H
Hakob Sarukhanyan
Researcher at Armenian National Academy of Sciences
Publications - 30
Citations - 172
Hakob Sarukhanyan is an academic researcher from Armenian National Academy of Sciences. The author has contributed to research in topics: Hadamard transform & Mobile device. The author has an hindex of 6, co-authored 30 publications receiving 158 citations.
Papers
More filters
Activity recognition using k-nearest neighbor algorithm on smartphone with tri-axial accelerometer
Sahak Kaghyan,Hakob Sarukhanyan +1 more
TL;DR: This paper is devoted to one approach that solves human activity classification problem with help of a mobile device carried by user and is based on K-Nearest Neighbor algorithm (K-NN).
Proceedings ArticleDOI
Automatic detection and concealment of specular reflections for endoscopic images
TL;DR: A method for segmentation of highlights based on adoptive colour thresholding and contour analyses and a powerful inpainting method, which fill the highlighted regions with information propagated from adjacent areas are proposed.
Proceedings ArticleDOI
Accelerometer and GPS sensor combination based system for human activity recognition
Sahak Kaghyan,Hakob Sarukhanyan +1 more
TL;DR: An approach which allows recognizing activity, performed by human, using smartphone acceleration and positioning sensors, that retrieves signal data and stores it SQLite portable mobile database and uses asynchronous model of signal retrieving and storing procedures.
Proceedings ArticleDOI
Human Movement Activity Classification Approaches that use Wearable Sensors and Mobile Devices
TL;DR: This paper reviews different approaches of human activity recognition and finds that healthcare applications exploiting build-in sensors are very promising.
Proceedings ArticleDOI
Short-term memory with read-only unit in neural image caption generator
TL;DR: A model that can automatically generate an image description and is based on a recurrent neural network with modified LSTM cell with an additional gate responsible for image features is presented, which results in generation of more accurate captions.