Real time unsupervised learning of visual stimuli in neuromorphic VLSI systems

Recent years have seen significant growth in the adoption of smart home devices. It involves a Smart Home System for better visualisation and analysis with time series. However, there are a few challenges faced by the system developers, such as data quality or data anomaly issues. These anomalies can be due to technical or non-technical faults. It is essential to detect the non-technical fault as it might incur economic cost. In this study, the main objective is to overcome the challenge of training learning models in the case of an unlabelled dataset. Another important consideration is to train the model to be able to discriminate abnormal consumption from seasonal-based consumption. This paper proposes a system using unsupervised learning for Time-Series data in the smart home environment. Initially, the model collected data from the real-time scenario. Following seasonal-based features are generated from the time-domain, followed by feature reduction technique PCA to 2-dimension data. This data then passed through four known unsupervised learning models and was evaluated using the Excess Mass and Mass-Volume method. The results concluded that LOF tends to outperform in the case of detecting anomalies in electricity consumption. The proposed model was further evaluated by benchmark anomaly dataset, and it was also proved that the system could work with the different fields containing time-series data. The model will cluster data into anomalies and not. The developed anomaly detector will detect all anomalies as soon as possible, triggering real alarms in real-time for time-series data's energy consumption. It has the capability to adapt to changing values automatically. Doi: 10.28991/esj-2021-01314 Full Text: PDF

Download Full-text

Real-Time Air Quality Feedback in Children’s Bedrooms Increases Compliance with Indoor Air Guidelines

10.31234/osf.io/48zqe ◽

2021 ◽

Author(s):

Vincent Berardi

Keyword(s):

Air Quality ◽

Real Time ◽

Indoor Air ◽

Visual Stimuli ◽

Quality Feedback ◽

Single Instance

Air particle monitors were placed in the children's bedrooms in 298 homes. We examined whether outfitting these monitors with immediate auditory and visual stimuli plus weekly coaching increased the probability of establishing a smoke-free home, operationalized by at least single instance of 30 consecutive days below the WHO 25 ug/m3 guideline.

Download Full-text

Microprocessor-controlled vector scan display system for generation of real-time visual stimuli

Medical & Biological Engineering & Computing ◽

10.1007/bf02441481 ◽

1989 ◽

Vol 27 (3) ◽

pp. 246-253 ◽

Cited By ~ 4

Author(s):

F. H. Schuling ◽

B. Vorenkamp ◽

W. H. Zaagman

Keyword(s):

Real Time ◽

Visual Stimuli ◽

Display System

Download Full-text

Real-Time Classification of Complex Patterns Using Spike-Based Learning in Neuromorphic VLSI

IEEE Transactions on Biomedical Circuits and Systems ◽

10.1109/tbcas.2008.2005781 ◽

2009 ◽

Vol 3 (1) ◽

pp. 32-42 ◽

Cited By ~ 135

Author(s):

Srinjoy Mitra ◽

Stefano Fusi ◽

Giacomo Indiveri

Keyword(s):

Real Time ◽

Complex Patterns ◽

Real Time Classification ◽

Neuromorphic Vlsi

Download Full-text

Using Eye-Tracking and Support Vector Machine to Measure Learning Attention in eLearning

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.311.9 ◽

2013 ◽

Vol 311 ◽

pp. 9-14 ◽

Cited By ~ 4

Author(s):

Chien Hung Liu ◽

Po Yin Chang ◽

Chun Yuan Huang

Keyword(s):

Support Vector Machine ◽

Learning Environment ◽

Unsupervised Learning ◽

Real Time ◽

Feedback System ◽

Integrated Approach ◽

Support Vector ◽

Region Detection ◽

Volumetric Features ◽

Finite State

For eLearning, how to naturally measure the learning attention of students with lower cost devices in an unsupervised learning environment is a crucial issue. Students often far away and out of teachers’ control in above situation which may cause students do not have strong learning motivation and might feel fatigued and inattentive for learning. A real-time and naturally learning attention measure approach can support instructor to better control the learning attention of students in unsupervised learning environment. This paper proposes an integrated approach, named Real-time Learning Attention Feedback System (RLAFS) which could naturally measure learning attention in unsupervised learning environments. The system architecture of RLAFS consists with three layers: first layer is Image preprocessing layer, which is responsible for image processing and motion detection. Second is eyebrow region detection layer, which is focus on the features of face and eyes capturing and positioning. Classifier layer is the third layer, in which integral image, volumetric features and finite-state-machine are used to capture the current state of learning attention of students. Consequently, support vector machine is utilized to classify the level of learning attention. The experiments are conducted in an unsupervised environment, and results showed RLAFS is a promising approach which can naturally measure learning attention and has a significant impact on learning efficient.

Download Full-text