In this paper, we propose a method for automatically detecting various types of snore sounds using image classification convolutional neural network (CNN) descriptors extracted from audio file spectrograms. The descriptors, denoted as deep spectrum features, are derived from forwarding spectrograms through very deep task-independent pre-trained CNNs. Specifically, activations of fully connected layers from two common image classification CNNs, AlexNet and VGG19, are used as feature vectors. Moreover, we investigate the impact of differing spectrogram colour maps and two CNN architectures on the performance of the system. Results presented indicate that deep spectrum features extracted from the activations of the second fully connected layer of AlexNet using a viridis colour map are well suited to the task. This feature space, when combined with a support vector classifier, outperforms the more conventional knowledge-based features of 6 373 acoustic functionals used in the INTERSPEECH ComParE 2017 Snoring sub-challenge baseline system. In comparison to the baseline, unweighted average recall is increased from 40.6 % to 44.8 % on the development partition, and from 58.5 % to 67.0 % on the test partition.
Technological developments along with the emergence of Industry 4.0 allow for new approaches to solve industrial problems, such as the Job-shop Scheduling Problem (JSP). In this sense, embedding Multi-Agent Systems (MAS) into Cyber-Physical Systems (CPS) is a highly promising approach to handle complex and dynamic JSPs. This paper proposes a data exchange framework in order to deal with the JSP considering the state-of-the-art technology regarding MAS, CPS and industrial standards. The proposed framework has self-configuring features to deal with disturbances in the production line. This is possible through the development of an intelligent system based on the use of agents and the Internet of Things (IoT) to achieve real-time data exchange and decision making in the job-shop. The performance of the proposed framework is tested in a simulation study based on a real industrial case. The results substantiate gains in flexibility, scalability and efficiency through the data exchange between factory layers. Finally, the paper presents insights regarding industrial applications in the Industry 4.0 era in general and in particular with regard to the framework implementation in the analyzed industrial case.
Today's assembly systems have to be flexible to adapt quickly to an increasing number and variety of products and changing demand volume. To manage these dynamics, flexible, reconfigurable, and autonomous assembly systems were proposed and partly realised in the last two decades. The flexibility and adaptability is realised by clustering the assembly system into subsystems and modules which get a certain degree of autonomy and control themselves in a decentralised way. This keynote paper will present the general principles of autonomy and the proposed concepts, methods and technologies to realise autonomous processes in assembly systems. Different approaches for design and autonomous operation of assembly will be explained and future trends towards fully autonomous components of an assembly system as well as autonomous parts and products will be discussed.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.