Search

In this paper, a scalable end-to-end tree boosting system called XGBoost has been applied for predicting the magnitude of an earthquake from the early part of earthquake waveform data. This model uses the features extracted from the early P wave phase of the records as an input. The model's effectiveness has been verified by using data on earthquakes occurring in the Eurasian plate of Japan Islands from 1996 to 2021. Feature engineering has given 29 new features identified from the early P wave phase of the record, which show a high correlation with the magnitude of an earthquake. The comparison of predicted and actual magnitude shows that a trained XGboost model, which uses a single input record for magnitude prediction, gives an average prediction error of 0.004 ± 0.57 for earthquakes in the test dataset. In contrast, the average prediction error of –1.1 ± 0.80 and –0.65 ± 0.69 has been obtained for the magnitude estimated from conventional τc and Pd methods using the same test dataset. It is further seen that the average predicted magnitude of a single earthquake of magnitude 4.5 and 6.1 (MJMA) obtained by using multiple nearfield records using XGBoost model is 4.58 ± 0.33 and 6.32 ± 0.29, which is close to the actual magnitude of the earthquake. The results presented in this paper clearly show that the structured data can be effectively used by complex machine learning or deep learning models to predict earthquake magnitude from single or multiple records. [ABSTRACT FROM AUTHOR]

Japan is one of the countries with the most frequent natural disasters in the world and is faced with various threats of natural disasters every year, which significantly impact Japan's social economy and people's lives. A great deal of information about disasters is preserved in Japanese literature. Interpreting and organizing this information help us to analyze the regularity of disasters and understand the preventive measures of ancient people. This paper aims to organize, analyze and save disaster data by collecting various information about disasters. Then a disaster spatiotemporal database system is constructed by using deep learning, image processing, and database technology. The system consists of two parts, namely, the disaster database and disaster website. The disaster database is the core of the whole system, which saves the disaster data after organizing and summarizing. The database collects disaster information from various sources, including key information such as disaster type, time, location, scale, and scope of impact. The Disaster website is the system's user interface, providing an interactive platform for users to access and use disaster data easily. The website has many functions, including search, visual display, disaster information query, etc. We also make a detailed analysis of the collected data, aiming to predict the causes and occurrence rules of disasters so as to achieve the target of disaster prediction. [ABSTRACT FROM AUTHOR]

Background: Landslide-affecting factors are uncorrelated or non-linearly correlated, limiting the predictive performance of traditional machine learning methods for landslide susceptibility assessment. Deep learning methods can take advantage of the high-level representation and reconstruction of information from landslide-affecting factors. In this paper, a novel deep learning-based algorithm that combine classifiers of both deep learning and machine learning is proposed for landslide susceptibility assessment. A stacked autoencoder (StAE) and a sparse autoencoder (SpAE) both consist of an input layer for raw data, hidden layer for feature extraction, and output layer for classification and prediction. As a study case, Oda City and Gotsu City in Shimane Prefecture, southwestern Japan, were used for susceptibility assessment and prediction of landslides triggered by extreme rainfall. Results: The prediction performance was compared by analyzing real landslide and non-landslide data. The prediction performance of random forest (RF) was evaluated as better than that of a support vector machine (SVM) in traditional machine learning, so RF was combined with both StAE and SpAE. The results show that the prediction ratio of the combined classifiers was 93.2% for StAE combined with RF model and 92.5% for SpAE combined with RF model, which were higher than those of the SVM (87.4%), RF (89.7%), StAE (84.2%), and SpAE (88.2%). Conclusions: This study provides an example of combined classifiers giving a better predictive ratio than a single classifier. The asymmetric and unsupervised autoencoder combined with RF can exploit optimal non-linear features from landslide-affecting factors successfully, outperforms some conventional machine learning methods, and is promising for landslide susceptibility assessment. [ABSTRACT FROM AUTHOR]

Pear pollination is performed by artificial pollination because the pollination rate through insect pollination is not stable. Pollen must be collected to secure sufficient pollen for artificial pollination. However, recently, collecting sufficient amounts of pollen in Japan has become difficult, resulting in increased imports from overseas. To solve this problem, improving the efficiency of pollen collection and strengthening the domestic supply and demand system is necessary. In this study, we proposed an Artificial Intelligence (AI)-based method to estimate the amount of pear pollen. The proposed method used a deep learning-based object detection algorithm, You Only Look Once (YOLO), to classify and detect flower shapes in five stages, from bud to flowering, and to estimate the pollen amount. In this study, the performance of the proposed method was discussed by analyzing the accuracy and error of classification for multiple flower varieties. Although this study only discussed the performance of estimating the amount of pollen collected, in the future, we aim to establish a technique for estimating the time of maximum pollen collection using the method proposed in this study. [ABSTRACT FROM AUTHOR]

Appearance-based gender identification of the horsehair crab [Erimacrus isenbeckii (Brandt, 1848)] is important for preventing indiscriminate fishing of female crabs. Although their gender is easily identified by visual observation of their abdomen because of a difference in the forms of their sex organs, most of the crabs settle with their shell side upward when placed on a floor, making visual gender identification difficult. Our objective is to use deep learning to identify the gender of the horsehair crab on the basis of images of their shell and abdomen sides. Deep learning was applied to a photograph of 60 males and 60 females captured in Funka Bay, Southern Hokkaido, Japan. The deep learning algorithms used the AlexNet, VGG-16, and ResNet-50 convolutional neural networks. The VGG-16 network achieved high accuracy. Heatmaps were enhanced near the forms of the sex organs in the abdomen side (F-1 measure: 98%). The bottom of the shell was enhanced in the heatmap of a male; by contrast, the upper part of the shell was enhanced in the heatmap of a female (F-1 measure: 95%). The image recognition of the shell side based on a deep learning algorithm enabled more precise gender identification than could be achieved by human-eye inspection. [ABSTRACT FROM AUTHOR]

We present a novel method for coastal tsunami prediction utilizing a denoising autoencoder (DAE) model, one of the deep learning algorithms. Our study focuses on the Tohoku coast, Japan, where dense offshore bottom pressure gauges (OBPGs), called S-net, are installed. To train the model, we generated 800 hypothetical tsunami scenarios by employing stochastic earthquake models (M7.0–8.8). We used synthetic tsunami waveforms at 44 OBPGs as input and the waveforms at four coastal tide gauges as output. Subsequently, we evaluated the model's performance using 200 additional hypothetical and two real tsunami events: the 2016 Fukushima earthquake and 2022 Tonga volcanic tsunamis. Our DAE model demonstrated high accuracy in predicting coastal tsunami waveforms for hypothetical events, achieving an impressive quality index of approximately 90%. Furthermore, it accurately forecasted the maximum amplitude of the 2016 Fukushima tsunami, achieving a quality index of 91.4% at 15 min after the earthquake. However, the prediction of coastal waveforms for the 2022 Tonga volcanic tsunami was not satisfactory. We also assessed the impact of the forecast time window and found that it had limited effects on forecast accuracy. This suggests that our method is suitable for providing rapid forecasts soon after an earthquake occurs. Our research is the first application of an artificial neural network to tsunami prediction using real observations. In the future, we will use more tsunami scenarios for model training to enhance its robustness for different types of tsunamis. [ABSTRACT FROM AUTHOR]

Issuing a disaster certificate, which is used to decide the contents of a victim's support, requires accuracy and rapidity. However, in Japan at large, issuing of damage certificates has taken a long time in past earthquake disasters. Hence, the government needs a more efficient mechanism for issuing damage certificates. This study developed an estimation system of roof-damaged buildings to obtain an overview of earthquake damage based on aero-photo images using deep learning. To provide speedy estimation, this system utilized the trimming algorithm, which automatically generates roof image data using the location information of building polygons on GIS (Geographic Information System). Consequently, the proposed system can estimate, if a house is covered with a blue sheet with 97.57 % accuracy and also detect whether a house is damaged, with 93.51 % accuracy. It would therefore be worth considering the development of an image recognition model and a method of collecting aero-photo data to operate this system during a real earthquake. [ABSTRACT FROM AUTHOR]

In 2017, the Japanese Ophthalmological Society (JOS) created the Japan Ocular Imaging (JOI) registry, a national database of images and clinical data in the field of ophthalmology in Japan. The JOI registry automatically transfers the information stored in the electronic medical records of each institution to the cloud storage governed by the JOS. This process adheres to Japanese laws and regulations such as the Act on the Protection of Personal Information and the relevant laws of Japan, the Ethical Guidelines for Medical and Biological Research Involving Human Subjects, and the Medical Information System Security Management Guideline of Japan. The JOI registry works closely with academia in ophthalmology, as well as the Japan Association for Medical Informatics, the Japan Ophthalmic Instrument Association, and manufacturers of ophthalmic medical devices. The collected data will be used in studies and surveys to improve the quality of ophthalmic care. The JOI registry has been created as a common asset for the entire ophthalmology field, and welcomes participation of related institutions to utilize the infrastructure to promote multicenter joint research and development. The JOI registry network intends to cover all ophthalmological facilities, including clinics. However, the project is currently being conducted with the cooperation of 22 university hospitals, one private hospital, and two health checkup facilities. As of October 2021, six facilities are connected to the JOI registry network; it comprises a total of 486,189 fundus images, 25,224 optical coherence tomography images, and 11,565 visual field results, among others, are stored in the JOI registry. This project was initially funded by a national funding agency, the Japan Agency for Medical Research and Development (AMED) (ICT Infrastructure Establishment and Implementation of Artificial Intelligence for Clinical and Medical Research, 2017–2019). The General Incorporated Association Japan Ocular Imaging Registry was established in 2019, following which the association is managing the JOI registry under the governance of the JOS. The JOI registry is aiming to become a large-scale real-world database comparable to the Intelligent Research in Sight registry of the United States. [ABSTRACT FROM AUTHOR]

In this study, we propose a method to automate fruit harvesting with a fruit harvesting robot equipped with robotic arms. Given the future growth of the world population, food shortages are expected to accelerate. Since much of Japan's agriculture is dependent on imports, it is expected to be greatly affected by this upcoming food shortage. In recent years, the number of agricultural workers in Japan has been decreasing and the population is aging. As a result, there is a need to automate and reduce labor in agricultural work using agricultural machinery. In particular, fruit cultivation requires a lot of manual labor due to the variety of orchard conditions and tree shapes, causing mechanization and automation to lag behind. In this study, a dual-armed fruit harvesting robot was designed and fabricated to reach most of the fruits on joint V-shaped trellis that was cultivated and adjusted for the robot. To harvest the fruit, the fruit harvesting robot uses sensors and computer vision to detect and estimate the position of the fruit and then inserts end-effectors into the lower part of the fruit. During this process, there is a possibility of collision within the robot itself or with other fruits depending on the position of the fruit to be harvested. In this study, inverse kinematics and a fast path planning method using random sampling is used to harvest fruits with robot arms. This method makes it possible to control the robot arms without interfering with the fruit or the other robot arm by considering them as obstacles. Through experiments, this study showed that these methods can be used to detect pears and apples outdoors and automatically harvest them using the robot arms. [ABSTRACT FROM AUTHOR]

In the present study, we propose a new approach for determining earthquake hypocentral parameters. This approach integrates computed theoretical seismograms and deep machine learning. The theoretical seismograms are generated through a realistic three-dimensional Earth model, and are then used to create spatial images of seismic wave propagation at the Earth's surface. These snapshots are subsequently utilized as a training data set for a convolutional neural network. Neural networks for determining hypocentral parameters such as the epicenter, depth, occurrence time, and magnitude are established using the temporal evolution of the snapshots. These networks are applied to seismograms from the seismic observation network in the Hakone volcanic region in Japan to demonstrate the suitability of the proposed approach for locating earthquakes. We demonstrate that the determination accuracy of hypocentral parameters can be improved by including theoretical seismograms for different earthquake locations and sizes, in the learning data set for the deep machine learning. Using the proposed method, the hypocentral parameters are automatically determined within seconds after detecting an event. This method can potentially serve in monitoring earthquake activity in active volcanic areas such as the Hakone region. [ABSTRACT FROM AUTHOR]

The aim of this study is to evaluate and compare the performances of 5 machine learning (ML) techniques for predicting the spatial probability of landslide at Atsuma, Japan, and Mt. Umyeon, Korea. 5 ML models used are Naïve Bayes (NB), ensemble learning (random forest (RF) and adaboost (AB)), and deep learning (multilayer perceptron (MLP) and convolutional neural network (CNN)) models. Real landslide events at the study areas are randomly separated to the training set for landslide mapping and the validation set for assessing performance. To assess the performance of the used models, the resulting models are validated using receiver operating characteristic (ROC) curve. The success rate curves show that the areas under the curve (AUC) for the NB, RF, AB, MLP, and CNN are 85.1, 88.8, 88.6, 87.5, and 95.0%, respectively, at Atsuma and 68.7, 85.6, 90.5, 81.6, and 92.0%, respectively, at Mt. Umyeon. Similarly, the validation results show that the areas under the curve for the NB, RF, AB, MLP, and CNN are 84.3, 87.1, 87.1, 86.7, and 89.7%, respectively, at Atsuma and 64.9, 85.5, 83.9, 84.7, and 90.5%, respectively, at Mt. Umyeon. In addition, statistical tests such as Chi-square test and difference of proportions test show that all classified landslide susceptibility maps have statistical significance and the significant difference in classified landslide susceptibility maps from different ML models. The comparison results among 5 ML models show that the CNN model had the best performance and NB model had the worst performance in both study areas. [ABSTRACT FROM AUTHOR]

In the last fifteen years, tsunami science has progressed at a rapid pace. Three major tsunamis: The Indian Ocean in 2004, the 2011 Tohoku tsunami, and the 2018 Palu tsunami were significant landmarks in the history of tsunami science. All the three tsunamis, as mentioned, suffered from either no warning or poor reception of the alerts issued. Various lessons learned, consequent numerical models proposed, post-2004 tsunami damage findings manifested into solutions. However, the misperceived solutions led to a disastrous impact of the 2011 Tohoku event. In the following years, numerous improvements in warning systems and community preparedness frameworks were proposed and implemented. The contributions and new findings have added multi-fold advancements to tsunami science progress. Later, the 2018 Palu tsunami happened and again led to a massive loss of life and property. The warning systems and community seemed un-prepared for this non-seismic tsunami. A significant change is to take place in tsunami science practices and solutions. The 2018 tsunami is one of the most discussed and researched events concerning the palaeotsunami records, damage assessment, and source findings. In the new era, using machine learning and deep learning prevails in all the fields related to tsunami science. This article presents a complete 15-year bibliometric analysis of tsunami research from Scopus and Web of Science (WoS). The review of majorly cited documents in the form of a progressing storyline has highlighted the need for multidisciplinary research to design and propose practical solutions. [ABSTRACT FROM AUTHOR]

Earthquake signal detection and seismic phase picking are challenging tasks in the processing of noisy data and the monitoring of microearthquakes. Here we present a global deep-learning model for simultaneous earthquake detection and phase picking. Performing these two related tasks in tandem improves model performance in each individual task by combining information in phases and in the full waveform of earthquake signals by using a hierarchical attention mechanism. We show that our model outperforms previous deep-learning and traditional phase-picking and detection algorithms. Applying our model to 5 weeks of continuous data recorded during 2000 Tottori earthquakes in Japan, we were able to detect and locate two times more earthquakes using only a portion (less than 1/3) of seismic stations. Our model picks P and S phases with precision close to manual picks by human analysts; however, its high efficiency and higher sensitivity can result in detecting and characterizing more and smaller events. The authors here present a deep learning model that simultaneously detects earthquake signals and measures seismic-phase arrival times. The model performs particularly well for cases with high background noise and the challenging task of picking the S wave arrival. [ABSTRACT FROM AUTHOR]

P-wave first-motion polarity is the most useful information in determining the focal mechanisms of earthquakes, particularly for smaller earthquakes. Algorithms have been developed to automatically determine P-wave first-motion polarity, but the performance level of the conventional algorithms remains lower than that of human experts. In this study, we develop a model of the convolutional neural networks (CNNs) to determine the P-wave first-motion polarity of observed seismic waveforms under the condition that P-wave arrival times determined by human experts are known in advance. In training and testing the CNN model, we use about 130 thousand 250 Hz and about 40 thousand 100 Hz waveform data observed in the San-in and the northern Kinki regions, western Japan, where three to four times larger number of waveform data were obtained in the former region than in the latter. First, we train the CNN models using 250 Hz and 100 Hz waveform data, respectively, from both regions. The accuracies of the CNN models are 97.9% for the 250 Hz data and 95.4% for the 100 Hz data. Next, to examine the regional dependence, we divide the waveform data sets according to the observation region, and then we train new CNN models with the data from one region and test them using the data from the other region. We find that the accuracy is generally high (≳ 95%) and the regional dependence is within about 2%. This suggests that there is almost no need to retrain the CNN model by regions. We also find that the accuracy is significantly lower when the number of training data is less than 10 thousand, and that the performance of the CNN models is a few percentage points higher when using 250 Hz data compared to 100 Hz data. Distribution maps, on which polarities determined by human experts and the CNN models are plotted, suggest that the performance of the CNN models is better than that of human experts. [ABSTRACT FROM AUTHOR]