In recent years, drones have increasingly supported First Responders (FRs) in monitoring incidents and providing additional information. However, analysing drone footage is time-intensive and cognitively demanding. In this research, we investigate the use of AI models for the detection of humans in drone footage to aid FRs in tasks such as locating victims. Detecting small-scale objects, particularly humans from high altitudes, poses a challenge for AI systems. We present first steps of introducing and evaluating a series of YOLOv8 Convolutional Neural Networks (CNNs) for human detection from drone images. The models are fine-tuned on a created drone image dataset of the Dutch Fire Services and were able to achieve a 53.1% F1-Score, identifying 439 out of 825 humans in the test dataset. These preliminary findings, validated by an incident commander, highlight the promising utility of these models. Ongoing efforts aim to further refine the models and explore additional technologies.
MULTIFILE
We present a novel anomaly-based detection approach capable of detecting botnet Command and Control traffic in an enterprise network by estimating the trustworthiness of the traffic destinations. A traffic flow is classified as anomalous if its destination identifier does not origin from: human input, prior traffic from a trusted destination, or a defined set of legitimate applications. This allows for real-time detection of diverse types of Command and Control traffic. The detection approach and its accuracy are evaluated by experiments in a controlled environment.
DOCUMENT
Frontline professionals such as social workers and civil servants play a crucial role in countering violent extremism.Because of their direct contac twith society,first liners are tasked with detecting individuals that may threaten national security and the democratic rule of law. Preliminary screening takes place during the pre-crime phase. However, without clear evidence or concrete indicators of unlawful action or physical violence, it is challenging to determine when someone poses a threat. There are no set patterns that can be used to identify cognitive radicalization processes that will result in violent extremism. Furthermore, prevention targets ideas and ideologies with no clear framework for assessing terrorism-risk. This article examines how civil servants responsible for public order, security and safety deal with their mandate to engage in early detection, and discusses the side effects that accompany this practice. Based on openinterviews with fifteen local security professionals in the Netherlands, we focus here on the risk assessments made by these professionals. To understand their performance, we used the following two research questions: First, what criteria do local security professionals use to determine whether or not someone forms a potential risk? Second, how do local security professionals substantiate their assessments of the radicalization processes that will develop into violent extremism? We conclude that such initial risk weightings rely strongly on ‘gut feelings’ or intuition. We conclude that this subjectivitymayleadto prejudiceand/oradministrativearbitrariness in relationtopreliminary risk assessment of particular youth.
DOCUMENT
In recent years, the fight against terrorism and political violence has focused more on anticipating the threats that they pose. Therefore, early detection of ideas by local professionals has become an important part of the preventive approach in countering radicalization. Frontline workers who operate in the arteries of society are encouraged to identify processes toward violent behavior at an early stage. To date, however, little is known about how these professionals take on this screening task at their own discretion. Research from the Netherlands suggests that subjective assessment appears to exist. In this article, we argue that the absence of a clear norm for preliminary judgments affects prejudice or administrative arbitrariness, which may cause side effects due to unjustified profiling.
DOCUMENT
Standard SARS-CoV-2 testing protocols using nasopharyngeal/throat (NP/T) swabs are invasive and require trained medical staff for reliable sampling. In addition, it has been shown that PCR is more sensitive as compared to antigen-based tests. Here we describe the analytical and clinical evaluation of our in-house RNA extraction-free saliva-based molecular assay for the detection of SARS-CoV-2. Analytical sensitivity of the test was equal to the sensitivity obtained in other Dutch diagnostic laboratories that process NP/T swabs. In this study, 955 individuals participated and provided NP/T swabs for routine molecular analysis (with RNA extraction) and saliva for comparison. Our RT-qPCR resulted in a sensitivity of 82,86% and a specificity of 98,94% compared to the gold standard. A false-negative ratio of 1,9% was found. The SARS-CoV-2 detection workflow described here enables easy, economical, and reliable saliva processing, useful for repeated testing of individuals.
LINK
Explainable Artificial Intelligence (XAI) aims to provide insights into the inner workings and the outputs of AI systems. Recently, there’s been growing recognition that explainability is inherently human-centric, tied to how people perceive explanations. Despite this, there is no consensus in the research community on whether user evaluation is crucial in XAI, and if so, what exactly needs to be evaluated and how. This systematic literature review addresses this gap by providing a detailed overview of the current state of affairs in human-centered XAI evaluation. We reviewed 73 papers across various domains where XAI was evaluated with users. These studies assessed what makes an explanation “good” from a user’s perspective, i.e., what makes an explanation meaningful to a user of an AI system. We identified 30 components of meaningful explanations that were evaluated in the reviewed papers and categorized them into a taxonomy of human-centered XAI evaluation, based on: (a) the contextualized quality of the explanation, (b) the contribution of the explanation to human-AI interaction, and (c) the contribution of the explanation to human- AI performance. Our analysis also revealed a lack of standardization in the methodologies applied in XAI user studies, with only 19 of the 73 papers applying an evaluation framework used by at least one other study in the sample. These inconsistencies hinder cross-study comparisons and broader insights. Our findings contribute to understanding what makes explanations meaningful to users and how to measure this, guiding the XAI community toward a more unified approach in human-centered explainability.
MULTIFILE
In mobile robotics, LASER scanners have a wide spectrum of indoor and outdoor applications, both in structured and unstructured environments, due to their accuracy and precision. Most works that use this sensor have their own data representation and their own case-specific modeling strategies, and no common formalism is adopted. To address this issue, this manuscript presents an analytical approach for the identification and localization of objects using 2D LiDARs. Our main contribution lies in formally defining LASER sensor measurements and their representation, the identification of objects, their main properties, and their location in a scene. We validate our proposal with experiments in generic semi-structured environments common in autonomous navigation, and we demonstrate its feasibility in multiple object detection and identification, strictly following its analytical representation. Finally, our proposal further encourages and facilitates the design, modeling, and implementation of other applications that use LASER scanners as a distance sensor.
DOCUMENT
The COVID-19 pandemic led to an accelerated implementation of digital solutions, such as online proctoring. In this paper we discuss how the use of an ethical matrix may influence the way in which digital solutions are applied. To initiate an ethical discussion, we conducted an online workshop with educators, examiners, controllers, and students to identify risks and opportunities of online proctoring for various stakeholders. We used the Ethical Matrix to structure the meeting. We compared the outcome of the workshop with the outcomes of a proctoring software pilot by examiners. We found that the two approaches led to complementary implementation criteria. The ethical session was less focused on making things work and more on transparency about conditions, processes, and rights. The ethical session also concentrated more on the values of all involved rather than on fraud detection effectiveness
MULTIFILE
Objectives: Animals with induced tinnitus showed difficulties in detecting silent gaps in sounds, suggesting that the tinnitus percept may be filling the gap. The main purpose of this study was to evaluate the applicability of this approach to detect tinnitus in human patients. The authors first hypothesized that gap detection would be impaired in patients with tinnitus, and second, that gap detection would be more impaired at frequencies close to the tinnitus frequency of the patient. Design: Twenty-two adults with bilateral tinnitus, 20 age-matched and hearing loss–matched subjects without tinnitus, and 10 young normal-hearing subjects participated in the study. To determine the characteristics of the tinnitus, subjects matched an external sound to their perceived tinnitus in pitch and loudness. To determine the minimum detectable gap, the gap threshold, an adaptive psychoacoustic test was performed three times by each subject. In this gap detection test, four different stimuli, with various frequencies and bandwidths, were presented at three intensity levels each. Results: Similar to previous reports of gap detection, increasing sensation level yielded shorter gap thresholds for all stimuli in all groups. Interestingly, the tinnitus group did not display elevated gap thresholds in any of the four stimuli. Moreover, visual inspection of the data revealed no relation between gap detection performance and perceived tinnitus pitch. Conclusions: These findings show that tinnitus in humans has no effect on the ability to detect gaps in auditory stimuli. Thus, the testing procedure in its present form is not suitable for clinical detection of tinnitus in humans.
LINK
The paper introduced an automatic score detection model using object detection techniques. The performance of sevenmodels belonging to two different architectural setups was compared. Models like YOLOv8n, YOLOv8s, YOLOv8m, RetinaNet-50, and RetinaNet-101 are single-shot detectors, while Faster RCNN-50 and Faster RCNN-101 belong to the two-shot detectors category. The dataset was manually captured from the shooting range and expanded by generating more versatile data using Python code. Before the dataset was trained to develop models, it was resized (640x640) and augmented using Roboflow API. The trained models were then assessed on the test dataset, and their performance was compared using matrices like mAP50, mAP50-90, precision, and recall. The results showed that YOLOv8 models can detect multiple objects with good confidence scores.
DOCUMENT
