The observation of geomagnetic anomalies appearing prior to earthquakes (EQs) is theorized to be generated by the underground seismic processes. However, these pre-EQ anomalies can only provide ‘postdiction’ and are still inadequate for practical applications. So, the present study was conducted to pursue the long-term quest for real EQ prediction models through the adoption of automated machine learning (AutoML), which automates many laborious routines of model development. In this study, more than 50 years geomagnetic field data recorded at 131 magnetometer observatories globally were acquired. Several features were extracted from them through wavelet scattering transform. The features were used as the input to model optimization, of which the strategy for automatic algorithm selection and hyperparameter tuning were performed based on the Asynchronous Successive Halving Algorithm. From the implementation of five classification algorithms, neural network yielded the best performing model with an accuracy of 83.29%. The results showed that practical EQ prediction models could be achievable even for complex systems like lithospheric and seismo-induced geomagnetic processes by employing AutoML.

Traditional monocular visual simultaneous localization and mapping (SLAM) algorithms have been extensively studied and proven to reliably recover a sparse structure and camera motion. Nevertheless, the sparse structure is still insufficient for scene interaction, e.g., visual navigation and augmented reality applications. To densify the scene reconstruction, the use of single-image absolute depth prediction from convolutional neural networks (CNNs) for filling in the missing structure has been proposed. However, the prediction accuracy tends to not generalize well on scenes that are different from the training datasets.In this paper, we propose a dense monocular SLAM system, named DeepRelativeFusion, that is capable to recover a globally consistent 3D structure. To this end, we use a visual SLAM algorithm to reliably recover the camera poses and semi-dense depth maps of the keyframes, and then use relative depth prediction to densify the semi-dense depth maps and refine the keyframe pose-graph. To improve the semi-dense depth maps, we propose an adaptive filtering scheme, which is a structure- preserving weighted average smoothing filter that takes into account the pixel intensity and depth of the neighbouring pixels, yielding substantial reconstruction accuracy gain in densification. To perform densification, we introduce two incremental improvements upon the energy minimization framework proposed by DeepFusion: (1) an improved cost function, and(2) the use of single-image relative depth prediction. After densification, we update the keyframes with two-view consistent optimized semi-dense and dense depth maps to improve pose- graph optimization, providing a feedback loop to refine the keyframe poses for accurate scene reconstruction. Our system outperforms the state-of-the-art dense SLAM systems quantitatively in dense reconstruction accuracy by a large margin.For more information, see the demo video and supplementary material.

Reliable feature correspondence between frames is a critical step in visual odometry (VO) and visual simultaneous localization and mapping (V-SLAM) algorithms. In comparison with existing VO and V-SLAM algorithms, semi-direct visual odometry (SVO) has two main advantages that lead to state-of-the-art frame rate camera motion estimation: direct pixel correspondence and efficient implementation of probabilistic mapping method. This paper improves the SVO mapping by initializing the mean and the variance of the depth at a feature location according to the depth prediction from a single-image depth prediction network. By significantly reducing the depth uncertainty of the initialized map point (i.e., small variance centred about the depth prediction), the benefits are twofold: reliable feature correspondence between views and fast convergence to the true depth in order to create new map points. We evaluate our method with two outdoor datasets: KITTI dataset and Oxford Robotcar dataset. The experimental results indicate that improved SVO mapping results in increased robustness and camera tracking accuracy. The implementation of this work is available at https: //github.com/yan99033/CNN-SVO

In this paper, hybridization of hill-climbing (HC) and elitism (E) with a specially tailored genetic algorithm (GA) for image registration of printed circuit boards (PCBs) placed arbitrarily on a conveyor belt during inspection is proposed to maximize the robustness of the existing framework. These hybrid methods are investigated individually and in combination for accuracy, reliability and performance. Experimental results highlight the potential of the hybrid GA (HGA) that consists of all methods in combination because of the most accurate findings and significantly more reliable than GA alone. However, there is a compensation on performance, though it converges efficiently in terms of number of generations.

This paper presents the implementation of a number of elitist schemes for a low cost printed circuit board (PCB) inspection system. This strategy also aims to explore the role of tournament and roulette-wheel in improving the existing system when using a deterministic selection scheme. In this system, GA is used to detect rotation angle and displacement of PCB placed arbitrarily on a conveyor belt passing under the camera. Deterministic, tournament and roulette-wheel selection scheme have been compared in terms of maximum fitness, rate of accuracy and computation time. The finding shows that deterministic outperformed the other two schemes in all categories and still proves to be an ideal candidate for GA-based PCB inspection system. The modifications on population size and implementation of center block image matching technique also contributed to the improvement of computational time of the system.