Neural guided visual slam system with Laplacian of Gaussian operator

Abstract

Simultaneous localization and mapping (SLAM) addresses the problem of constructing the map from noisy sensor data and tracking the robot's path within the built map. After decades of development, a lot of mature systems achieve competent results in feature-based implementations. However, there are still problems when migrating the technology to practical applications. One typical example is the accuracy and robustness of SLAM in environment with illuminance and texture variations. To this end, two modules in the existing systems are improved here namely tracking and camera relocalization. In tracking module, image pyramid is processed with Laplacian of Gaussian (LoG) operator in feature extraction for enhanced edges and details. A majority voting mechanism is proposed to dynamically evaluate and redetermine the zero-mean sum of square difference threshold according to the matching error estimation in patch search. In camera relocalization module, full convolutional neural network which focuses on certain parts of the input data is utilized in guiding for accurate output predictions. The authors implement the two modules into OpenvSLAM and propose a neural guided visual SLAM system named LoG-SLAM. Experiments on publicly available datasets show that the accuracy and efficiency increase with LoG-SLAM when compared with other feature-based methods, and relocalization accuracy also improves compared with the recently proposed deep learning pipelines. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

Narrow gap welding seam deflection correction study based on passive vision

“…The average greyness a is calculated using equation (5): The expression for the greyscale variance is given in equation (6): Gaussian of Laplacian (LOG) The Laplace algorithm is used to detect the two-dimensional isotropic measure of the second-order spatial derivative of the image for the interference of scratches and sections. Weld features are distinguished for the difference between scratches/sections and welds that undergo rapid changes in intensity in the image (Ge et al , 2021; Ou et al , 2018). The image is smoothed with a Gaussian smoothing filter prior to the Laplacian operation to reduce the sensitivity of the Laplacian operation to noise.…”

“…In narrow gap welding, the torch oscillates laterally in the weld path to increase the melt pool coverage area and improve the quality of the weld. Because narrow gap welding is mostly used within precision fields such as aerospace and nuclear power stations, a high level of precision is required (Ge et al, 2021;KyungEun et al, 2023;Chen et al, 2023). The current research trend is how to quickly and accurately identify whether there is a deviation in position between the weld seam and the welding torch.…”