scholarly journals Convolutional neural network identification of galaxy post-mergers in UNIONS using IllustrisTNG

2021 ◽  
Vol 504 (1) ◽  
pp. 372-392
Author(s):  
Robert W Bickley ◽  
Connor Bottrell ◽  
Maan H Hani ◽  
Sara L Ellison ◽  
Hossen Teimoorinia ◽  
...  
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Formation Rate ◽  
Hybrid Approach ◽  
Environmental Parameters ◽  
Decision Threshold ◽  
Trade Offs ◽  
Training Samples ◽  
Wide Range ◽  
Order Of Magnitude

ABSTRACT The Canada–France Imaging Survey (CFIS) will consist of deep, high-resolution r-band imaging over ∼5000 deg2 of the sky, representing a first-rate opportunity to identify recently merged galaxies. Because of the large number of galaxies in CFIS, we investigate the use of a convolutional neural network (CNN) for automated merger classification. Training samples of post-merger and isolated galaxy images are generated from the IllustrisTNG simulation processed with the observational realism code RealSim. The CNN’s overall classification accuracy is 88 per cent, remaining stable over a wide range of intrinsic and environmental parameters. We generate a mock galaxy survey from IllustrisTNG in order to explore the expected purity of post-merger samples identified by the CNN. Despite the CNN’s good performance in training, the intrinsic rarity of post-mergers leads to a sample that is only ∼6 per cent pure when the default decision threshold is used. We investigate trade-offs in purity and completeness with a variable decision threshold and find that we recover the statistical distribution of merger-induced star formation rate enhancements. Finally, the performance of the CNN is compared with both traditional automated methods and human classifiers. The CNN is shown to outperform Gini–M20 and asymmetry methods by an order of magnitude in post-merger sample purity on the mock survey data. Although the CNN outperforms the human classifiers on sample completeness, the purity of the post-merger sample identified by humans is frequently higher, indicating that a hybrid approach to classifications may be an effective solution to merger classifications in large surveys.

scholarly journals Automatic Design System with Generative Adversarial Network and Convolutional Neural Network for Optimization Design of Interior Permanent Magnet Synchronous Motor

2021 ◽  
Author(s):  
Yuki Shimizu ◽  
Shigeo Morimoto ◽  
Masayuki Sanada ◽  
Yukinori Inoue
Keyword(s):  
Neural Network ◽  
Prediction Model ◽  
Convolutional Neural Network ◽  
Permanent Magnet ◽  
Optimization Design ◽  
Design System ◽  
Automatic Design ◽  
Wide Range ◽  
Interior Permanent Magnet ◽  
Long Time

The optimal design of interior permanent magnet synchronous motors requires a long time because finite element analysis (FEA) is performed repeatedly. To solve this problem, many researchers have used artificial intelligence to construct a prediction model that can replace FEA. However, because the training data are generated by FEA, it takes a very long time to obtain a sufficient amount of data, making it impossible to train a large-scale prediction model. Here, we propose a method for generating a large amount of data from a small number of FEA results using machine learning. An automatic design system with a deep generative model and a convolutional neural network is then constructed. With its sufficient data, the proposed system can handle three topologies and three motor parameters in a wide range of current vector regions. The proposed system was applied to multi-objective optimization design, with the optimization completed in 13-15 seconds.

scholarly journals Automatic Design System with Generative Adversarial Network and Convolutional Neural Network for Optimization Design of Interior Permanent Magnet Synchronous Motor

2021 ◽  
Author(s):  
Yuki Shimizu ◽  
Shigeo Morimoto ◽  
Masayuki Sanada ◽  
Yukinori Inoue
Keyword(s):  
Neural Network ◽  
Prediction Model ◽  
Convolutional Neural Network ◽  
Permanent Magnet ◽  
Optimization Design ◽  
Design System ◽  
Automatic Design ◽  
Wide Range ◽  
Interior Permanent Magnet ◽  
Long Time

The optimal design of interior permanent magnet synchronous motors requires a long time because finite element analysis (FEA) is performed repeatedly. To solve this problem, many researchers have used artificial intelligence to construct a prediction model that can replace FEA. However, because the training data are generated by FEA, it takes a very long time to obtain a sufficient amount of data, making it impossible to train a large-scale prediction model. Here, we propose a method for generating a large amount of data from a small number of FEA results using machine learning. An automatic design system with a deep generative model and a convolutional neural network is then constructed. With its sufficient data, the proposed system can handle three topologies and three motor parameters in a wide range of current vector regions. The proposed system was applied to multi-objective optimization design, with the optimization completed in 13-15 seconds.

An adaptive convolutional neural network model for ionosphere prediction

2020 ◽  
Author(s):  
Maria Kaselimi ◽  
Nikolaos Doulamis ◽  
Demitris Delikaraoglou
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Electron Density ◽  
Electron Content ◽  
Complex Data ◽  
Navigation Satellite ◽  
Ionospheric Electron Density ◽  
Satellite Systems ◽  
Wide Range ◽  
Adaptive Convolutional Neural Network

<p>Knowledge of the ionospheric electron density is essential for a wide range of applications, e.g., telecommunications, satellite positioning and navigation, and Earth observation from space. Therefore, considerable efforts have been concentrated on modeling this ionospheric parameter of interest. Ionospheric electron density is characterized by high complexity and is space−and time−varying, as it is highly dependent on local time, latitude, longitude, season, solar cycle and activity, and geomagnetic conditions. Daytime disturbances cause periodic changes in total electron content (diurnal variation) and additionally, there are multi-day periodicities, seasonal variations, latitudinal variations, or even ionospheric perturbations that cause fluctuations in signal transmission.</p><p>Because of its multiple band frequencies, the current Global Navigation Satellite Systems (GNSS) offer an excellent example of how we can infer ionosphere conditions from its effect on the radiosignals from different GNSS band frequencies. Thus, GNSS techniques provide a way of directly measuring the electron density in the ionosphere. The main advantage of such techniques is the provision of the integrated electron content measurements along the satellite-to-receiver line-of-sight at a large number of sites over a large geographic area.</p><p>Deep learning techniques are essential to reveal accurate ionospheric conditions and create representations at high levels of abstraction. These methods can successfully deal with non-linearity and complexity and are capable of identifying complex data patterns, achieving accurate ionosphere modeling. One application that has recently attracted considerable attention within the geodetic community is the possibility of applying these techniques in order to model the ionosphere delays based on GNSS satellite signals.</p><p>This paper deals with a modeling approach suitable for predicting the ionosphere delay at different locations of the IGS network stations using an adaptive Convolutional Neural Network (CNN). As experimental data we used actual GNSS observations from selected stations of the global IGS network which were participating in the still-ongoing MGEX project that provides various satellite signals from the currently available multiple navigation satellite systems. Slant TEC data (STEC) were obtained using the undifferenced and unconstrained PPP technique. The STEC data were provided by GAMP software and converted to VTEC data values. The proposed CNN uses the following basic information: GNSS signal azimuth and elevation angle, GNSS satellite position (x and y). Then, the adaptive CNN utilizes these data inputs along with the predicted VTEC values of the first CNN for the previous observation epochs. Topics to be discussed in the paper include the design of the CNN network structure, training strategy, data analysis, as well as preliminary testing results of the ionospheric delays predictions as compared with the IGS ionosphere products.   </p>

scholarly journals Detection of Dense Small Rigid Targets Based on Convolutional Neural Network and Synthetic Images

2021 ◽  
Vol 38 (1) ◽  
pp. 61-71
Author(s):  
Xianrong Zhang ◽  
Gang Chen
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Synthetic Image ◽  
Initial Training ◽  
Image Processing Algorithm ◽  
Training Set ◽  
Network Training ◽  
Training Samples ◽  
Image Set ◽  
Synthetic Images

Facing the image detection of dense small rigid targets, the main bottleneck of convolutional neural network (CNN)-based algorithms is the lack of massive correctly labeled training images. To make up for the lack, this paper proposes an automatic end-to-end synthesis algorithm to generate a huge amount of labeled training samples. The synthetic image set was adopted to train the network progressively and iteratively, realizing the detection of dense small rigid targets based on the CNN and synthetic images. Specifically, the standard images of the target classes and the typical background mages were imported, and the color, brightness, position, orientation, and perspective of real images were simulated by image processing algorithm, creating a sufficiently large initial training set with correctly labeled images. Then, the network was preliminarily trained on this set. After that, a few real images were compiled into the test set. Taking the missed and incorrectly detected target images as inputs, the initial training set was progressively expanded, and then used to iteratively train the network. The results show that our method can automatically generate a training set that fully substitutes manually labeled dataset for network training, eliminating the dependence on massive manually labeled images. The research opens a new way to implement the tasks similar to the detection of dense small rigid targets, and provides a good reference for solving similar problems through deep learning (DL).

scholarly journals Divide-and-Conquer Dual-Architecture Convolutional Neural Network for Classification of Hyperspectral Images

2019 ◽  
Vol 11 (5) ◽  
pp. 484 ◽  
Author(s):  
Jie Feng ◽  
Lin Wang ◽  
Haipeng Yu ◽  
Licheng Jiao ◽  
Xiangrong Zhang
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Classification Performance ◽  
Hyperspectral Images ◽  
Divide And Conquer ◽  
Fixed Size ◽  
Edge Preservation ◽  
Fine Grained ◽  
Training Samples ◽  
Spatial Window

Convolutional neural network (CNN) is well-known for its powerful capability on image classification. In hyperspectral images (HSIs), fixed-size spatial window is generally used as the input of CNN for pixel-wise classification. However, single fixed-size spatial architecture hinders the excellent performance of CNN due to the neglect of various land-cover distributions in HSIs. Moreover, insufficient samples in HSIs may cause the overfitting problem. To address these problems, a novel divide-and-conquer dual-architecture CNN (DDCNN) method is proposed for HSI classification. In DDCNN, a novel regional division strategy based on local and non-local decisions is devised to distinguish homogeneous and heterogeneous regions. Then, for homogeneous regions, a multi-scale CNN architecture with larger spatial window inputs is constructed to learn joint spectral-spatial features. For heterogeneous regions, a fine-grained CNN architecture with smaller spatial window inputs is constructed to learn hierarchical spectral features. Moreover, to alleviate the problem of insufficient training samples, unlabeled samples with high confidences are pre-labeled under adaptively spatial constraint. Experimental results on HSIs demonstrate that the proposed method provides encouraging classification performance, especially region uniformity and edge preservation with limited training samples.

Sign in / Sign up

Export Citation Format

Share Document