No abstract

The coding of observational data is commonly used to analyse and evaluate human behaviours. The technique can help researchers inform the design and impact of, for example, an Ubicomp system by studying specific behaviours of interest. There are some tools that can alleviate the burden of observational coding, like those that help to collect and organise data, but can still be error-prone and time-consuming. Moreover, most of these tools lack automation, requiring intense human interaction. In order to mitigate these issues, computer vision (CV) and machine learning (ML) techniques could be used to automate observational coding, but little work has focused on analysing the feasibility of such an approach, with the goal of reducing the total coding time while maintaining accuracy. In this work, we address this question by proposing an automated approach for a real-world case study and compare it to manual coding. The study is composed of 10 videos with an average duration of 17 min each, where the goal is to determine the attention of children with autism that participate in a neurofeedback therapy session. Each video was hand-coded by three human observers to define the ground truth and to measure the manual coding time. Results show that it is feasible to automate the coding of observational behaviours and obtain a noticeable reduction in coding time, but with a slight loss in accuracy. Moreover, we illustrate that the best solution would be a hybrid approach, using a semi-automated system that combines human expertise and

This paper presents the proposal of a method to recognize emotional states through EEG analysis. The novelty of this work lies in its feature improvement strategy, based on multiclass genetic programming with multidimensional populations (M3GP), which builds features by implementing an evolutionary technique that selects, combines, deletes, and constructs the most suitable features to ease the classification process of the learning method. In this way, the problem data can be mapped into a more favorable search space that best defines each class. After implementing the M3GP, the results showed an increment of 14.76% in the recognition rate without changing any settings in the learning method. The tests were performed on a biometric EEG dataset (BED), designed to evoke emotions and record the cerebral cortex’s electrical response; this dataset implements a low cost device to collect the EEG signals, allowing greater viability for the application of the results. The proposed methodology achieves a mean classification rate of 92.1%, and simplifies the feature management process by increasing the separability of the spectral features.

The detection and description of locally salient regions is one of the most widely used low-level processes in modern computer vision systems. The general approach relies on the detection of stable and invariant image features that can be uniquely characterized using compact descriptors. Many detection and description algorithms have been proposed, most of them derived using different assumptions or problem models. This work presents a comparison of different approaches towards the feature extraction problem, namely: (1) standard computer vision techniques; (2) automatic synthesis techniques based on genetic programming (GP); and (3) a new local descriptor based on composite correlation filtering, proposed for the first time in this paper. The considered methods are evaluated on a difficult real-world problem, vision-based simultaneous localization and mapping (SLAM). Using three experimental scenarios, results indicate that the GP-based methods and the correlation filtering techniques outperform widely used computer vision algorithms such as the Harris and Shi-Tomasi detectors and the Speeded Up Robust Features descriptor.