The last decade has seen an increasing demand from the industrial field of computerized visual inspection. Applications rapidly become more complex and often with more demanding real time constraints. However, from 2004 onwards the clock frequency of CPUs has not increased significantly. Computer Vision applications have an increasing demand for more processing power but are limited by the performance capabilities of sequential processor architectures. The only way to get more performance using commodity hardware, like multi-core processors and graphics cards, is to go for parallel programming. This article focuses on the practical question: How can the processing time for vision algorithms be improved, by parallelization, in an economical way and execute them on multiple platforms?
DOCUMENT
Estimation of the factor model by unweighted least squares (ULS) is distribution free, yields consistent estimates, and is computationally fast if the Minimum Residuals (MinRes) algorithm is employed. MinRes algorithms produce a converging sequence of monotonically decreasing ULS function values. Various suggestions for algorithms of the MinRes type are made for confirmatory as well as for exploratory factor analysis. These suggestions include the implementation of inequality constraints and the prevention of Heywood cases. A simulation study, comparing the bootstrap standard deviations for the parameters with the standard errors from maximum likelihood, indicates that these are virtually equal when the score vectors are sampled from the normal distribution. Two empirical examples demonstrate the usefulness of constrained exploratory and confirmatory factor analysis by ULS used in conjunction with the bootstrap method.
DOCUMENT
The aviation industry needs led to an increase in the number of aircraft in the sky. When the number of flights within an airspace increases, the chance of a mid-air collision increases. Systems such as the Traffic Alert and Collision Avoidance System (TCAS) and Airborne Collision Avoidance System (ACAS) are currently used to alert pilots for potential mid-air collisions. The TCAS and the ACAS use algorithms to perform Aircraft Trajectory Predictions (ATPs) to detect potential conflicts between aircrafts. In this paper, three different aircraft trajectory prediction algorithms named Deep Neural Network (DNN), Random Forest (RF) and Extreme Gradient Boosting were implemented and evaluated in terms of their accuracy and robustness to predict the future aircraft heading. These algorithms were as well evaluated in the case of adversarial samples. Adversarial training is applied as defense method in order to increase the robustness of ATPs algorithms against the adversarial samples. Results showed that, comparing the three algorithm’s performance, the extreme gradient boosting algorithm was the most robust against adversarial samples and adversarial training may benefit the robustness of the algorithms against lower intense adversarial samples. The contributions of this paper concern the evaluation of different aircraft trajectory prediction algorithms, the exploration of the effects of adversarial attacks, and the effect of the defense against adversarial samples with low perturbation compared to no defense mechanism.
DOCUMENT
The increasing amount of electronic waste (e-waste) urgently requires the use of innovative solutions within the circular economy models in this industry. Sorting of e-waste in a proper manner are essential for the recovery of valuable materials and minimizing environmental problems. The conventional e-waste sorting models are time-consuming processes, which involve laborious manual classification of complex and diverse electronic components. Moreover, the sector is lacking in skilled labor, thus making automation in sorting procedures is an urgent necessity. The project “AdapSort: Adaptive AI for Sorting E-Waste” aims to develop an adaptable AI-based system for optimal and efficient e-waste sorting. The project combines deep learning object detection algorithms with open-world vision-language models to enable adaptive AI models that incorporate operator feedback as part of a continuous learning process. The project initiates with problem analysis, including use case definition, requirement specification, and collection of labeled image data. AI models will be trained and deployed on edge devices for real-time sorting and scalability. Then, the feasibility of developing adaptive AI models that capture the state-of-the-art open-world vision-language models will be investigated. The human-in-the-loop learning is an important feature of this phase, wherein the user is enabled to provide ongoing feedback about how to refine the model further. An interface will be constructed to enable human intervention to facilitate real-time improvement of classification accuracy and sorting of different items. Finally, the project will deliver a proof of concept for the AI-based sorter, validated through selected use cases in collaboration with industrial partners. By integrating AI with human feedback, this project aims to facilitate e-waste management and serve as a foundation for larger projects.
Horse riding falls under the “Sport for Life” disciplines, where a long-term equestrian development can provide a clear pathway of developmental stages to help individuals, inclusive of those with a disability, to pursue their goals in sport and physical activity, providing long-term health benefits. However, the biomechanical interaction between horse and (disabled) rider is not wholly understood, leaving challenges and opportunities for the horse riding sport. Therefore, the purpose of this KIEM project is to start an interdisciplinary collaboration between parties interested in integrating existing knowledge on horse and (disabled) rider interaction with any novel insights to be gained from analysing recently collected sensor data using the EquiMoves™ system. EquiMoves is based on the state-of-the-art inertial- and orientational-sensor system ProMove-mini from Inertia Technology B.V., a partner in this proposal. On the basis of analysing previously collected data, machine learning algorithms will be selected for implementation in existing or modified EquiMoves sensor hardware and software solutions. Target applications and follow-ups include: - Improving horse and (disabled) rider interaction for riders of all skill levels; - Objective evidence-based classification system for competitive grading of disabled riders in Para Dressage events; - Identifying biomechanical irregularities for detecting and/or preventing injuries of horses. Topic-wise, the project is connected to “Smart Technologies and Materials”, “High Tech Systems & Materials” and “Digital key technologies”. The core consortium of Saxion University of Applied Sciences, Rosmark Consultancy and Inertia Technology will receive feedback to project progress and outcomes from a panel of international experts (Utrecht University, Sport Horse Health Plan, University of Central Lancashire, Swedish University of Agricultural Sciences), combining a strong mix of expertise on horse and rider biomechanics, veterinary medicine, sensor hardware, data analysis and AI/machine learning algorithm development and implementation, all together presenting a solid collaborative base for derived RAAK-mkb, -publiek and/or -PRO follow-up projects.
The utilization of drones in various industries, such as agriculture, infrastructure inspection, and surveillance, has significantly increased in recent years. However, navigating low-altitude environments poses a challenge due to potential collisions with “unseen” obstacles like power lines and poles, leading to safety concerns and equipment damage. Traditional obstacle avoidance systems often struggle with detecting thin and transparent obstacles, making them ill-suited for scenarios involving power lines, which are essential yet difficult to perceive visually. Together with partners that are active in logistics and safety and security domains, this project proposal aims at conducting feasibility study on advanced obstacle detection and avoidance system for low-flying drones. To that end, the main research question is, “How can AI-enabled, robust and module invisible obstacle avoidance technology can be developed for low-flying drones? During this feasibility study, cutting-edge sensor technologies, such as LiDAR, radar, camera and advanced machine learning algorithms will be investigated to what extent they can be used be to accurately detect “Not easily seen” obstacles in real-time. The successful conclusion of this project will lead to a bigger project that aims to contribute to the advancement of drone safety and operational capabilities in low-altitude environments, opening new possibilities for applications in industries where low-flying drones and obstacle avoidance are critical.
