Spatio-temporal Filtering Research Articles

An adaptive spatiotemporal filter for ultrasound localization microscopy based on density canopy clustering

Ultrasound Localization Microscopy (ULM) facilitates structural and hemodynamic imaging of microvessels with a resolution of tens of micrometers. In ULM, the extraction of effective microbubble signals is crucial for image quality. Singular Value Decomposition (SVD) is currently the most prevalent method for microbubble signal extraction in ULM. Most existing ULM studies employ a fixed SVD filter threshold using empirical values which will lead to imaging quality degradation due to the insufficient separation of blood signals. In this study, we propose an adaptive and non-threshold SVD filter based on canopy-density clustering, termed DCC-SVD. This filter automatically classifies the components of the SVD based on the density of their spatiotemporal features, eliminating the need for parameter selection. In in vitro tube phantom, DCC-SVD demonstrated its ability to adaptive separation of blood and bubble signal at varying microbubble concentrations and flow rates. We compared the proposed DCC-SVD method with the Block-match 3D (BM3D) filter and a classical adaptive method called spatial similarity matrix (SSM), using concentration-variable in vivo rat brain data, as well as open-source rat kidney and mouse tumor datasets. The proposed DCC-SVD improved the global spatial resolution by approximately 4 μm from 30.39 μm to 26.02 μm. It also captured vessel structure absent in images obtained by other methods and yielded a smoother vessel intensity profile, making it a promising spatiotemporal filter for ULM imaging.

ISLAND: Interpolating Land Surface Temperature using land cover

Cloud occlusion is a common problem in the field of remote sensing, particularly for retrieving Land Surface Temperature (LST). Remote sensing thermal instruments onboard operational satellites are supposed to enable frequent and high-resolution observations over land; unfortunately, clouds adversely affect thermal signals by blocking outgoing longwave radiation emission from the Earth’s surface, interfering with the retrieved ground emission temperature. Such cloud contamination severely reduces the set of serviceable LST images for downstream applications, making it impractical to perform intricate time-series analysis of LST. In this paper, we introduce a novel method to remove cloud occlusions from Landsat 8 LST images. We call our method ISLAND, an acronym for Interpolating Land Surface Temperature using land cover. Our approach uses LST images from Landsat 8 (at 30m resolution with 16-day revisit cycles) and the NLCD land cover dataset. Inspired by Tobler’s first law of Geography, ISLAND predicts occluded LST through a set of spatio-temporal filters that perform distance-weighted spatio-temporal interpolation. A critical feature of ISLAND is that the filters are land cover-class aware, making it particularly advantageous in complex urban settings with heterogeneous land cover types and distributions. Through qualitative and quantitative analysis, we show that ISLAND achieves robust reconstruction performance across a variety of cloud occlusion and surface land cover conditions, and with a high spatio-temporal resolution. We provide a public dataset of 20 U.S. cities with pre-computed ISLAND LST outputs. Using several case studies, we demonstrate that ISLAND opens the door to a multitude of high-impact urban and environmental applications across the continental United States.

Spatiotemporal Filter Velocimetry With Arbitrary-Shaped Measurement Region For Accurate Measurement In The Vicinity Of A Boundary

Viscous stress at a boundary or an interface is a source of drag acting on a body or a fluid particle and therefore its measurement is of great importance for understanding the motion of the body or the fluid particle. In this study, spatiotemporal filter velocimetry (SFV: Hosokawa & Tomiyama, 2012) is extended to arbitrary quadrilateral measurement areas fitted to an interface shape for accurate measurement of flow about a deformed drop, and it is applied to flows about single drops of glycerol-water solution falling in stagnant silicon oil under clean and contaminated conditions. Figure 1 shows the trajectories of tracer particles and the measured velocity distributions. No separation is observed for the spherical and ellipsoidal drops. On the other hand, the stagnant region is observed in the dimple region of the tail of the cap drop. These results clearly indicate that the flow about the deformed drop is successfully measured by SFV with arbitrary quadrilateral measurement areas. Figure 2 shows the trajectories of tracer particles and the measured velocity distributions of single ellipsoidal drops in clean and contaminated systems. Triton X-100 was used as the surfactant and its concentration C in the drop ranges from 0.0 to 0.1 mol/m3 . The drop tail becomes flat and then slightly dimpled shape, and the internal circulation becomes weak as C increases. The internal circulation remains even at the high surfactant concentration condition under which the internal circulation in single spherical drop is fully damped (Hosokawa et al., 2017). We also evaluate interfacial velocity, interfacial shear stresses, the Marangoni stress and surfactant concentration at the interface from measured velocity distributions in the vicinity of the interface. As a result, we confirmed that SFV enables us to measure velocity distribution about a single deformed drop and enables us to evaluate interfacial velocity, interfacial shear stresses, the Marangoni stress and surfactant concentration at the interface.

EEG decoding with spatiotemporal convolutional neural network for visualization and closed-loop control of sensorimotor activities: A simultaneous EEG-fMRI study.

Closed-loop neurofeedback training utilizes neural signals such as scalp electroencephalograms (EEG) to manipulate specific neural activities and the associated behavioral performance. A spatiotemporal filter for high-density whole-head scalp EEG using a convolutional neural network can overcome the ambiguity of the signaling source because each EEG signal includes information on the remote regions. We simultaneously acquired EEG and functional magnetic resonance images in humans during the brain-computer interface (BCI) based neurofeedback training and compared the reconstructed and modeled hemodynamic responses of the sensorimotor network. Filters constructed with a convolutional neural network captured activities in the targeted network with spatial precision and specificity superior to those of the EEG signals preprocessed with standard pipelines used in BCI-based neurofeedback paradigms. The middle layers of the trained model were examined to characterize the neuronal oscillatory features that contributed to the reconstruction. Analysis of the layers for spatial convolution revealed the contribution of distributed cortical circuitries to reconstruction, including the frontoparietal and sensorimotor areas, and those of temporal convolution layers that successfully reconstructed the hemodynamic response function. Employing a spatiotemporal filter and leveraging the electrophysiological signatures of the sensorimotor excitability identified in our middle layer analysis would contribute to the development of a further effective neurofeedback intervention.

Estimating receptive fields of simple and complex cells in early visual cortex: A convolutional neural network model with parameterized rectification.

Neurons in the primary visual cortex respond selectively to simple features of visual stimuli, such as orientation and spatial frequency. Simple cells, which have phase-sensitive responses, can be modeled by a single receptive field filter in a linear-nonlinear model. However, it is challenging to analyze phase-invariant complex cells, which require more elaborate models having a combination of nonlinear subunits. Estimating parameters of these models is made additionally more difficult by cortical neurons' trial-to-trial response variability. We develop a simple convolutional neural network method to estimate receptive field models for both simple and complex visual cortex cells from their responses to natural images. The model consists of a spatiotemporal filter, a parameterized rectifier unit (PReLU), and a two-dimensional Gaussian "map" of the receptive field envelope. A single model parameter determines the simple vs. complex nature of the receptive field, capturing complex cell responses as a summation of homogeneous subunits, and collapsing to a linear-nonlinear model for simple type cells. The convolutional method predicts simple and complex cell responses to natural image stimuli as well as grating tuning curves. The fitted models yield a continuum of values for the PReLU parameter across the sampled neurons, showing that the simple/complex nature of cells can vary in a continuous manner. We demonstrate that complex-like cells respond less reliably than simple-like cells. However, compensation for this unreliability with noise ceiling analysis reveals predictive performance for complex cells proportionately closer to that for simple cells. Most spatial receptive field structures are well fit by Gabor functions, whose parameters confirm well-known properties of cat A17/18 receptive fields.

FAST: Filter-Adapted Spatio-Temporal Sampling for Real-Time Rendering

Stochastic sampling techniques are ubiquitous in real-time rendering, where performance constraints force the use of low sample counts, leading to noisy intermediate results. To remove this noise, the post-processing step of temporal and spatial denoising is an integral part of the real-time graphics pipeline. The main insight presented in this paper is that we can optimize the samples used in stochastic sampling such that the post-processing error is minimized. The core of our method is an analytical loss function which measures post-filtering error for a class of integrands --- multidimensional Heaviside functions. These integrands are an approximation of the discontinuous functions commonly found in rendering. Our analysis applies to arbitrary spatial and spatiotemporal filters, scalar and vector sample values, and uniform and non-uniform probability distributions. We show that the spectrum of Monte Carlo noise resulting from our sampling method is adapted to the shape of the filter, resulting in less noisy final images. We demonstrate improvements over state-of-the-art sampling methods in three representative rendering tasks: ambient occlusion, volumetric ray-marching, and color image dithering. Common use noise textures, and noise generation code is available at https://github.com/electronicarts/fastnoise.

Adaptive target response-based spatio-temporal regularized correlation filter for UAV-based object tracking

Adaptive target response-based spatio-temporal regularized correlation filter for UAV-based object tracking

Research on Land-Based Wind/Solar Power Station Site Deformation Monitoring Based on SBAS-InSAR Technology

INTRODUCTION: In recent years, China has been building extensive wind/solar power stations. During the construction and operation of land-based wind/solar power stations, deformation monitoring is an important method to investigate the station stability. 
 OBJECTIVES: Therefore, this study uses Sentinel-1 data and time-series InSAR technology to monitor the deformation of photovoltaic and wind power stations in Qingyuan County. 
 METHODS: InSAR technology obtains deformation rate maps in the radar line of sight (LOS) direction for a wide area around the power station sites. Since wind/solar power stations are mainly located in natural environments with relatively dense vegetation coverage, this paper proposes a SBAS-InSAR method integrated with spatiotemporal filtering to accurately extract the time series deformation over a large area. Based on the statistical characteristic difference between the deformation and the atmospheric delay, spatiotemporal filterings are applied to remove the atmospheric delay from the InSAR derived deformation results. 
 RESULTS: The experimental results show that spatiotemporal filtering is an effective and fast method to remove atmospheric delay. 
 CONCLUSION: The integration of BSAS-InSAR with spatiotemporal filtering has great potential applications in the deformation monitoring of land-based wind/solar power station sites, which is critical for the construction and operation of land-based wind/solar power stations.

Dynamic Multiple Object Segmentation with Spatio-Temporal Filtering.

This article primarily focuses on the localization and extraction of multiple moving objects in images taken from a moving camera platform, such as image sequences captured by drones. The positions of moving objects in the images are influenced by both the camera's motion and the movement of the objects themselves, while the background position in the images is related to the camera's motion. The main objective of this article was to extract all moving objects from the background in an image. We first constructed a motion feature space containing motion distance and direction, to map the trajectories of feature points. Subsequently, we employed a clustering algorithm based on trajectory distinctiveness to differentiate between moving objects and the background, as well as feature points corresponding to different moving objects. The pixels between the feature points were then designated as source points. Within local regions, complete moving objects were segmented by identifying these pixels. We validated the algorithm on some sequences in the Video Verification of Identity (VIVID) program database and compared it with relevant algorithms. The experimental results demonstrated that, in the test sequences when the feature point trajectories exceed 10 frames, there was a significant difference in the feature space between the feature points on the moving objects and those on the background. Correctly classified frames with feature points accounted for 67% of the total frames.The positions of the moving objects in the images were accurately localized, with an average IOU value of 0.76 and an average contour accuracy of 0.57. This indicated that our algorithm effectively localized and segmented the moving objects in images captured by moving cameras.

Nonlinear spatio-temporal filter to reduce crosstalk in bipolar electromyogram

Objective. The wide detection volume of surface electromyogram (EMG) makes it prone to crosstalk, i.e. the signal from other muscles than the target one. Removing this perturbation from bipolar recordings is an important open problem for many applications. Approach. An innovative nonlinear spatio-temporal filter is developed to estimate the EMG generated by the target muscle by processing noisy signals from two bipolar channels, placed over the target and the crosstalk muscle, respectively. The filter is trained on some calibration data and then can be applied on new signals. Tests are provided in simulations (considering different thicknesses of the subcutaneous tissue, inter-electrode distances, locations of the EMG channels, force levels) and experiments (from pronator teres and flexor carpi radialis of 8 healthy subjects). Main results. The proposed filter allows to reduce the effect of crosstalk in all investigated conditions, with a statistically significant reduction of its root mean squared of about 20%, both in simulated and experimental data. Its performances are also superior to those of a blind source separation method applied to the same data. Significance. The proposed filter is simple to be applied and feasible in applications in which single bipolar channels are placed over the muscles of interest. It can be useful in many fields, such as in gait analysis, tests of myoelectric fatigue, rehabilitation with EMG biofeedback, clinical studies, prosthesis control.

Multi-parametric Retinal Microvascular Functional Perfusion Imaging Based on Dynamic Fundus Fluorescence Angiography.

Retinal microvascular disease has caused serious visual impairment widely in the world, which can be hopefully prevented via early and precision microvascular hemodynamic diagnosis. Due to artifacts from choroidal microvessels and tiny movements, current fundus microvascular imaging techniques including fundus fluorescein angiography (FFA) precisely identify retinal microvascular microstructural damage and abnormal hemodynamic changes difficulty, especially in the early stage. Therefore, this study proposes an FFA-based multi-parametric retinal microvascular functional perfusion imaging (RM-FPI) scheme to assess the microstructural damage and quantify its hemodynamic distribution precisely. Herein, a spatiotemporal filter based on singular value decomposition combined with a lognormal fitting model was used to remove the above artifacts. Dynamic FFAs of patients (n = 7) were collected first. The retinal time fluorescence intensity curves were extracted and the corresponding perfusion parameters were estimated after decomposition filtering and model fitting. Compared with in vivo results without filtering and fitting, the signal-to-clutter ratio of retinal perfusion curves, average contrast, and resolution of RM-FPI were up to 7.32 ± 0.43 dB, 14.34 ± 0.24 dB, and 11.0 ± 2.0 μm, respectively. RM-FPI imaged retinal microvascular distribution and quantified its spatial hemodynamic changes, which further characterized the parabolic distribution of local blood flow within diameters ranging from 9 to 400 μm. Finally, RM-FPI was used to quantify, visualize, and diagnose the retinal hemodynamics of retinal vein occlusion from mild to severe. Therefore, this study provided a scheme for early and precision diagnosis of retinal microvascular disease, which might be beneficial in preventing its development.

Adaptive Unsupervised Learning-Based 3D Spatiotemporal Filter for Event-Driven Cameras.

In the evolving landscape of robotics and visual navigation, event cameras have gained important traction, notably for their exceptional dynamic range, efficient power consumption, and low latency. Despite these advantages, conventional processing methods oversimplify the data into 2 dimensions, neglecting critical temporal information. To overcome this limitation, we propose a novel method that treats events as 3D time-discrete signals. Drawing inspiration from the intricate biological filtering systems inherent to the human visual apparatus, we have developed a 3D spatiotemporal filter based on unsupervised machine learning algorithm. This filter effectively reduces noise levels and performs data size reduction, with its parameters being dynamically adjusted based on population activity. This ensures adaptability and precision under various conditions, like changes in motion velocity and ambient lighting. In our novel validation approach, we first identify the noise type and determine its power spectral density in the event stream. We then apply a one-dimensional discrete fast Fourier transform to assess the filtered event data within the frequency domain, ensuring that the targeted noise frequencies are adequately reduced. Our research also delved into the impact of indoor lighting on event stream noise. Remarkably, our method led to a 37% decrease in the data point cloud, improving data quality in diverse outdoor settings.

Clutter filtering of angular domain data for contrast-free ultrafast microvascular imaging

Objective. Contrast-free microvascular imaging is clinically valuable for the assessment of physiological status and the early diagnosis of diseases. Effective clutter filtering is essential for microvascular visualization without contrast enhancement. Singular value decomposition (SVD)-based spatiotemporal filter has been widely used to suppress clutter. However, clinical real-time imaging relies on short ensembles (dozens of frames), which limits the implementation of SVD filtering due to the large error of eigen-correlated estimations and high dependence on optimal threshold when used in such ensembles. Approach. To address the above challenges of imaging in short ensembles, two optimized filters of angular domain data are proposed in this paper: grouped angle SVD (GA-SVD) and angular-coherence-based higher-order SVD (AC-HOSVD). GA-SVD applies SVD to the concatenation of all angular data to improve clutter rejection performance in short ensembles, while AC-HOSVD applies HOSVD to the angular data tensor and utilizes angular coherence in addition to spatial and temporal features for filtering. Feasible threshold selection strategies in each feature space are provided. The clutter rejection performance of the proposed filters and SVD was evaluated with Doppler phantom and in vivo studies at different cases. Moreover, the robustness of the filters was explored under wrong singular value threshold estimation, and their computational complexity was studied. Main results. Qualitative and quantitative results indicated that GA-SVD and AC-HOSVD can effectively improve clutter rejection performance in short ensembles, especially AC-HOSVD. Notably, the proposed methods using 20 frames had similar image quality to SVD using 100 frames. In vivo studies showed that compared to SVD, GA-SVD increased the signal-to-noise-ratio (SNR) by 6.03 dB on average, and AC-HOSVD increased the SNR by 8.93 dB on average. Furthermore, AC-HOSVD remained better power Doppler image quality under non-optimal thresholds, followed by GA-SVD. Significance. The proposed filters can greatly enhance contrast-free microvascular visualization in short ensembles and have potential for different clinical translations due to the performance differences.

Sparse Bayesian Learning for End-to-End EEG Decoding.

Decoding brain activity from non-invasive electroencephalography (EEG) is crucial for brain-computer interfaces (BCIs) and the study of brain disorders. Notably, end-to-end EEG decoding has gained widespread popularity in recent years owing to the remarkable advances in deep learning research. However, many EEG studies suffer from limited sample sizes, making it difficult for existing deep learning models to effectively generalize to highly noisy EEG data. To address this fundamental limitation, this paper proposes a novel end-to-end EEG decoding algorithm that utilizes a low-rank weight matrix to encode both spatio-temporal filters and the classifier, all optimized under a principled sparse Bayesian learning (SBL) framework. Importantly, this SBL framework also enables us to learn hyperparameters that optimally penalize the model in a Bayesian fashion. The proposed decoding algorithm is systematically benchmarked on five motor imagery BCI EEG datasets ( N=192) and an emotion recognition EEG dataset ( N=45), in comparison with several contemporary algorithms, including end-to-end deep-learning-based EEG decoding algorithms. The classification results demonstrate that our algorithm significantly outperforms the competing algorithms while yielding neurophysiologically meaningful spatio-temporal patterns. Our algorithm therefore advances the state-of-the-art by providing a novel EEG-tailored machine learning tool for decoding brain activity.Code is available at https://github.com/EEGdecoding/Code-SBLEST.

The Detection and Control Factor Analysis of Active Landslides in Guizhou Province, China, Using Sentinel-1 SAR Imagery

Catastrophic landslides occur frequently in Guizhou Province, China, and the landslides in this area have special geomorphological, geological, and anthropogenic features. In order to detect and explore the distribution pattern and control factors of active landslides in Guizhou, firstly, a total of 693 active landslides throughout Guizhou Province were mapped based on the deformation rate, which was obtained by spatiotemporal filtering and Intermittent Small Baseline Subset (ISBAS) Interferometric Synthetic Aperture Radar (InSAR) techniques. Then, the relationships between the detected landslides and elevation, aspect, slope gradient, and stratigraphic lithology were analysed. Moreover, it was found that the landslides were mainly concentrated in three stratigraphic combinations, that is T1f~P2l−d, T1f~T1yn, and T2g~T1yn. Thereafter, the correlation coefficients between the landslide density and elevation and distance to the stratigraphic boundary were 0.54 and −0.19, indicating that the distribution of landslides was significantly controlled by the elevation and the boundary of specific stratigraphic combinations. Finally, we chose a typical landslide to explore how landslide development was controlled by the combined effects of elevation and stratigraphy by using ascending and descending InSAR results. We revealed that landslides occurred primarily in areas with a steep slope and a stratigraphy characterized by mudstone and sandstone.

Urban water bodies in China: Spatial distribution patterns, temporal change characteristics, and relationship with economic development

Urban water bodies constitute a vital element within the urban ecosystem. Knowledge of the spatiotemporal changes in urban water bodies and their relationship with economic development in China is critical for sustainable development. However, previous studies usually investigated the spatiotemporal changes of urban water bodies at discrete time points and ignored the relationship between urban water bodies and economic development. Here, this study first corrected freely available water body products based on spatiotemporal majority filter, thereby obtaining an improved yearly urban water body dataset. Then, the research adopted Moran’s I and Anselin Local Moran’s I to explore the spatial distribution patterns of urban water bodies and employed linear regression and segmented linear regression to investigate the temporal abrupt change and long-term trend of urban water bodies. Finally, we used geodetector to evaluate the relative contributions of GDP to the spatial heterogeneity of trends in urban water bodies. The study revealed a significant spatial autocorrelation in the distribution of urban water bodies, particularly notable in the “High-High” cluster within the Yangtze and Pearl River Deltas. It was observed that approximately 40 % and 60 % of total cities experienced an abrupt change in permanent and seasonal urban water bodies, respectively. Specifically, the majority of these abrupt changes occurred during 2010–2013. When considering both long-term trends and abrupt changes, permanent urban water bodies in Chinese cities exhibited primarily a steadily increasing trend, while seasonal water bodies in Chinese cities displayed two distinct temporal patterns: a steadily increasing trend and a decreasing trend followed by a subsequent increasing trend. Furthermore, it was noted that GDP exhibited a significantly positive correlation with the temporal changes of permanent urban water bodies from 2000 to 2020 at the national scale while it first showed a significantly negative correlation with the temporal changes of seasonal urban water bodies from 2000 to 2012 and then showed a significantly positive correlation from 2012 to 2020 on a national scale. In conclusion, we found that GDP played a more important role in driving the spatial heterogeneity of trends in urban water bodies during 2012–2020 compared to 2000–2012, with the relative contribution of ∼ 32.5 %.

Robust vehicle lane keeping control with networked proactive adaptation

Road condition is an important environmental factor for autonomous vehicle control. A dramatic change in the road condition from the nominal status is a source of uncertainty that can lead to a system failure. Once the vehicle encounters an uncertain environment, such as hitting an ice patch, it is too late to reduce the speed, and the vehicle can lose control. To cope with unforeseen uncertainties in advance, we study a proactive robust adaptive control architecture for autonomous vehicles' lane-keeping control problems. In the proposed framework, the data center generates a prior environmental uncertainty estimate with a quantified uncertainty by combining weather forecasts and measurements from anonymous vehicles through a spatio-temporal filter. The prior estimate and quantified uncertainty contribute to designing a robust heading controller and nominal longitudinal velocity for proactive adaptation to each new abnormal condition. Then the control parameters are updated based on posterior information fusion with on-board measurements.

Deep sound-field denoiser: optically-measured sound-field denoising using deep neural network.

This paper proposes a deep sound-field denoiser, a deep neural network (DNN) based denoising of optically measured sound-field images. Sound-field imaging using optical methods has gained considerable attention due to its ability to achieve high-spatial-resolution imaging of acoustic phenomena that conventional acoustic sensors cannot accomplish. However, the optically measured sound-field images are often heavily contaminated by noise because of the low sensitivity of optical interferometric measurements to airborne sound. Here, we propose a DNN-based sound-field denoising method. Time-varying sound-field image sequences are decomposed into harmonic complex-amplitude images by using a time-directional Fourier transform. The complex images are converted into two-channel images consisting of real and imaginary parts and denoised by a nonlinear-activation-free network. The network is trained on a sound-field dataset obtained from numerical acoustic simulations with randomized parameters. We compared the method with conventional ones, such as image filters, a spatiotemporal filter, and other DNN architectures, on numerical and experimental data. The experimental data were measured by parallel phase-shifting interferometry and holographic speckle interferometry. The proposed deep sound-field denoiser significantly outperformed the conventional methods on both the numerical and experimental data. Code is available on GitHub (https://github.com/nttcslab/deep-sound-field-denoiser).

Monitoring monthly tropical humid forest disturbances with Planet NICFI images in Cameroon

Selective logging and smallholder clearing are the dominant drivers of tropical forest disturbances in Cameroon (CAM). However, they are difficult to monitor accurately by satellite remote sensing because openings in the canopy can be very small, the vegetation is generally fast-growing, and cloud cover is common. Norway's International Climate and Forest Initiative (NICFI) provides access to monthly and biannual collections of 5 m Planet images in the tropics, creating a great opportunity for mapping tropical forest disturbances. In this paper, we develop a method to monitor monthly small-scale tropical humid forest disturbances using 2021 Planet NICFI images. First, a cloud mask for each of the monthly Planet NICFI images was predicted by integrating a cloud cover possibility map with a haze optimal transformation (HOT) index image. Second, possible monthly forest disturbances were mapped from a self-referenced Hue_forest (rHue_forest) index image. Finally, an adjusted monthly forest disturbance map was produced by eliminating many false positives with a spatio-temporal filter. Results in CAM demonstrated that the method applied to monthly Planet NICFI images was effective in identifying numerous small-scale tropical forest disturbances that were short-lived, lasting only a few months. After filtering out new possible forest disturbances in 2021 which did not meet a temporal permanence criterion based on the monthly images, the adjusted user´s and producer´s accuracies for CAM were 84.7 ± 2.9% and 61.5 ± 46.4%, respectively (±95% confidence intervals). Our results provide much greater spatial detail than forest disturbance methods based on Sentinel-1 and Landsat images. The adjusted disturbed area of humid forests in CAM was estimated as 1,168 ± 882 km2 in 2021. The proposed method for monthly mapping of forest disturbance using Planet NICFI images has great potential to complement existing forest cover change products and monitor hitherto neglected tropical forest disturbances due to small-scale clearing.

Improved Decoding of Attentional Selection in Multi-Talker Environments with Self-Supervised Learned Speech Representation.

Auditory attention decoding (AAD) is a technique used to identify and amplify the talker that a listener is focused on in a noisy environment. This is done by comparing the listener's brainwaves to a representation of all the sound sources to find the closest match. The representation is typically the waveform or spectrogram of the sounds. The effectiveness of these representations for AAD is uncertain. In this study, we examined the use of self-supervised learned speech representation in improving the accuracy and speed of AAD. We recorded the brain activity of three subjects using invasive electrocorticography (ECoG) as they listened to two conversations and focused on one. We used WavLM to extract a latent representation of each talker and trained a spatiotemporal filter to map brain activity to intermediate representations of speech. During the evaluation, the reconstructed representation is compared to each speaker's representation to determine the target speaker. Our results indicate that speech representation from WavLM provides better decoding accuracy and speed than the speech envelope and spectrogram. Our findings demonstrate the advantages of self-supervised learned speech representation for auditory attention decoding and pave the way for developing brain-controlled hearable technologies.