Search Results (8)

Passive electric field sensing can be utilized in a wide variety of application areas, although it has certain limitations. In order to better understand what these limitations are and how countervailing measures to these limitations could be implemented, this paper contributes an in-depth discussion of problems with passive electric field sensing and how to bypass or solve them. The focus lies on the explanation of how commonly known signal processing techniques and hardware build-up schemes can be used to improve passive electric field sensors and the corresponding data processing. Full article

Heterogeneous cyberattacks against industrial control systems (ICSs) have had a strong impact on the physical world in recent decades. Connecting devices to the internet enables new attack surfaces for attackers. The intrusion of ICSs, such as the manipulation of industrial sensory or actuator data, can be the cause for anomalous ICS behaviors. This poses a threat to the infrastructure that is critical for the operation of a modern city. Nowadays, the best techniques for detecting anomalies in ICSs are based on machine learning and, more recently, deep learning. Cybersecurity in ICSs is still an emerging field, and industrial datasets that can be used to develop anomaly detection techniques are rare. In this paper, we propose an unsupervised deep learning methodology for anomaly detection in ICSs, specifically, a lightweight long short-term memory variational auto-encoder (LW-LSTM-VAE) architecture. We successfully demonstrate our solution under two ICS applications, namely, water purification and water distribution plants. Our proposed method proves to be efficient in detecting anomalies in these applications and improves upon reconstruction-based anomaly detection methods presented in previous work. For example, we successfully detected 82.16% of the anomalies in the scenario of the widely used Secure Water Treatment (SWaT) benchmark. The deep learning architecture we propose has the added advantage of being extremely lightweight. Full article

This work addresses the challenge of building an accurate and generalizable periocular recognition model with a small number of learnable parameters. Deeper (larger) models are typically more capable of learning complex information. For this reason, knowledge distillation (kd) was previously proposed to carry this knowledge from a large model (teacher) into a small model (student). Conventional KD optimizes the student output to be similar to the teacher output (commonly classification output). In biometrics, comparison (verification) and storage operations are conducted on biometric templates, extracted from pre-classification layers. In this work, we propose a novel template-driven KD approach that optimizes the distillation process so that the student model learns to produce templates similar to those produced by the teacher model. We demonstrate our approach on intra- and cross-device periocular verification. Our results demonstrate the superiority of our proposed approach over a network trained without KD and networks trained with conventional (vanilla) KD. For example, the targeted small model achieved an equal error rate (EER) value of 22.2% on cross-device verification without KD. The same model achieved an EER of 21.9% with the conventional KD, and only 14.7% EER when using our proposed template-driven KD. Full article

Smartphones have become an essential part of our lives. Especially its computing power and its current specifications make a modern smartphone a powerful device for human activity recognition tasks. Equipped with various integrated sensors, a modern smartphone can be leveraged for lots of smart applications. We already investigated the possibility of using an unmodified commercial smartphone to recognize eight strength-based exercises. App-based workouts have become popular in the last few years. The advantage of using a mobile device is that you can practice anywhere at anytime. In our previous work, we proved the possibility of turning a commercial smartphone into an active sonar device to leverage the echo reflected from exercising movement close to the device. By conducting a test study with 14 participants, we showed the first results for cross person evaluation and the generalization ability of our inference models on disjoint participants. In this work, we extended another model to further improve the model generalizability and provided a thorough comparison of our proposed system to other existing state-of-the-art approaches. Finally, a concept of counting the repetitions is also provided in this study as a parallel task to classification. Full article

The concept of the quantified self has gained popularity in recent years with the hype of miniaturized gadgets to monitor vital fitness levels. Smartwatches or smartphone apps and other fitness trackers are overwhelming the market. Most aerobic exercises such as walking, running, or cycling can be accurately recognized using wearable devices. However whole-body exercises such as push-ups, bridges, and sit-ups are performed on the ground and thus cannot be precisely recognized by wearing only one accelerometer. Thus, a floor-based approach is preferred for recognizing whole-body activities. Computer vision techniques on image data also report high recognition accuracy; however, the presence of a camera tends to raise privacy issues in public areas. Therefore, we focus on combining the advantages of ubiquitous proximity-sensing with non-optical sensors to preserve privacy in public areas and maintain low computation cost with a sparse sensor implementation. Our solution is the ExerTrack, an off-the-shelf sports mat equipped with eight sparsely distributed capacitive proximity sensors to recognize eight whole-body fitness exercises with a user-independent recognition accuracy of 93.5% and a user-dependent recognition accuracy of 95.1% based on a test study with 9 participants each performing 2 full sessions. We adopt a template-based approach to count repetitions and reach a user-independent counting accuracy of 93.6%. The final model can run on a Raspberry Pi 3 in real time. This work includes data-processing of our proposed system and model selection to improve the recognition accuracy and data augmentation technique to regularize the network. Full article

Tracking objects over time, i.e., identity (ID) consistency, is important when dealing with multiple object tracking (MOT). Especially in complex scenes with occlusion and interaction of objects this is challenging. Significant improvements in single object tracking (SOT) methods have inspired the introduction of SOT to MOT to improve the robustness, that is, maintaining object identities as long as possible, as well as helping alleviate the limitations from imperfect detections. SOT methods are constantly generalized to capture appearance changes of the object, and designed to efficiently distinguish the object from the background. Hence, simply extending SOT to a MOT scenario, which consists of a complex scene with spatially mixed, occluded, and similar objects, will encounter problems in computational efficiency and drifted results. To address this issue, we propose a binary-channel verification model that deeply excavates the potential of SOT in refining the representation while maintaining the identities of the object. In particular, we construct an integrated model that jointly processes the previous information of existing objects and new incoming detections, by using a unified correlation filter through the whole process to maintain consistency. A delay processing strategy consisting of the three parts—attaching, re-initialization, and re-claiming—is proposed to tackle drifted results caused by occlusion. Avoiding the fuzzy appearance features of complex scenes in MOT, this strategy can improve the ability to distinguish specific objects from each other without contaminating the fragile training space of a single object tracker, which is the main cause of the drift results. We demonstrate the effectiveness of our proposed approach on the MOT17 challenge benchmarks. Our approach shows better overall ID consistency performance in comparison with previous works. Full article

Quantified Self has seen an increased interest in recent years, with devices including smartwatches, smartphones, or other wearables that allow you to monitor your fitness level. This is often combined with mobile apps that use gamification aspects to motivate the user to perform fitness activities, or increase the amount of sports exercise. Thus far, most applications rely on accelerometers or gyroscopes that are integrated into the devices. They have to be worn on the body to track activities. In this work, we investigated the use of a speaker and a microphone that are integrated into a smartphone to track exercises performed close to it. We combined active sonar and Doppler signal analysis in the ultrasound spectrum that is not perceivable by humans. We wanted to measure the body weight exercises bicycles, toe touches, and squats, as these consist of challenging radial movements towards the measuring device. We have tested several classification methods, ranging from support vector machines to convolutional neural networks. We achieved an accuracy of 88% for bicycles, 97% for toe-touches and 91% for squats on our test set. Full article

Even when the Global Positioning System (GPS) signal is blocked, a rate gyroscope (gyro) north finder is capable of providing the required azimuth reference information to a certain extent. In order to measure the azimuth between the observer and the north direction very accurately, we propose a multi-position non-continuous rotation gyro north finding scheme. Our new generalized mathematical model analyzes the elements that affect the azimuth measurement precision and can thus provide high precision azimuth reference information. Based on the gyro’s principle of detecting a projection of the earth rotation rate on its sensitive axis and the proposed north finding scheme, we are able to deduct an accurate mathematical model of the gyro outputs against azimuth with the gyro and shaft misalignments. Combining the gyro outputs model and the theory of propagation of uncertainty, some approaches to optimize north finding are provided, including reducing the gyro bias error, constraining the gyro random error, increasing the number of rotation points, improving rotation angle measurement precision, decreasing the gyro and the shaft misalignment angles. According them, a north finder setup is built and the azimuth uncertainty of 18” is obtained. This paper provides systematic theory for analyzing the details of the gyro north finder scheme from simulation to implementation. The proposed theory can guide both applied researchers in academia and advanced practitioners in industry for designing high precision robust north finder based on different types of rate gyroscopes. Full article