Pencarian gambarmenggunakan keyword berupa teks telah dirasakan kurang efektif. Hal inidisebabkan karena adanya batasan kemampuan teks dalam mewakili keseluruhan isidari gambar, terutama pada basisdata gambar yang besar. Keterbatasan tersebutmeliputi penilaian yang subjektif dalam mengartikan gambar dan pemberian namaberkas gambar yang belum tentu dapat mendeskripsikan isi gambar sepenuhnya.Pendekatan lain yang dilakukan dalam pencarian gambar adalah berdasarkan isidari gambar (content based imageretrieval). Penelitian ini membangun sebuah aplikasi untuk mencari gambarmelalui pendekatan content based imageretrieval dengan menggunakan kombinasi fitur warna dan tekstur. Fitur warnadiperoleh dengan menggunakan algoritma colormoment berdasarkan distribusi warna, yaitu nilai mean, variance dan skewness. Terdapat dua cara untukmendapatkan fitur warna yaitu secara global (whole) dan berdasarkan region.Fitur tekstur diperoleh dengan menggunakan algoritma Gabor texture. Fitur warna dan tekstur juga dikombinasikan untukmengetahui kemampuannya dalam proses pencarian gambar. Proses pengukurankemiripan gambar dihitung dengan menggunakan Conberra Distance. Hasil evaluasi diperoleh dengan membandingkannilai presisi dan recall pada saat proses pencarian gambar pada dataset. Hasileksperimen menunjukkan bahwa kombinasi colormoment region dan gabor texturedapat menampilkan hasil pencarian gambar yang lebih relevan yang ditunjukkandengan nilai presisi dan recall yang lebih tinggi dibandingkan dengan kombinasi lainnya.
The Short Term Averaging/Long Term Averaging (STA/LTA) has been widely used to detect earthquake arrival time. The method simply calculates the ratio of moving average of the waveform amplitude at short and long-time windows. However, although STA/LTA signals can distinguish between real events and noise, we still recognize some lack of accuracies in first P wave arrival pickings. In this study, we attempt to implement one machine learning method popularly, Artificial Neural Network (ANN) that employ input, hidden and output layer similar as human brain works. Note that in this study, we also try to add input parameters with another derivative signal attributes such as Recursive STA/LTA and Carl STA/LTA. The processing step started by collecting event waveforms from the Agency of Meteorology, Climatology and Geophysics. We chose regional events with moment magnitude higher than 3 in the Maluku region Indonesia. Next, we apply all STA/LTA attributes to the input waveforms. We also tested our STA/LTA with synthetic data and additional noise. Further step, we manually picked the arrival of P wave events and used this as the output for ANN. In total, we used 100 events for arrival data training in P wave phases. In the validation process, an accuracy of more than 0.98 can be obtained after 200 iterations. Final outputs showed, that in average, the difference between manual picking and automatic picking from ANN is 0.45 s. We are able to increase the accuracy by band pass filter (0.1 – 3 Hz) all signal and improve the mean into 0.15s difference between manual picking and ANN picks.
The traditional method in determining first arrival time of earthquake dataset is time consuming process due to waveform manual inspection. Additional waveform attributes can help determine events detection. One of the widely used attribute is The Short Term Averaging/Long Term Averaging (STA/LTA) which is simply division moving average of waveform amplitude between short time and longer time. Alternatively, waveform attribute can also be built using kurtosis and skewness. The kurtosis attribute is defined as sharpness of data distribution. By definition, the maximum signal should be at or close to the P wave arrival. The skewness is typically used to show normal distribution of the data. The uniqueness of this method is that it has an ability to determine whether the first P wave arrival has positive of negative number. The skewness calculation is similar to kurtosis but it uses the power of 3 instead of 4. With the objective of generating efficient scheme to pick first time arrival, we explore use artificial neural network and a combination of kurtosis and skewness attributes. We use a collection of magnitude events with moment magnitude larger than 3 located close to Moluccas island, Indonesia. We collected all events information from the Indonesian Agency of Meteorology, Climatology and Geophysics. The process is started with manually pick all P wave arrivals using manual inspection. Next, we trained the artificial neural network with attributes numbers as inputs and arrival time we manually picked as the output. In total we used 100 regional events that has clear P wave phases. Then, we validated the results until reaching 0.99 accuracy. In the last step, we tested the once trained procedures on new waveforms contained events. Current result shows an average of 0.4s different between manual pick and trained picked from machine learning. The accuracy can be improved by applying a band pass 0.1-2 Hz filtering with an average of 0.2s.
The tectonic setting of Java Island is mainly controlled by the collision of Indo-Australian plate subducting the Eurasian plate. The high collision activity of Eurasian and Indo-Australian plates often causes megathrust earthquakes and the rise of arc magmatism that includes volcanic eruption. This study aims to determine the tectonic pattern beneath Central Java based on P-wave tomography inversion. We used the fast-marching method as ray tracing and subspace inversion to image subsurface velocity model to a depth of 150 km. The data used in this study are catalogue events data derived from a temporary seismometer network MERAMEX installed around central Java and DOMERAPI installed surround Mt. Merapi and Mt. Merbabu. We also include events collected from the International Seismological Centre. In total, we processed 563 earthquake events to illustrate velocity structures under central Java. The checker-board model shows that good resolutions can be identified at shallow depth, including offshore south Java contributed from Ocean Bottom Seismometer data. In vertical axis, good resolution models can be expected down to a depth 150 km following rich events from the Benioff zone. Current P wave model show a distinct low velocity zone under Mt Merapi that can be seen down to a depth of 40 km, suggesting a possible separated deep magma reservoir. To the south of Mt Merapi area also shows a low-velocity band that may be related with the southern mountain arc. Additionally, the northern part of Mt. Merapi displays a band of strong low-velocity anomaly to the East and West with the anomaly in the Eastern Part seems to have a deeper extension to a depth of ~50 km. We related this anomaly with Merapi Lawu Anomaly and Kendeng basin. Our results show a similar result with the previous tomography models in this region.
Our study area is located near island Sumbawa, Sumba, Flores, West Timor, Indonesia and East Timor, popularly known as Sunda-Banda arc transition zone. The tectonic setting is mainly controlled by the movement of the oceanic lithosphere Indo-Australian plate subducting the Eurasian plate and Northward migration of Australian continental lithosphere into western Banda-arc in the region of Flores, Sumba and Timor island. We tried to image velocity structure beneath these regions using regional events and tomography inversion model. We collected 5 years of regional events from the Indonesian Agency of Meteorology, Climatology and Geophysics. In total, we reserved 3186 events recorded on 29 stations. For data processing, we used fast marching method as ray tracing between sources and receiver. We then employed subspace inversion as the tomography procedure to estimate the best velocity model representing the tectonic model in the region. Hypocenter data distribution is concentrated on shallow parts of the region and along the Benioff zone down to a maximum depth of 400 km. One of challenge of this study is that although events are abundance, the stations used are mostly located onshore and does not extend in the south-north direction that leads us to under determined problem in the inversion process. However, checker-board models show most our target area can be retrieved to its initial model with sign of smearing effects shown start from a depth of 50 km. After six iteration and optimized selection of damping and smoothing parameters, we observed low velocity anomaly under Bali, Lombok, Sumba, East Nusa Tenggara at shallow depth that may be related with volcanic activity. Deeper low anomaly can also be seen that may be related with partial melting process. A band of fast velocity is clearly seen that goes deepen to the north depicting subducting slabs own to a depth of 300 km. We also observed a possible of fast velocity in the northern part of our stations at shallow depth that we believe may represent the back arc thrust.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.