A Study on Distortion Estimation Based on Image Gradients

Image noise is a variation of uneven pixel values that occurs randomly. A good estimation of image noise parameters is crucial in image noise modeling, image denoising, and image quality assessment. To the best of our knowledge, there is no single estimator that can predict all noise parameters for multiple noise types. The first contribution of our research was to design a noise data feature extractor that can effectively extract noise information from the image pair. The second contribution of our work leveraged other noise parameter estimation algorithms that can only predict one type of noise. Our proposed method, DE-G, can estimate additive noise, multiplicative noise, and impulsive noise from single-source images accurately. We also show the capability of the proposed method in estimating multiple corruptions.

Studies of Seismic Velocities in Subduction Zones from Continuous Ocean Bottom Seismometer Data

<p><b>This thesis uses continuous ambient noise data recorded by Ocean Bottom Seismometers (OBSs) to study seismic velocities in the upper crust of the overriding plate. The first and second projects (Chapters 3 and 4) focus on temporal seismic velocity variations in the northern Hikurangi subduction zone offshore the North Island, New Zealand, while the third project (Chapter 5) focuses on shear wave velocities in the southwestern Okinawa Trough offshore northeastern Taiwan. In the first project (Chapters 3), we investigate a region of frequent slow slip events (SSEs) offshore Gisborne, North Island, New Zealand. From September to October 2014, an SSE occurred with a slip over 250 mm and was recorded successfully by the Hikurangi Ocean Bottom Investigation of Tremor and Slow Slip deployment II (HOBITSS II). We apply coda wave interferometry on the ambient noise data acquired by nine OBSs deployed by the HOBITSS II to study the seismic velocity variations related to the SSE. The average velocity variations display a decrease on the order of 0.05% during the SSE, followed by an increase of similar magnitude afterwards. Two hypotheses are proposed to explain our observation. The first hypothesis, which has been suggested by previous studies, considers that the velocity decrease during the SSE is caused by more fluids migrating into the upper plate as the SSE breaks a low-permeability seal on the plate boundary. After the SSE, the fluids in the upper plate diffuse gradually and the velocity increases; The second hypothesis is that before the SSE, elastic strain accumulates causing contraction and reduction of porosity and therefore increase of velocity (the velocity increase between SSEs). During the SSE, the velocity decrease is caused by increased porosity as the SSE relieves the accumulated elastic strain on the plate interface, which results in dilation. After the SSE, stress and strain accumulate again, causing a porosity decrease and a velocity increase back to the original value. This study demonstrates that the velocity variations related to SSEs are observable and provides evidence for slow slip mechanism hypotheses.</b></p> <p>The second project (Chapter 4) focuses on the temporal seismic velocity variations associated with an SSE in 2019 offshore Gisborne, North Island, New Zealand. This is a later SSE in the same area as the first project (Chapters 3). Based on the success of the HOBITSS II, more ocean bottom instruments were deployed in the northern Hikurangi subduction zone from 2018 to 2019 (HOBITSS V). An SSE lasting approximately one month from the end of March to the beginning of May 2019 occurred during the deployment and was recorded by the network. The main slip was south of the deployment and the slip beneath the deployment was up to 150 mm. This study applies coda wave interferometry on the ambient noise data acquired by five OBSs and computes seismic velocity variations to investigate their relation to the SSE. A velocity decrease on the order of 0.015% during the SSE and an increase back to the original velocity value are observed at 1–2.5 s. This supports the two hypotheses proposed in Chapters 3: fluid migration and/or stain changes through the SSE cycle. In addition, velocity variations computed from individual stations show velocity increases before the SSE, which are destructively interfered in their average. Such a situation could occur if the SSE migrated across the network. If the velocity increases before the SSE from individual stations are real, they can be only explained by the hypothesis of crustal strain changes (the second hypothesis in project 1). However, fluid migration (the first hypothesis in project 1) may still happen concomitantly.</p> <p>The third project focuses on the tectonics in southwestern Okinawa Trough offshore northeastern Taiwan. The southwestern Okinawa Trough is an active back-arc basin, extending and rifting within the continental lithosphere. The tectonic development of the back-arc basin is still not well-understood. This study uses continuous ambient noise data recorded by 34 OBSs deployed by Academia Sinica at various periods from 2010 to 2018. Cross-correlations on vertical seismic components and pressure gauges are computed to construct Rayleigh/Scholte waves to study the shear wave velocity structure in the southwestern Okinawa Trough. Phase velocities are measured from the Rayleigh/Scholte waves. Shear velocities are inverted from the phase velocities. Results show the velocity in the south of the back-arc rifting axis near the axis is slower than the velocity in the north of the rifting axis, suggesting the velocity structure in the southwestern Okinawa Trough is asymmetric along the rifting axis. Previous studies have shown high heat flows (about 110mW/m 2 on average) in the south of the rifting axis. The low velocity in the south could be caused by the high heat flow that may be related to asymmetric back-arc extension and/or rifting. This study presents the shear wave velocity structure in the southwest Okinawa Trough is asymmetric along the rifting axis, which implies the back-arc extending/rifting is asymmetric in the study region. This study also suggests effective techniques for OBS noise corrections and unwrapping the cycle skipping of phase velocity measurements.</p> <p>In summary, this thesis represents three projects focusing on seismic velocities in two subduction zones using ambient noise data collected by OBSs. The first and second projects study the temporal velocity variations and the relation to SSEs. Both studies observe velocity decreases during the SSEs and increases after the SSEs, supporting two hypotheses of fluid migration and/or stain changes through the SSE cycle. The third project finds the shear velocity structure in the southwestern Okinawa Trough is asymmetric along the rifting center, which may imply the back-arc extension is asymmetric.</p>

Studies of Seismic Velocities in Subduction Zones from Continuous Ocean Bottom Seismometer Data

<p><b>This thesis uses continuous ambient noise data recorded by Ocean Bottom Seismometers (OBSs) to study seismic velocities in the upper crust of the overriding plate. The first and second projects (Chapters 3 and 4) focus on temporal seismic velocity variations in the northern Hikurangi subduction zone offshore the North Island, New Zealand, while the third project (Chapter 5) focuses on shear wave velocities in the southwestern Okinawa Trough offshore northeastern Taiwan. In the first project (Chapters 3), we investigate a region of frequent slow slip events (SSEs) offshore Gisborne, North Island, New Zealand. From September to October 2014, an SSE occurred with a slip over 250 mm and was recorded successfully by the Hikurangi Ocean Bottom Investigation of Tremor and Slow Slip deployment II (HOBITSS II). We apply coda wave interferometry on the ambient noise data acquired by nine OBSs deployed by the HOBITSS II to study the seismic velocity variations related to the SSE. The average velocity variations display a decrease on the order of 0.05% during the SSE, followed by an increase of similar magnitude afterwards. Two hypotheses are proposed to explain our observation. The first hypothesis, which has been suggested by previous studies, considers that the velocity decrease during the SSE is caused by more fluids migrating into the upper plate as the SSE breaks a low-permeability seal on the plate boundary. After the SSE, the fluids in the upper plate diffuse gradually and the velocity increases; The second hypothesis is that before the SSE, elastic strain accumulates causing contraction and reduction of porosity and therefore increase of velocity (the velocity increase between SSEs). During the SSE, the velocity decrease is caused by increased porosity as the SSE relieves the accumulated elastic strain on the plate interface, which results in dilation. After the SSE, stress and strain accumulate again, causing a porosity decrease and a velocity increase back to the original value. This study demonstrates that the velocity variations related to SSEs are observable and provides evidence for slow slip mechanism hypotheses.</b></p> <p>The second project (Chapter 4) focuses on the temporal seismic velocity variations associated with an SSE in 2019 offshore Gisborne, North Island, New Zealand. This is a later SSE in the same area as the first project (Chapters 3). Based on the success of the HOBITSS II, more ocean bottom instruments were deployed in the northern Hikurangi subduction zone from 2018 to 2019 (HOBITSS V). An SSE lasting approximately one month from the end of March to the beginning of May 2019 occurred during the deployment and was recorded by the network. The main slip was south of the deployment and the slip beneath the deployment was up to 150 mm. This study applies coda wave interferometry on the ambient noise data acquired by five OBSs and computes seismic velocity variations to investigate their relation to the SSE. A velocity decrease on the order of 0.015% during the SSE and an increase back to the original velocity value are observed at 1–2.5 s. This supports the two hypotheses proposed in Chapters 3: fluid migration and/or stain changes through the SSE cycle. In addition, velocity variations computed from individual stations show velocity increases before the SSE, which are destructively interfered in their average. Such a situation could occur if the SSE migrated across the network. If the velocity increases before the SSE from individual stations are real, they can be only explained by the hypothesis of crustal strain changes (the second hypothesis in project 1). However, fluid migration (the first hypothesis in project 1) may still happen concomitantly.</p> <p>The third project focuses on the tectonics in southwestern Okinawa Trough offshore northeastern Taiwan. The southwestern Okinawa Trough is an active back-arc basin, extending and rifting within the continental lithosphere. The tectonic development of the back-arc basin is still not well-understood. This study uses continuous ambient noise data recorded by 34 OBSs deployed by Academia Sinica at various periods from 2010 to 2018. Cross-correlations on vertical seismic components and pressure gauges are computed to construct Rayleigh/Scholte waves to study the shear wave velocity structure in the southwestern Okinawa Trough. Phase velocities are measured from the Rayleigh/Scholte waves. Shear velocities are inverted from the phase velocities. Results show the velocity in the south of the back-arc rifting axis near the axis is slower than the velocity in the north of the rifting axis, suggesting the velocity structure in the southwestern Okinawa Trough is asymmetric along the rifting axis. Previous studies have shown high heat flows (about 110mW/m 2 on average) in the south of the rifting axis. The low velocity in the south could be caused by the high heat flow that may be related to asymmetric back-arc extension and/or rifting. This study presents the shear wave velocity structure in the southwest Okinawa Trough is asymmetric along the rifting axis, which implies the back-arc extending/rifting is asymmetric in the study region. This study also suggests effective techniques for OBS noise corrections and unwrapping the cycle skipping of phase velocity measurements.</p> <p>In summary, this thesis represents three projects focusing on seismic velocities in two subduction zones using ambient noise data collected by OBSs. The first and second projects study the temporal velocity variations and the relation to SSEs. Both studies observe velocity decreases during the SSEs and increases after the SSEs, supporting two hypotheses of fluid migration and/or stain changes through the SSE cycle. The third project finds the shear velocity structure in the southwestern Okinawa Trough is asymmetric along the rifting center, which may imply the back-arc extension is asymmetric.</p>

Studies of Seismic Velocities in Subduction Zones from Continuous Ocean Bottom Seismometer Data

<p><b>This thesis uses continuous ambient noise data recorded by Ocean Bottom Seismometers (OBSs) to study seismic velocities in the upper crust of the overriding plate. The first and second projects (Chapters 3 and 4) focus on temporal seismic velocity variations in the northern Hikurangi subduction zone offshore the North Island, New Zealand, while the third project (Chapter 5) focuses on shear wave velocities in the southwestern Okinawa Trough offshore northeastern Taiwan. In the first project (Chapters 3), we investigate a region of frequent slow slip events (SSEs) offshore Gisborne, North Island, New Zealand. From September to October 2014, an SSE occurred with a slip over 250 mm and was recorded successfully by the Hikurangi Ocean Bottom Investigation of Tremor and Slow Slip deployment II (HOBITSS II). We apply coda wave interferometry on the ambient noise data acquired by nine OBSs deployed by the HOBITSS II to study the seismic velocity variations related to the SSE. The average velocity variations display a decrease on the order of 0.05% during the SSE, followed by an increase of similar magnitude afterwards. Two hypotheses are proposed to explain our observation. The first hypothesis, which has been suggested by previous studies, considers that the velocity decrease during the SSE is caused by more fluids migrating into the upper plate as the SSE breaks a low-permeability seal on the plate boundary. After the SSE, the fluids in the upper plate diffuse gradually and the velocity increases; The second hypothesis is that before the SSE, elastic strain accumulates causing contraction and reduction of porosity and therefore increase of velocity (the velocity increase between SSEs). During the SSE, the velocity decrease is caused by increased porosity as the SSE relieves the accumulated elastic strain on the plate interface, which results in dilation. After the SSE, stress and strain accumulate again, causing a porosity decrease and a velocity increase back to the original value. This study demonstrates that the velocity variations related to SSEs are observable and provides evidence for slow slip mechanism hypotheses.</b></p> <p>The second project (Chapter 4) focuses on the temporal seismic velocity variations associated with an SSE in 2019 offshore Gisborne, North Island, New Zealand. This is a later SSE in the same area as the first project (Chapters 3). Based on the success of the HOBITSS II, more ocean bottom instruments were deployed in the northern Hikurangi subduction zone from 2018 to 2019 (HOBITSS V). An SSE lasting approximately one month from the end of March to the beginning of May 2019 occurred during the deployment and was recorded by the network. The main slip was south of the deployment and the slip beneath the deployment was up to 150 mm. This study applies coda wave interferometry on the ambient noise data acquired by five OBSs and computes seismic velocity variations to investigate their relation to the SSE. A velocity decrease on the order of 0.015% during the SSE and an increase back to the original velocity value are observed at 1–2.5 s. This supports the two hypotheses proposed in Chapters 3: fluid migration and/or stain changes through the SSE cycle. In addition, velocity variations computed from individual stations show velocity increases before the SSE, which are destructively interfered in their average. Such a situation could occur if the SSE migrated across the network. If the velocity increases before the SSE from individual stations are real, they can be only explained by the hypothesis of crustal strain changes (the second hypothesis in project 1). However, fluid migration (the first hypothesis in project 1) may still happen concomitantly.</p> <p>The third project focuses on the tectonics in southwestern Okinawa Trough offshore northeastern Taiwan. The southwestern Okinawa Trough is an active back-arc basin, extending and rifting within the continental lithosphere. The tectonic development of the back-arc basin is still not well-understood. This study uses continuous ambient noise data recorded by 34 OBSs deployed by Academia Sinica at various periods from 2010 to 2018. Cross-correlations on vertical seismic components and pressure gauges are computed to construct Rayleigh/Scholte waves to study the shear wave velocity structure in the southwestern Okinawa Trough. Phase velocities are measured from the Rayleigh/Scholte waves. Shear velocities are inverted from the phase velocities. Results show the velocity in the south of the back-arc rifting axis near the axis is slower than the velocity in the north of the rifting axis, suggesting the velocity structure in the southwestern Okinawa Trough is asymmetric along the rifting axis. Previous studies have shown high heat flows (about 110mW/m 2 on average) in the south of the rifting axis. The low velocity in the south could be caused by the high heat flow that may be related to asymmetric back-arc extension and/or rifting. This study presents the shear wave velocity structure in the southwest Okinawa Trough is asymmetric along the rifting axis, which implies the back-arc extending/rifting is asymmetric in the study region. This study also suggests effective techniques for OBS noise corrections and unwrapping the cycle skipping of phase velocity measurements.</p> <p>In summary, this thesis represents three projects focusing on seismic velocities in two subduction zones using ambient noise data collected by OBSs. The first and second projects study the temporal velocity variations and the relation to SSEs. Both studies observe velocity decreases during the SSEs and increases after the SSEs, supporting two hypotheses of fluid migration and/or stain changes through the SSE cycle. The third project finds the shear velocity structure in the southwestern Okinawa Trough is asymmetric along the rifting center, which may imply the back-arc extension is asymmetric.</p>

GIS Mapping of Short-Term Noisy Event of Diwali Night in Lucknow City

Noise is a universal problem that is particularly prominent in developing nations like India. Short-term noise-sensitive events like New Year’s Eve, derby matches, DJ night, Diwali night (celebration with firecracker) in India, etc. create lots of noise in a short period. There is a need to come up with a system that can predict the noise level for an area for a short period indicating its detailed variations. GIS (Geographic Information System)-based google maps for terrain data and crowd-sourced or indirect collection of noise data can overcome this challenge to a great extent. Authors have tried to map the highly noisy Diwali night for Lucknow, a northern city of India. The mapping was done by collecting the data from 100 points using the noise capture app (30% were close to the source and 70% were away from the source (receiver). Noise data were predicted for 750 data points using the modeling interpolation technique. A noise map is generated for this Diwali night using the crowd-sourcing technique for Diwali night. The results were also varied with 50 test points and are found to be within ±4.4 dB. Further, a noise map is also developed for the same site using indirect data of noise produced from the air pollution open-sourced data. The produced noise map is also verified with 50 test points and found to be ±6.2 dB. The results are also corroborated with the health assessment survey report of the residents of nearby areas.

Unsupervised Clustering of Continuous Ambient Noise Data to Get Higher Signal Quality in Seismic Surveys

Seismic interferometry has been shown to extract body wave arrivals from ambient noise seismic data. However, surface waves dominate ambient noise data, so cross-correlating and stacking all available data may not succeed in extracting body wave arrivals. A better strategy is to find portions of the data in which body wave energy dominates and to process only those portions. One challenge is that passive seismic recordings comprise huge volumes of data, so identifying portions with strong body-wave energy could be difficult or time-consuming. We use spatio-temporal features, calculated with data recorded by all receivers together, to perform unsupervised clustering. Using data recorded by a dense seismic array in Sweetwater, TX we were able to identify five clusters, representing a subsets of the complete dataset that contain similar features, and extract a 7 km/s body wave arrival from one cluster. This arrival did not emerge when we performed the same cross-correlation and stacking regimen on the entire dataset.

Transport receptor occupancy in Nuclear Pore Complex mimics

Nuclear Pore Complexes (NPCs) regulate all molecular transport between the nucleus and the cytoplasm in eukaryotic cells. Intrinsically disordered Phe-Gly nucleoporins (FG Nups) line the central conduit of NPCs to impart a selective barrier where large proteins are excluded unless bound to a transport receptor (karyopherin; Kap). Here, we assess 'Kap-centric' NPC models, which postulate that Kaps participate in establishing the selective barrier. We combine biomimetic nanopores, formed by tethering Nsp1 to the inner wall of a solid-state nanopore, with coarse-grained modeling to show that yeast Kap95 exhibits two populations in Nsp1-coated pores: one population that is transported across the pore in milliseconds, and a second population that is stably assembled within the FG mesh of the pore. Ionic current measurements show a conductance decrease for increasing Kap concentrations and noise data indicate an increase in rigidity of the FG-mesh. Modeling reveals an accumulation of Kap95 near the pore wall, yielding a conductance decrease. We find that Kaps only mildly affect the conformation of the Nsp1 mesh and that, even at high concentrations, Kaps only bind at most 8% of the FG-motifs in the nanopore, indicating that Kap95 occupancy is limited by steric constraints rather than by depletion of available FG-motifs. Our data provide an alternative explanation of the origin of bimodal NPC binding of Kaps, where a stable population of Kaps binds avidly to the NPC periphery, while fast transport proceeds via a central FG-rich channel through lower affinity interactions between Kaps and the cohesive domains of Nsp1.

Improving the accuracy and robustness of RRAM-based in-memory computing against RRAM hardware noise and adversarial attacks

We present a novel deep neural network (DNN) training scheme and RRAM in-memory computing (IMC) hardware evaluation towards achieving high robustness to the RRAM device/array variations and adversarial input attacks. We present improved IMC inference accuracy results evaluated on state-of-the-art DNNs including ResNet-18, AlexNet, and VGG with binary, 2-bit, and 4-bit activation/weight precision for the CIFAR-10 dataset. These DNNs are evaluated with measured noise data obtained from three different RRAM-based IMC prototype chips. Across these various DNNs and IMC chip measurements, we show that our proposed hardware noise-aware DNN training consistently improves DNN inference accuracy for actual IMC hardware, up to 8% accuracy improvement for the CIFAR-10 dataset. We also analyze the impact of our proposed noise injection scheme on the adversarial robustness of ResNet-18 DNNs with 1-bit, 2-bit, and 4-bit activation/weight precision. Our results show up to 6% improvement in the robustness to black-box adversarial input attacks.

Experimental Approach on the Study of Ground-Motion Amplification at the Cerro Prieto Volcano, Mexicali Valley, Baja California, Mexico

We conducted experimental work to explain the large peak ground accelerations observed at the Cerro Prieto volcano in Mexicali Valley, Mexico. Using ambient noise and earthquake data, we compared horizontal-to-vertical spectral ratios (HVSRs) computed for sites on the volcano against those calculated for locations outside it. High-HVSR values (∼11 at ∼2 Hz) were obtained on the top of the volcano at 183 m of altitude, decreasing for sites located at lower elevations. We calculated a median HVSR of ∼1 at 2 Hz from HVSRs computed for nine sites located along an N18°E transect and at an average elevation of ∼25 m. The earlier comparison suggests a relative amplification on the volcano. In addition, we calculated HVSRs from accelerograms generated by 62 earthquakes (2.6≤ML≤5.4; 4.6≤Mw≤7.2) recorded at four locations: two on the volcano (at 194 and 110 m of elevation) and two outside it. These last two sites, located up to 6 km away in a north-northwest and south-southwest direction relative to the volcano, are at an average altitude of 22 m. For the four locations, we also computed the HVSRs from ambient noise data. Although the HVSR results derived from both types of data are slightly different, we also found high HVSRs for the two sites on the volcano and low HVSRs for the two sites outside it, corroborating the relative amplification on the volcano. Using the 1D wave propagation modeling, based on the stiffness matrix method, we modeled the experimental HVSRs to analyze the local site effects. Therefore, we propose that the ground-motion amplification at the Cerro Prieto volcano may be due to a combination of its topography and shallow site effects.

Research on Recommendation of Personalized Exercises in English Learning Based on Data Mining

Aiming at the problems of traditional method of exercise recommendation precision, recall rate, long recommendation time, and poor recommendation comprehensiveness, this study proposes a personalized exercise recommendation method for English learning based on data mining. Firstly, a personalized recommendation model is designed, based on the model to preprocess the data in the Web access log, and cleaning the noise data to avoid its impact on the accuracy of the recommendation results is focused; secondly, the DINA model to diagnose the degree of mastery of students’ knowledge points is used and the students’ browsing patterns through fuzzy similar relationships are clustered; and finally, according to the clustering results, the similarity between students and the similarity between exercises are measured, and the collaborative filtering recommendation of personalized exercises for English learning is realized. The experimental results show that the exercise recommendation precision and recall rate of this method are higher, the recommendation time is shorter, and the recommendation results are comprehensive.