Neural Network (FCNN) architecture specifically designed for object localization and counting. A field of vectors pointing to the nearest object centroid is trained and c...
Conference Paper
Full-text available
In precision agriculture, counting and precise localization of crops is important for optimizing crop yield. In this paper CentroidNet is introduced which is a Fully Convolutional Neural Network (FCNN) architecture specifically designed for object localization and counting. A field of vectors pointing to the nearest object centroid is trained and c...
Conference Paper
Full-text available
This study focuses on supplementing data sets with data of absent classes by using other, similar data sets in which these classes are represented. The data is generated using Gener-ative Adversarial Nets (GANs) trained on the CelebA and MNIST datasets. In particular we use and compare Coupled GANs (CoGANs), Auxiliary Classifier GANs (AC-GANs) and...
Conference Paper
Full-text available
Reducing the use of pesticides by early visual detection of diseases in precision agriculture is important. Because of the color similarity between potato-plant diseases, narrow band hyper-spectral imaging is required. Payload constraints on unmanned aerial vehicles require reduction of spectral bands. Therefore, we present a methodology for per-pa...
Article
Full-text available
The quality of sample inoculation is critical to achieve optimal yield of discrete colonies in both monomicrobial and polymicrobial samples to perform identification and antibiotic susceptibility testing. Consequently, we compared the performance between the InoqulA (BD Kiestra), the WASP (Copan) and manual inoculation. Defined mono- and polymicrob...
Article
This study quantifies the effect of oil layer thickness on entrainment and dispersion of oil into seawater, using a plunging jet with a camera system. In contrast to what is generally assumed, we revealed that for the low viscosity "surrogate MC252 oil" we used, entrainment rate is directly proportional to layer thickness. Furthermore, the volume o...
Conference Paper
Full-text available
Twirre is a new architecture for mini-UAV plat-forms designed for autonomous flight in both GPS-enabled and GPS-deprived applications. The architecture consists of low-cost hardware and software components. High-level control software enables autonomous operation. Ex-changing or upgrading hardware components is straightforward and the architecture...
Conference Paper
Full-text available
Twirre Architecture for autonomous mini-UAVs using interchangeable commodity components • All sensors and processing on-board • Low-cost components • Upgradable and extendable • Useful in multiple applications Result of experiments Conclusions Architecture Cascade control system • High level: simulation of human stick inputs • Low level : exchangea...
Conference Paper
Objectives The emergence of automation in bacteriology opens a new era in clinical diagnostic laboratories. Automation is impacting the management and the laboratory workflow but also offers new perspectives for research and development in bacteriology. Sample inoculation is a fastidious and repetitive process that represents a significant laborat...
Article
Full-text available
A survey [1] was performed on 22 stan-dards for parallel computing and pro-gramming. OpenMP was chosen as the standard for multi-core CPU program-ming and OpenCL as the standard for GPU programming. Portability, vendor independence and efficiently parallelizing the code were key decision factors. The economical parallelization with respect to devel...
Conference Paper
Full-text available
The increasing popularity of machine vision based solutions in common applications calls for a structured approach for incorporating the end user's domain knowledge and limiting the solution's dependency on expert knowledge. We propose a framework facilitating optimized classification results and will show several approaches in which prior knowledg...
Conference Paper
Full-text available
Objectives Methods Conclusions Results BD Kiestra provides a workflow where digital images of Petri-dishes used in antibiotic susceptibility testing by disc diffusion can be automatically analyzed using machine vision algorithms. The objective of this study is to develop and test a system which automatically optimizes a zone measurement algorithm t...
Conference Paper
Full-text available
The increasing popularity of machine vision based solutions in common applications calls for a structured approach for incorporating the end user's domain knowledge and limiting the solution's dependency on expert knowledge. We propose a framework facilitating optimized classifi-cation results and will show several approaches in which prior knowled...

Network

Cited By