Physical activity classification has grown in importance lately, for reasons such as positioning or health issues. Given the ubiquity of smartphones and the plethora of sensors they contain, these devices have become an extremely useful tool for this task. In that direction, this project provides an algorithm to count steps using the accelerometer of an Android smartphone. This algorithm searches for patterns in the accelerometer’s signal based on the correlation between consecutive fragments of the signal after a pre-processing step that adapts the data, to count steps under relatively unconstrained ways of carrying the smartphone. The accuracy of the designed algorithm is 92.5% using a database of eleven subjects and four different tests ...
AbstractNowadays, smartphones play an ubiquitous role in accessing and processing information, most ...
Background Traditional activity recognition solutions are not widely applicable due to a high cost a...
Traditional activity recognition solutions are not widely applicable due to a high cost and inconven...
Physical activity classification has grown in importance lately, for reasons such as positioning or ...
Programming in Android and using the accelerometer we implement a movements' classifier.[ANGLÈS] Hum...
The accuracy of physical activity intensity (PAI) quantifiers that rely solely on smartphone acceler...
Programming in Android and using the accelerometer we implement a movements' classifier.[ANGLÈS] Hum...
AbstractNowadays, smartphones play an ubiquitous role in accessing and processing information, most ...
In healthcare sector, getting to track the activity patterns of an individual is vital when providin...
This paper proposes a reliable step counting system for pedestrians based on pattern recognition, mo...
Physical activity recognition of everyday activities such as sitting, standing, laying, walking, and...
The main aim of the project is to develop an algorithm which will classify the activity performed by...
The focus of this thesis is the detection of walking and running, followed by a step count. Testing ...
Background Traditional activity recognition solutions are not widely applicable due to a high cost a...
Traditional activity recognition solutions are not widely applicable due to a high cost and inconven...
AbstractNowadays, smartphones play an ubiquitous role in accessing and processing information, most ...
Background Traditional activity recognition solutions are not widely applicable due to a high cost a...
Traditional activity recognition solutions are not widely applicable due to a high cost and inconven...
Physical activity classification has grown in importance lately, for reasons such as positioning or ...
Programming in Android and using the accelerometer we implement a movements' classifier.[ANGLÈS] Hum...
The accuracy of physical activity intensity (PAI) quantifiers that rely solely on smartphone acceler...
Programming in Android and using the accelerometer we implement a movements' classifier.[ANGLÈS] Hum...
AbstractNowadays, smartphones play an ubiquitous role in accessing and processing information, most ...
In healthcare sector, getting to track the activity patterns of an individual is vital when providin...
This paper proposes a reliable step counting system for pedestrians based on pattern recognition, mo...
Physical activity recognition of everyday activities such as sitting, standing, laying, walking, and...
The main aim of the project is to develop an algorithm which will classify the activity performed by...
The focus of this thesis is the detection of walking and running, followed by a step count. Testing ...
Background Traditional activity recognition solutions are not widely applicable due to a high cost a...
Traditional activity recognition solutions are not widely applicable due to a high cost and inconven...
AbstractNowadays, smartphones play an ubiquitous role in accessing and processing information, most ...
Background Traditional activity recognition solutions are not widely applicable due to a high cost a...
Traditional activity recognition solutions are not widely applicable due to a high cost and inconven...