A robust event-driven approach to always-on object recognition

<div> <div> <div> <p>We propose a neuromimetic architecture able to perform always-on pattern recognition. To achieve this, we extended an existing event-based algorithm [1], which introduced novel spatio-temporal features as a Hierarchy Of Time-Surfaces (HOTS). Built from asynchronous events acquired by a neuromorphic camera, these time surfaces allow to code the local dynamics of a visual scene and to create an efficient event-based pattern recognition architecture. Inspired by neuroscience, we extended this method to increase its performance. Our first contribution was to add a homeostatic gain control on the activity of neurons to improve the learning of spatio-temporal patterns [2]. A second contribution is to draw an analogy between the HOTS algorithm and Spiking Neural Networks (SNN). Following that analogy, our last contribution is to modify the classification layer and remodel the offline pattern categorization method previously used into an online and event-driven one. This classifier uses the spiking output of the network to define novel time surfaces and we then perform online classification with a neuromimetic implementation of a multinomial logistic regression. Not only do these improvements increase consistently the performances of the network, they also make this event-driven pattern recognition algorithm online and bio-realistic. Results were validated on different datasets: DVS barrel [3], Poker-DVS [4] and N-MNIST [5]. We foresee to develop the SNN version of the method and to extend this fully event-driven approach to more naturalistic tasks, notably for always-on, ultra-fast object categorization. </p> </div> </div> </div>

A robust event-driven approach to always-on object recognition

<div> <div> <div> <p>We propose a neuromimetic architecture able to perform always-on pattern recognition. To achieve this, we extended an existing event-based algorithm [1], which introduced novel spatio-temporal features as a Hierarchy Of Time-Surfaces (HOTS). Built from asynchronous events acquired by a neuromorphic camera, these time surfaces allow to code the local dynamics of a visual scene and to create an efficient event-based pattern recognition architecture. Inspired by neuroscience, we extended this method to increase its performance. Our first contribution was to add a homeostatic gain control on the activity of neurons to improve the learning of spatio-temporal patterns [2]. A second contribution is to draw an analogy between the HOTS algorithm and Spiking Neural Networks (SNN). Following that analogy, our last contribution is to modify the classification layer and remodel the offline pattern categorization method previously used into an online and event-driven one. This classifier uses the spiking output of the network to define novel time surfaces and we then perform online classification with a neuromimetic implementation of a multinomial logistic regression. Not only do these improvements increase consistently the performances of the network, they also make this event-driven pattern recognition algorithm online and bio-realistic. Results were validated on different datasets: DVS barrel [3], Poker-DVS [4] and N-MNIST [5]. We foresee to develop the SNN version of the method and to extend this fully event-driven approach to more naturalistic tasks, notably for always-on, ultra-fast object categorization. </p> </div> </div> </div>

A deep neural network model for multi-view human activity recognition

Multiple cameras are used to resolve occlusion problem that often occur in single-view human activity recognition. Based on the success of learning representation with deep neural networks (DNNs), recent works have proposed DNNs models to estimate human activity from multi-view inputs. However, currently available datasets are inadequate in training DNNs model to obtain high accuracy rate. Against such an issue, this study presents a DNNs model, trained by employing transfer learning and shared-weight techniques, to classify human activity from multiple cameras. The model comprised pre-trained convolutional neural networks (CNNs), attention layers, long short-term memory networks with residual learning (LSTMRes), and Softmax layers. The experimental results suggested that the proposed model could achieve a promising performance on challenging MVHAR datasets: IXMAS (97.27%) and i3DPost (96.87%). A competitive recognition rate was also observed in online classification.

A Hybrid Online Classifier System for Internet Traffic Based on Statistical Machine Learning Approach and Flow Port Number

Internet traffic classification is a beneficial technique in the direction of intrusion detection and network monitoring. After several years of searching, there are still many open problems in Internet traffic classification. The hybrid classifier combines more than one classification method to identify Internet traffic. Using only one method to classify Internet traffic poses many risks. In addition, an online classifier is very important in order to manage threats on traffic such as denial of service, flooding attack and other similar threats. Therefore, this paper provides some information to differentiate between real and live internet traffic. In addition, this paper proposes a hybrid online classifier (HOC) system. HOC is based on two common classification methods, port-base and ML-base. HOC is able to perform an online classification since it can identify live Internet traffic at the same time as it is generated. HOC was used to classify three common Internet application classes, namely web, WhatsApp and Twitter. HOC produces more than 90% accuracy, which is higher than any individual classifiers.

Learning Geometry Information of Target for Visual Object Tracking with Siamese Networks

Recently, Siamese architecture has been widely used in the field of visual tracking, and has achieved great success. Most Siamese network based trackers aggregate the target information of two branches by cross-correlation. However, since the location of the sampling points in the search feature area is pre-fixed in cross-correlation operation, these trackers suffer from either background noise influence or missing foreground information. Moreover, the cross-correlation between the template and the search area neglects the geometry information of the target. In this paper, we propose a Siamese deformable cross-correlation network to model the geometric structure of target and improve the performance of visual tracking. We propose to learn an offset field end-to-end in cross-correlation. With the guidance of the offset field, the sampling in the search image area can adapt to the deformation of the target, and realize the modeling of the geometric structure of the target. We further propose an online classification sub-network to model the variation of target appearance and enhance the robustness of the tracker. Extensive experiments are conducted on four challenging benchmarks, including OTB2015, VOT2018, VOT2019 and UAV123. The results demonstrate that our tracker achieves state-of-the-art performance.

A Fast Online Classification Method of Solid Wood Floors Based on Stochastic Sampling and Machine Learning

Solid wood floors are widely used as an interior decoration material, and the color of solid wood surfaces plays a decisive role in the final decoration effect. Therefore, the color classification of solid wood floors is the final and most important step before laying. However, research on floor classification usually focuses on recognizing complex and diverse features but ignores execution speed, which causes common methods to not meet the requirements of online classification in practical production. In this paper, a new online classification method of solid wood floors was proposed by combining probability theory and machine learning. Firstly, a probability-based feature extraction method (stochastic sampling feature extractor) was developed to obtain rapid key features regardless of the disturbance of wood grain. The stochastic features were determined by a genetic algorithm. Then, an extreme learning machine—as a fast classification neural network—was selected and trained with the selected stochastic features to classify solid wood floors. Several experiments were carried out to evaluate the performance of the proposed method, and the results showed that the proposed method achieved a classification accuracy of 97.78% and less than 1 ms for each solid wood floor. The proposed method has advantages including a high execution speed, great accuracy, and flexible adaptability. Overall, it is suitable for online industry production.

Optimization of Task Allocation for Collaborative Brain–Computer Interface Based on Motor Imagery

ObjectiveCollaborative brain–computer interfaces (cBCIs) can make the BCI output more credible by jointly decoding concurrent brain signals from multiple collaborators. Current cBCI systems usually require all collaborators to execute the same mental tasks (common-work strategy). However, it is still unclear whether the system performance will be improved by assigning different tasks to collaborators (division-of-work strategy) while keeping the total tasks unchanged. Therefore, we studied a task allocation scheme of division-of-work and compared the corresponding classification accuracies with common-work strategy’s.ApproachThis study developed an electroencephalograph (EEG)-based cBCI which had six instructions related to six different motor imagery tasks (MI-cBCI), respectively. For the common-work strategy, all five subjects as a group had the same whole instruction set and they were required to conduct the same instruction at a time. For the division-of-work strategy, every subject’s instruction set was a subset of the whole one and different from each other. However, their union set was equal to the whole set. Based on the number of instructions in a subset, we divided the division-of-work strategy into four types, called “2 Tasks” … “5 Tasks.” To verify the effectiveness of these strategies, we employed EEG data collected from 19 subjects who independently performed six types of MI tasks to conduct the pseudo-online classification of MI-cBCI.Main resultsTaking the number of tasks performed by one collaborator as the horizontal axis (two to six), the classification accuracy curve of MI-cBCI was mountain-like. The curve reached its peak at “4 Tasks,” which means each subset contained four instructions. It outperformed the common-work strategy (“6 Tasks”) in classification accuracy (72.29 ± 4.43 vs. 58.53 ± 4.36%).SignificanceThe results demonstrate that our proposed task allocation strategy effectively enhanced the cBCI classification performance and reduced the individual workload.

Nuclear Fusion Pattern Recognition by Ensemble Learning

Nuclear fusion is the process by which two or more atomic nuclei join together to form a single heavier nucleus. This is usually accompanied by the release of large quantities of energy. This energy could be cheaper, cleaner, and safer than other technology currently in use. Experiments in nuclear fusion generate a large number of signals that are stored in huge databases. It is impossible to do a complete analysis of this data manually, and it is essential to automate this process. That is why machine learning models have been used to this end in previous years. In the literature, several popular algorithms can be found to carry out the automatic classification of signals. Among these, ensemble methods provide a good balance between success rate and internal information about models. Specifically, AdaBoost algorithm will allow obtaining an explicit set of rules that explains the class for each input data, adding interpretability to the models. In this paper, an innovative approach to perform an online classification, that is, to identify the discharge before it actually ends, using interpretable models is presented. In order to evaluate and reveal the benefits of rule-based models, an illustrative example has been implemented to perform an online classification of five different signals of the TJ-II stellarator fusion device located in Madrid, Spain.