With the development of remote sensing technology and the advent of high-resolution images, obtaining data has become increasingly convenient. However, the acquisition of small manhole cover information still has shortcomings including low efficiency of manual surveying and high leakage rate. Recently, deep learning models, especially deep convolutional neural networks (DCNNs), have proven to be effective at object detection. However, several challenges limit the applications of DCNN in manhole cover object detection using remote sensing imagery: (1) Manhole cover objects often appear at different scales in remotely sensed images and DCNNs’ fixed receptive field cannot match the scale variability of such objects; (2) Manhole cover objects in large-scale remotely-sensed images are relatively small in size and densely packed, while DCNNs have poor localization performance when applied to such objects. To address these problems, we propose an effective method for detecting manhole cover objects in remotely-sensed images. First, we redesign the feature extractor by adopting the visual geometry group (VGG), which can increase the variety of receptive field size. Then, detection is performed using two sub-networks: a multi-scale output network (MON) for manhole cover object-like edge generation from several intermediate layers whose receptive fields match different object scales and a multi-level convolution matching network (M-CMN) for object detection based on fused feature maps, which combines several feature maps that enable small and densely packed manhole cover objects to produce a stronger response. The results show that our method is more accurate than existing methods at detecting manhole covers in remotely-sensed images.
Purpose
This paper aims to provide the proper preset temperatures of the convection reflow oven when reflowing a printed circuit board (PCB) assembly with varied sizes of components simultaneously.
Design/methodology/approach
In this study, computational fluid dynamics modeling is used to simulate the reflow soldering process. The training data provided to the machine learning (ML) model is generated from a programmed system based on the physics model. Support vector regression and an artificial neural network are used to validate the accuracy of ML models.
Findings
Integrated physical and ML models synergistically can accurately predict reflow profiles of solder joints and alleviate the expense of repeated trials. Using this system, the reflow oven temperature settings to achieve the desired reflow profile can be obtained at substantially reduced computation cost.
Practical implications
The prediction of the reflow profile subjected to varied temperature settings of the reflow oven is beneficial to process engineers when reflowing bulky components. The study of reflowing a new PCB assembly can be started at the early stage of board design with no need for a physical profiling board prototype.
Originality/value
This study provides a smart solution to determine the optimal preset temperatures of the reflow oven, which is usually relied on experience. The hybrid physics–ML model providing accurate prediction with the significantly reduced expense is used in this application for the first time.
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.