A scSE-LinkNet Deep Learning Model for Daytime Sea Fog Detection

Sea fog is a precarious weather disaster affecting transportation on the sea. The accuracy of the threshold method for sea fog detection is limited by time and region. In comparison, the deep learning method learns features of objects through different network layers and can therefore accurately extract fog data and is less affected by temporal and spatial factors. This study proposes a scSE-LinkNet model for daytime sea fog detection that leverages residual blocks to encoder feature maps and attention module to learn the features of sea fog data by considering spectral and spatial information of nodes. With the help of satellite radar data from Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP), a ground sample database was extracted from Moderate Resolution Imaging Spectroradiometer (MODIS) L1B data. The scSE-LinkNet was trained on the training set, and quantitative evaluation was performed on the test set. Results showed the probability of detection (POD), false alarm rate (FAR), critical success index (CSI), and Heidke skill scores (HSS) were 0.924, 0.143, 0.800, and 0.864, respectively. Compared with other neural networks (FCN, U-Net, and LinkNet), the CSI of scSE-LinkNet was improved, with a maximum increase of nearly 8%. Moreover, the sea fog detection results were consistent with the measured data and CALIOP products.

Real-Time Detection of Freezing Motions in Parkinson's Patients for Adaptive Gait Phase Synchronous Cueing

Parkinson's disease is the second most common neurodegenerative disease worldwide reducing cognitive and motoric abilities of affected persons. Freezing of Gait (FoG) is one of the severe symptoms that is observed in the late stages of the disease and considerably impairs the mobility of the person and raises the risk of falls. Due to the pathology and heterogeneity of the Parkinsonian gait cycle, especially in the case of freezing episodes, the detection of the gait phases with wearables is challenging in Parkinson's disease. This is addressed by introducing a state-automaton-based algorithm for the detection of the foot's motion phases using a shoe-placed inertial sensor. Machine-learning-based methods are investigated to classify the actual motion phase as normal or FoG-affected and to predict the outcome for the next motion phase. For this purpose, spatio-temporal gait and signal parameters are determined from the segmented movement phases. In this context, inertial sensor fusion is applied to the foot's 3D acceleration and rate of turn. Support Vector Machine (SVM) and AdaBoost classifiers have been trained on the data of 16 Parkinson's patients who had shown FoG episodes during a clinical freezing-provoking assessment course. Two clinical experts rated the video-recorded trials and marked episodes with festination, shank trembling, shuffling, or akinesia. Motion phases inside such episodes were labeled as FoG-affected. The classifiers were evaluated using leave-one-patient-out cross-validation. No statistically significant differences could be observed between the different classifiers for FoG detection (p>0.05). An SVM model with 10 features of the actual and two preceding motion phases achieved the highest average performance with 88.5 ± 5.8% sensitivity, 83.3 ± 17.1% specificity, and 92.8 ± 5.9% Area Under the Curve (AUC). The performance of predicting the behavior of the next motion phase was significantly lower compared to the detection classifiers. No statistically significant differences were found between all prediction models. An SVM-predictor with features from the two preceding motion phases had with 81.6 ± 7.7% sensitivity, 70.3 ± 18.4% specificity, and 82.8 ± 7.1% AUC the best average performance. The developed methods enable motion-phase-based FoG detection and prediction and can be utilized for closed-loop systems that provide on-demand gait-phase-synchronous cueing to mitigate FoG symptoms and to prevent complete motoric blockades.

Prediction and detection of freezing of gait in Parkinson’s disease from plantar pressure data using long short-term memory neural-networks

Background Freezing of gait (FOG) is a walking disturbance in advanced stage Parkinson’s disease (PD) that has been associated with increased fall risk and decreased quality of life. Freezing episodes can be mitigated or prevented with external intervention such as visual or auditory cues, activated by FOG prediction and detection systems. While most research on FOG detection and prediction has been based on inertial measurement unit (IMU) and accelerometer data, plantar-pressure data may capture subtle weight shifts unique to FOG episodes. Different machine learning algorithms have been used for FOG detection and prediction; however, long short-term memory (LSTM) deep learning methods hold an advantage when dealing with time-series data, such as sensor data. This research aimed to determine if LSTM can be used to detect and predict FOG from plantar pressure data alone, specifically for use in a real-time wearable system. Methods Plantar pressure data were collected from pressure-sensing insole sensors worn by 11 participants with PD as they walked a predefined freeze-provoking path. FOG instances were labelled, 16 features were extracted, and the dataset was balanced and normalized (z-score). The resulting datasets were classified using long short-term memory neural-network models. Separate models were trained for detection and prediction. For prediction models, data before FOG were included in the target class. Leave-one-freezer-out cross validation was used for model evaluation. In addition, the models were tested on all non-freezer data to determine model specificity. Results The best FOG detection model had 82.1% (SD 6.2%) mean sensitivity and 89.5% (SD 3.6%) mean specificity for one-freezer-held-out cross validation. Specificity improved to 93.3% (SD 4.0%) when ignoring inactive state data (standing) and analyzing the model only on active states (turning and walking). The model correctly detected 95% of freeze episodes. The best FOG prediction method achieved 72.5% (SD 13.6%) mean sensitivity and 81.2% (SD 6.8%) mean specificity for one-freezer-held-out cross validation. Conclusions Based on FOG data collected in a laboratory, the results suggest that plantar pressure data can be used for FOG detection and prediction. However, further research is required to improve FOG prediction performance, including training with a larger sample of people who experience FOG.

Detection of agglomerate fog based on a shallow convolutional neural network

As a kind of frequent bad weather, Agglomerate fog is a serious danger to people's safe driving, especially on the highway. Therefore, the research on the detection of fog is of great practical significance to ensure the safety of pedestrians. This paper proposes a shallow convolutional neural network for agglomerate fog detection in images, including the framework of the network and the detailed design of each component. Firstly, the image is divided into several sub-images; and then a shallow convolutional neural network is constructed and employed to identify the existence of fog for each of the sub-area images; lastly, the decision results of each sub-area images were integrated to determine whether the whole image contained agglomerate fog. A large quantity of simulation data and real data were used to test the performance of the proposed method, the experimental results show that the presented method can achieve more than 90% detection accuracy, which demonstrated that the advantage of the proposed method comparing with several existed methods.

Grouping successive freezing of gait episodes has neutral to detrimental effect on freeze detection and prediction in Parkinson’s disease

Freezing of gait (FOG) is an intermittent walking disturbance experienced by people with Parkinson’s disease (PD). Wearable FOG identification systems can improve gait and reduce the risk of falling due to FOG by detecting FOG in real-time and providing a cue to reduce freeze duration. However, FOG prediction and prevention is desirable. Datasets used to train machine learning models often generate ground truth FOG labels based on visual observation of specific lower limb movements (event-based definition) or an overall inability to walk effectively (period of gait disruption based definition). FOG definition ambiguity may affect model performance, especially with respect to multiple FOG in rapid succession. This research examined whether merging multiple freezes that occurred in rapid succession could improve FOG detection and prediction model performance. Plantar pressure and lower limb acceleration data were used to extract a feature set and train decision tree ensembles. FOG was labeled using an event-based definition. Additional datasets were then produced by merging FOG that occurred in rapid succession. A merging threshold was introduced where FOG that were separated by less than the merging threshold were merged into one episode. FOG detection and prediction models were trained for merging thresholds of 0, 1, 2, and 3 s. Merging slightly improved FOG detection model performance; however, for the prediction model, merging resulted in slightly later FOG identification and lower precision. FOG prediction models may benefit from using event-based FOG definitions and avoiding merging multiple FOG in rapid succession.

Integration of Image-Based Fog Detection with Autonomous Decision System for Intelligent Road Sign

The paper presents the description of the decision system implemented for Intelligent Road Signs. It focuses on the implementation of the novel air transparency analysis system and its integration with the rule system and the speed control infrastructure. Moreover, there are presented issues of making decisions about the content displayed in the case of autonomous and cooperating signs. To reflect more closely on real-life situations, it is assumed that the content presented by the IRS changes dynamically, depending on the road traffic and weather parameters. The IRS system operation was presented using fog detection as an example.

Dealing with the Heterogeneous Presentations of Freezing of Gait: How Reliable are the Freezing Index and Heart Rate for Freezing Detection?

Background: Freezing of gait (FOG) is an unpredictable gait arrest that hampers the lives of 40% of people with Parkinson’s disease. Because the symptom is heterogeneous in phenotypical presentation (it can present as trembling, shuffling, or akinesia) and manifests during various circumstances (it can be triggered by e.g. turning, passing doors, and dual-tasking), it is particularly difficult to detect with motion sensors. The freezing index (FI) is one of the most frequently used accelerometer-based methods for FOG detection. However, it might not adequately distinguish FOG from voluntary stops, certainly for the akinetic type of FOG. Interestingly, a previous study showed that heart rate signals could distinguish FOG from stopping and turning movements. This study aimed to investigate for which phenotypes and evoking circumstances the FI and heart rate might provide reliable signals for FOG detection.Methods: Sixteen people with Parkinson’s disease and daily freezing completed a gait trajectory designed to provoke FOG including turns, narrow passages, starting, and stopping, with and without a cognitive or motor dual-task. We compared the FI and heart rate of 406 FOG events to baseline levels, and to stopping and normal gait events (i.e. turns and narrow passages without FOG) using mixed-effects models. We specifically evaluated the influence of different types of FOG (trembling vs akinesia) and triggering situations (turning vs narrow passages; no dual-task vs cognitive dual-task vs motor dual-task) on both outcome measures. Results: The FI increased significantly for trembling FOG, but not for akinetic FOG. Furthermore, the index increased similarly during stopping and was therefore not significantly different from FOG. In contrast, heart rate change during FOG was for all types and during all triggering situations statistically different from stopping, but not from normal gait events. Conclusion: The FI has issues to distinguish FOG from voluntary stopping, especially of the akinetic type. In contrast, the clear distinction in heart rate change between FOG and voluntary stops, which was independent of the heterogeneous presentation of FOG, might provide a solution for this issue. Therefore, we suggest that combining a heart rate monitor with a motion sensor may be promising for future FOG detection.

Developing a long-term high-resolution winter fog climatology over south Asia using satellite observations from 2002 to 2020

The vast Indo-Gangetic Plains (IGP) south of the Himalaya are subject to dense fog every year during winter months (December-January), severely disrupting rail, air and public transport of millions of people living in northern India, Pakistan, Nepal and Bangladesh. Air pollution combined with high moisture availability in the shallow boundary layer, are important factors affecting the persistence and widespread nature of fog over the IGP. Despite the environmental significance and impacts on the public at-large, an in depth understanding of the long-term spatial-temporal distribution of the south Asian fog, is presently not available in the literature. We utilize infrared remote sensing techniques to develop a high-resolution (≈1 km x 1 km) fog detection climatology over the past two decades (2002 – 2020), using Aqua/MODIS satellite observations. A dynamic brightness temperature difference threshold (involving 3.96 μm and 11.03 μm bands) for nighttime fog detection is constructed based on systematic radiative transfer simulations involving cloud effective radius, cloud top height, cloud optical depth and satellite viewing geometry. Our satellite-based fog detection is consistent with theoretical simulations of fog characterization and is also found to be well-correlated with near-surface visibility observations of dense fog (r = 0.87, p-value << 0.01). We also provide satellite-derived nighttime estimates of fog/low-cloud effective radius which is in general agreement with the operational daytime MODIS cloud data product and limited in situ observations. In terms of fog frequency, the IGP is relatively uniformly covered by widespread fog occurrences with the largest frequency found in the low-lying Terai region, bordering India and Nepal, which is also consistently observed in our daytime fog detection results over the last two decades. Additionally, the interannual variations in fog occurrences track closely with that of relative humidity in the IGP, which is associated with shallow boundary layer conditions during winter-time favoring the formation and persistence of fog. Overall, these long-term satellite-derived results present new high-resolution data and insights into the dense and often intense winter fog occurrences which routinely engulf the entire stretch of the Indo-Gangetic Plains and cause significant degradation to ground visibility in one of the world’s most densely populated regions.

Towards a climatology of fog frequency in the Atacama Desert via multi-spectral satellite data and machine learning techniques

In many hyper-arid ecosystems, such as the Atacama Desert, fog is the most important fresh water source. To study biological and geological processes in such water-limited regions, knowledge about the spatio-temporal distribution and variability of fog presence is necessary. In this study, in-situ measurements provided by a network of climate stations equipped, inter alia, with leaf wetness sensors are utilized to create a reference fog data set which enables the validation of satellite-based fog retrieval methods. Further, a new satellite-based fog detection approach is introduced which uses brightness temperatures measured by the Moderate Resolution Imaging Spectroradiometer (MODIS) as input for a neural network. Such a machine learning technique can exploit all spectral information of the satellite data and represent potential non-linear relationships. Compared to a second fog detection approach based on MODIS cloud top height retrievals, the neural network reaches a higher detection skill (Heidke skill score of 0.56 compared to 0.49). A suitable representation of temporal variability on subseasonal time scales is provided with correlations mostly greater than 0.7 between fog occurrence time series derived from the neural network and the reference data for individual climate stations, respectively. Furthermore, a suitable spatial representativity of the neural network approach to expand the application to the whole region is indicated. Three-year averages of fog frequencies reveal similar spatial patterns for the austral winter season for both approaches. However, differences are found for the summer and potential reasons are discussed.