In this paper the optical linear sensor, a representative of low resolution sensors, was investigated in the multi-class recognition of near field hand gestures. The Recurrent Neural Network (RNN) with a GRU (Gated Recurrent Unit) memory cell was utilized as a gestures classifier. A set of 27 gestures was collected from a group of volunteers. The 27 000 sequences obtained were divided into training, validation, and test subsets. The primary research goal was to define the most appropriate model architecture in terms of the accurate recognition of each gesture. An additional aim of the research was to investigate the kind of input data, i.e., raw data, or preprocessed (feature) data, which generally produces better results. Therefore, three datasets were generated: raw data, simple features data, and high level features data (this includes information about hand poses which are already recognized). The random search method was applied to achieve hyperparameter optimization to find the best possible topology for the neural network. The analysis performed shows that selected models were characterized by a test score at a level of 96.89% for the raw data, 95.75% for simple features, and 93.38% for high level features. Results indicate that the direct use of raw data obtained from the optical linear sensor evaluated on the RNN with GRU memory cells allows for the reliable recognition of even complex gestures. Therefore, such solutions may have the potential to serve as a support to, or as an alternative to video based sensors especially for mobile devices.

Recently, capabilities of many computer vision tasks have significantly improved due to advances in Convolutional Neural Networks. In our research, we demonstrate that it can be also used for face detection from low resolution thermal images, acquired with a portable camera. The physical size of the camera used in our research allows for embedding it in a wearable device or indoor remote monitoring solution for elderly and disabled people. The benefits of the proposed architecture were experimentally verified on the thermal video sequences, acquired in various scenarios to address possible limitations of remote diagnostics: movements of the person performing a diagnose and movements of the examined person. The achieved short processing time (42.05±0.21ms) along with high model accuracy (false positives -0.43%; true positives for the patient focused on a certain task -89.2%) clearly indicates that the current state of the art in the area of image classification and face tracking in thermography was significantly outperformed.

Estimation of vital signs using image processing techniques have already been proved to have a potential for supporting remote medical diagnostics and replacing traditional measurements that usually require special hardware and electrodes placed on a body. In this paper, we further extend studies on contactless Respiratory Rate (RR) estimation from extremely low resolution thermal imagery by enhancing acquired sequences using Deep Neural Networks (DNN). To perform extensive benchmark evaluation, we acquired two thermal datasets using FLIR^® cameras with a spatial resolution of 80 × 60 and 320 × 240 from 71 volunteers in total. In-depth analysis of the proposed Convolutional-based Super Resolution model showed that for images downscaled with a factor of 2 and then super-resolved using Deep Learning (DL) can lead to better RR estimation accuracy than from original high-resolution sequences. In addition, if an estimator based on a dominating peak in the frequency domain is used, SR can outperform original data for a down-scale factor of 4 and images as small as 20 × 15 pixels. Our study also showed that RR estimation accuracy is better for super-resolved data than for images with color changes magnified using algorithms previously applied in the literature for enhancing vital signs patterns.

With the technology advancements in smart home sector, voice control and automation are key components that can make a real difference in people's lives. The voice recognition technology market continues to involve rapidly as almost all smart home devices are providing speaker recognition capability today. However, most of them provide cloud-based solutions or use very deep Neural Networks for speaker recognition task, which are not suitable models to run on smart home devices. In this paper, we compare relatively small Convolutional Neural Networks (CNN) and evaluate effectiveness of speaker recognition using these models on edge devices. In addition, we also apply transfer learning technique to deal with a problem of limited training data. By developing solution suitable for running inference locally on edge devices, we eliminate the well-known cloud computing issues, such as data privacy and network latency, etc. The preliminary results proved that the chosen model adapts the benefit of computer vision task by using CNN and spectrograms to perform speaker classification with precision and recall ~84 % in time less than 60 ms on mobile device with Atom Cherry Trail processor.

Imaging photoplethysmography has already been proved to be successful in short distance (below 1m). However, most of the real-life use cases of measuring vital signs require the system to work at longer distances, to be both more reliable and convenient for the user. The possible scenarios that system designers must have in mind include monitoring of the vital signs of residents in nursing homes, disabled people, who can't move, constant support for people regardless of the performed activity (e.g. during sleeping), infants, etc. In this work we verified the possibility of remote pulse estimation at a distance above 5m. Additionally, we integrated the deep learning algorithm for person tracking and identification, even when facial features are not visible. In this way, we enabled the collection of user specific measurements to create personalized vital signs patterns and we provided the support for monitoring of multiple people using one video stream. The preliminary results showed that it is possible to accurately (RMSE <; 2.8 beats per minute) extract pulse from visible light sequences acquired with a webcam at a distance of 6m after applying a proper image pre-processing algorithm.