A Highly Effective Deep Learning Based Escape Route Recognition Module for Autonomous Robots in Crisis and Emergency Situations

Buettner, Ricardo; Baumgartl, Hermann

doi:10.24251/hicss.2019.081

Cited by 37 publications

(21 citation statements)

References 41 publications

(85 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While we intensively evaluated other traditional machine learning approaches such as clustering [46] and also most modern convolutional neural networks, which are outstanding in other domains such as image recognition [47][48][49], we achieved the best results here with our novel tree-based method proposed in [34]. However, the method of choice always limits scientific understanding.…”

Section: Limitationmentioning

confidence: 97%

Development of a Machine Learning Based Algorithm To Accurately Detect Schizophrenia based on One-minute EEG Recordings

Buettner¹,

Beil²,

Scholtz³

et al. 2020

Proceedings of the Annual Hawaii International Conference on System Sciences

Self Cite

View full text Add to dashboard Cite

While diagnosing schizophrenia by physicians based on patients' history and their overall mental health is inaccurate, we report on promising results using a novel, fast and reliable machine learning approach based on electroencephalography (EEG) recordings. We show that a fine granular division of EEG spectra in combination with the Random Forest classifier allows a distinction to be made between paranoid schizophrenic (ICD-10 F20.0) and nonschizophrenic persons with a very good balanced accuracy of 96.77 percent. We evaluate our approach on EEG data from an open neurological and psychiatric repository containing 499 one-minute recordings of n=28 participants (14 paranoid schizophrenic and 14 healthy controls). Since the fact that neither diagnostic tests nor biomarkers are available yet to diagnose paranoid schizophrenia, our approach paves the way to a quick and reliable diagnosis with a high accuracy. Furthermore, interesting insights about the most predictive subbands were gained by analyzing the electroencephalographic spectrum up to 100 Hz.

show abstract

Section: Limitationmentioning

confidence: 97%

Development of a Machine Learning Based Algorithm To Accurately Detect Schizophrenia based on One-minute EEG Recordings

Buettner¹,

Beil²,

Scholtz³

et al. 2020

Proceedings of the Annual Hawaii International Conference on System Sciences

Self Cite

View full text Add to dashboard Cite

show abstract

“…In recent years, convolutional neural networks have become the prime algorithm for solving many complex computer vision problems [38][39][40]. In the field of image recognition, the convolutional neural networks are among the latest deeplearning methods.…”

Section: Convolutional Neural Networkmentioning

confidence: 99%

“…In the field of image recognition, the convolutional neural networks are among the latest deeplearning methods. Convolutional neural networks are based on the multi-layered structure of real brain structures of the visual cortex and have shown remarkable results in many highly complex application scenarios [40][41][42][43].…”

Section: Convolutional Neural Networkmentioning

confidence: 99%

“…Our network architecture consists of three blocks of convolutional and batch-normalization layers. Inspired by architectures in [40,41,55], depthwise-separable convolutions were used in blocks 2 and 3, effectively reducing the amount of trainable parameters. The convolutional blocks are followed by a global average pooling layer and a fully connected softmax layer for the final classification.…”

Section: Our Network Architecturementioning

confidence: 99%

“…Heatmap for one example of a delamination defect: a the original thermographic image; b the class activation heatmap of our network where important spatial areas are highlighted, with the red circles indicating the area of high importance in the Grad-CAM heatmap and provide high accuracy values, while being light on computational resources[40].…”

mentioning

confidence: 99%

See 2 more Smart Citations

A deep learning-based model for defect detection in laser-powder bed fusion using in-situ thermographic monitoring

et al. 2020

Self Cite

View full text Add to dashboard Cite

Additive manufacturing of metal components with laser-powder bed fusion is a very complex process, since powder has to be melted and cooled in each layer to produce a part. Many parameters influence the printing process; however, defects resulting from suboptimal parameter settings are usually detected after the process. To detect these defects during the printing, different process monitoring techniques such as melt pool monitoring or off-axis infrared monitoring have been proposed. In this work, we used a combination of thermographic off-axis imaging as data source and deep learning-based neural network architectures, to detect printing defects. For the network training, a k-fold cross validation and a hold-out cross validation were used. With these techniques, defects such as delamination and splatter can be recognized with an accuracy of 96.80%. In addition, the model was evaluated with computing class activation heatmaps. The architecture is very small and has low computing costs, which means that it is suitable to operate in real time even on less powerful hardware.

show abstract

Modeling belief‐desire‐intention reasoning agents for situation‐aware formalisms

Haque

Saleem

Khan

2021

Concurrency and Computation

View full text Add to dashboard Cite

A natural disaster is an inevitable situation that can occur at any time and anywhere along with varied forms such as earthquakes, floods, hurricane, wildfire, and so forth, and different level of occurrences has been recorded from mild to an intense level. Timely disaster response plays an important role in reducing its deteriorate after-effects and can save countless lives. Over the years, people have been developing guidelines and processes to cope up with such kinds of hazardous situations. In recent years, with the advent of the pervasive computing paradigm, situation-awareness has been considered to be the most fascinating approach for situation assessment and provides decision support accordingly. Situation-aware systems observe/perceive dynamic changes in the environment, understand/comprehend the situation, and perform actions according to the environment. Although state-of-the-art formalisms have been developed to handle such kinds of hazardous situations intelligently and rescue the victims. However, belief-desire-intention (BDI) reasoning mechanism with the incorporation of situation-awareness is still the thirsty area of research to manage hazardous situations. In this article, we propose a temporal epistemic situation-aware formalism for BDI reasoning agents to model the context-aware decision support system.To demonstrate the work of the proposed formalism, we develop a case study based on a disaster situation, in which BDI agents are modeled and simulated to present the results in terms of agents' reasoning processes. We demonstrate the scalability in temporal aspects of the system using different levels of disasters to monitor the hazardous situations and evaluate the overall behavior of the system.

show abstract

A Highly Effective Deep Learning Based Escape Route Recognition Module for Autonomous Robots in Crisis and Emergency Situations

Cited by 37 publications

References 41 publications

Development of a Machine Learning Based Algorithm To Accurately Detect Schizophrenia based on One-minute EEG Recordings

Development of a Machine Learning Based Algorithm To Accurately Detect Schizophrenia based on One-minute EEG Recordings

A deep learning-based model for defect detection in laser-powder bed fusion using in-situ thermographic monitoring

Modeling belief‐desire‐intention reasoning agents for situation‐aware formalisms

Contact Info

Product

Resources

About