Intelligent Wearable Interfaces

Xu, Yangsheng; Li, Wen J.; Lee, Ka Keung

doi:10.1002/9780470222867

Cited by 29 publications

(17 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Applications that can specifically benefit from ASC include the design of context-aware services [45], intelligent wearable devices [52], robotics navigation systems [11] and audio archive management [32].…”

Section: Introductionmentioning

confidence: 99%

Acoustic Scene Classification: Classifying environments from the sounds they produce

Barchiesi

Giannoulis

Stowell

et al. 2015

IEEE Signal Process. Mag.

351

198

View full text Add to dashboard Cite

In this article we present an account of the state-of-the-art in acoustic scene classification (ASC), the task of classifying environments from the sounds they produce. Starting from a historical review of previous research in this area, we define a general framework for ASC and present different implementations of its components. We then describe a range of different algorithms submitted for a data challenge that was held to provide a general and fair benchmark for ASC techniques. The dataset recorded for this purpose is presented, along with the performance metrics that are used to evaluate the algorithms and statistical significance tests to compare the submitted methods. We use a baseline method that employes MFCCS, GMMS and a maximum likelihood criterion as a benchmark, and only find sufficient evidence to conclude that three algorithms significantly outperform it. We also evaluate the human classification accuracy in performing a similar classification task. The best performing algorithm achieves a mean accuracy that matches the median accuracy obtained by humans, and common pairs of classes are misclassified by both computers and humans. However, all acoustic scenes are correctly classified by at least some individuals, while there are scenes that are misclassified by all algorithms.

show abstract

Section: Introductionmentioning

confidence: 99%

Acoustic Scene Classification: Classifying environments from the sounds they produce

Barchiesi

Giannoulis

Stowell

et al. 2015

IEEE Signal Process. Mag.

351

198

View full text Add to dashboard Cite

show abstract

“…We trained the final scene classification systems using one-vs-one support vector machines (SVM) with χ 2 kernel. For LTE+, we used nonlinear SVMs with the extended Gaussian kernel given in (6). The hyperparameters of the SVMs were tuned via 10-fold cross-validation.…”

Section: Recognition Using Lte Representationsmentioning

confidence: 99%

“…Acoustic scene classification (ASC) is an important problem of computational auditory scene analysis [2,3]. Solving this problem will allow a device to recognize a surrounding environment via the sound it captures, and hence, enables a wide range of applications, such as surveillance [4], robotic navigation [5], and context-aware services [6,7]. However, due to its complex sound composition, it is challenging to obtain a good representation for recognition.…”

Section: Introductionmentioning

confidence: 99%

CNN-LTE: A class of 1-X pooling convolutional neural networks on label tree embeddings for audio scene classification

Phan

Koch

Hertel

et al. 2017

2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

We describe in this report our audio scene recognition system submitted to the DCASE 2016 challenge [1]. Firstly, given the label set of the scenes, a label tree is automatically constructed. This category taxonomy is then used in the feature extraction step in which an audio scene instance is represented by a label tree embedding image. Different convolutional neural networks, which are tailored for the task at hand, are finally learned on top of the image features for scene recognition. Our system reaches an overall recognition accuracy of 81.2% and 83.3% and outperforms the DCASE 2016 baseline with absolute improvements of 8.7% and 6.1% on the development and test data, respectively.

show abstract

“…Shi et.al. [5], [6] developed an inflated airbag system which can be activated when the system detect the wearer in falling based on IMUS. In these ways, people have to wear the special waist belt, and the data processor.…”

Section: Introductionmentioning

confidence: 99%

A Real-time intelligent shoe system for fall detection

Tao

Qian

Chen

et al. 2011

2011 IEEE International Conference on Robotics and Biomimetics

Self Cite

View full text Add to dashboard Cite

Injury duo to falling has accounted for a significant portion of accident. In order to provide prompt firstaid service to the victims, automatic and wearable devices are necessary to report the accident as soon as it occurs. This paper presents an intelligent shoe system which can not only detect the fall, but also classify the fall direction, especially the serious backward fall. In the prototype, eight pairs of force sensing resistors (FSRs) acquire the forces in different location of the insole. To reduce the computational cost and power consumption, and enhance the real-time performance, we propose an approach to reduce the sensor number based on principle component analysis (PCA), and lighten the system into a four-pair version. By means of artificial neural network (ANN), we classify the system input into three observations, and develop a finite state machine to trigger correct alarm and prevent false alarm by other complex human actions. To overcome the problem in shortage of learning data to detect the falling direction, nearest neighbor approach is utilized, which learns the necessary pattern from abundant tilted standing data. The experiment validates system and approaches to detect fall and fall direction.

show abstract

Intelligent Wearable Interfaces

Cited by 29 publications

References 0 publications

Acoustic Scene Classification: Classifying environments from the sounds they produce

Acoustic Scene Classification: Classifying environments from the sounds they produce

CNN-LTE: A class of 1-X pooling convolutional neural networks on label tree embeddings for audio scene classification

A Real-time intelligent shoe system for fall detection

Contact Info

Product

Resources

About