Real time event-based segmentation to classify locomotion activities through a single inertial sensor

Fida, Benish; Bibbo, Daniele; Bernabucci, Ivan; Proto, Antonino; Conforto, Silvia; Schmid, Maurizio

doi:10.4108/eai.14-10-2015.2261695

Cited by 14 publications

(6 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…IMU-based Activity Recognition. This dataset is collected by 9 participants [5], which contains 1200000 samples. 8 ADLs are selected as a subset of our paper.…”

Section: Data Settingmentioning

confidence: 99%

Deep Neural Network Hyperparameter Optimization with Orthogonal Array Tuning

Zhang

Chen

Yao

et al. 2019

Communications in Computer and Information Science

View full text Add to dashboard Cite

Deep learning algorithms have achieved excellent performance lately in a wide range of fields (e.g., computer version). However, a severe challenge faced by deep learning is the high dependency on hyperparameters. The algorithm results may fluctuate dramatically under the different configuration of hyper-parameters. Addressing the above issue, this paper presents an efficient Orthogonal Array Tuning Method (OATM) for deep learning hyper-parameter tuning. We describe the OATM approach in five detailed steps and elaborate on it using two widely used deep neural network structures (Recurrent Neural Networks and Convolutional Neural Networks). The proposed method is compared to the state-of-the-art hyper-parameter tuning methods including manually (e.g., grid search and random search) and automatically (e.g., Bayesian Optimization) ones. The experiment results state that OATM can significantly save the tuning time compared to the state-of-the-art methods while preserving the satisfying performance.

show abstract

“…IMU-based Activity Recognition. This dataset is collected by 9 participants [5], which contains 1200000 samples. 8 ADLs are selected as a subset of our paper.…”

Section: Data Settingmentioning

confidence: 99%

Deep Neural Network Hyperparameter Optimization with Orthogonal Array Tuning

Zhang

Chen

Yao

et al. 2019

Communications in Computer and Information Science

View full text Add to dashboard Cite

show abstract

“…Some of the proposed solutions relied on the use of wearable sensors [22,23,24,25], while others exploited non-invasive approaches in which the sensors were applied on the environment (e.g., chairs [26,27,28,29,30,31,32,33]). In this way, participants were not aware of being monitored, thus allowing the reproduction of real-life conditions.…”

Section: Introductionmentioning

confidence: 99%

A Sitting Posture Monitoring Instrument to Assess Different Levels of Cognitive Engagement

Bibbo

Carli

Conforto

et al. 2019

Sensors

Self Cite

View full text Add to dashboard Cite

An office chair for analyzing the seated posture variation during the performance of a stress-level test is presented in this work. To meet this aim, we placed a set of textile pressure sensors both on the backrest and on the seat of the chair. The position of the sensors was selected for maximizing the detection of variations of user’s posture. The effectiveness of the designed system was evaluated through an experiment where increasing stress levels were obtained by administering a Stroop test. The collected results had been analyzed by considering three different time intervals based on the difficulty level of the test (low, medium, and high). A transition analysis conducted on postures assumed during the test showed that participants reached a different posture at the end of the test, when the cognitive engagement increased, with respect to the beginning. This evidence highlighted the presence of movement presumably due to the increased cognitive engagement. Overall, the performed analysis showed the proposed monitoring system could be used to identify body posture variations related to different levels of engagement of a seated user while performing cognitive tasks.

show abstract

“…Thus, Shenck and Paradiso [ 2 ] developed an energy harvesting system, mounted on the shoes, that enables one to power a wide range of body-worn devices. Then, Gonzalez et al [ 3 ] and Niu et al [ 4 ] analyzed the feasibility of using the energy harvested from the human body to power wearable sensors, which include the functions of data processing and wireless communication [ 5 , 6 ].…”

Section: Introductionmentioning

confidence: 99%

Measurements of Generated Energy/Electrical Quantities from Locomotion Activities Using Piezoelectric Wearable Sensors for Body Motion Energy Harvesting

Proto

Penhaker

Bibbo

et al. 2016

Sensors

Self Cite

View full text Add to dashboard Cite

In this paper, two different piezoelectric transducers—a ceramic piezoelectric, lead zirconate titanate (PZT), and a polymeric piezoelectric, polyvinylidene fluoride (PVDF)—were compared in terms of energy that could be harvested during locomotion activities. The transducers were placed into a tight suit in proximity of the main body joints. Initial testing was performed by placing the transducers on the neck, shoulder, elbow, wrist, hip, knee and ankle; then, five locomotion activities—walking, walking up and down stairs, jogging and running—were chosen for the tests. The values of the power output measured during the five activities were in the range 6 µW–74 µW using both transducers for each joint.

show abstract

Real time event-based segmentation to classify locomotion activities through a single inertial sensor

Cited by 14 publications

References 23 publications

Deep Neural Network Hyperparameter Optimization with Orthogonal Array Tuning

Deep Neural Network Hyperparameter Optimization with Orthogonal Array Tuning

A Sitting Posture Monitoring Instrument to Assess Different Levels of Cognitive Engagement

Measurements of Generated Energy/Electrical Quantities from Locomotion Activities Using Piezoelectric Wearable Sensors for Body Motion Energy Harvesting

Contact Info

Product

Resources

About