An end-to-end fault diagnostics method based on convolutional neural network for rotating machinery with multiple case studies

The available stereo matching algorithms produce large number of false positive matches or only produce a few true-positives across oblique stereo images with large baseline. This undesired result happens due to the complex perspective deformation and radiometric distortion across the images. To address this problem, we propose a novel affine invariant feature matching algorithm with subpixel accuracy based on an end-to-end convolutional neural network (CNN). In our method, we adopt and modify a Hessian affine network, which we refer to as IHesAffNet, to obtain affine invariant Hessian regions using deep learning framework. To improve the correlation between corresponding features, we introduce an empirical weighted loss function (EWLF) based on the negative samples using K nearest neighbors, and then generate deep learning-based descriptors with high discrimination that is realized with our multiple hard network structure (MTHardNets). Following this step, the conjugate features are produced by using the Euclidean distance ratio as the matching metric, and the accuracy of matches are optimized through the deep learning transform based least square matching (DLT-LSM). Finally, experiments on Large baseline oblique stereo images acquired by ground close-range and unmanned aerial vehicle (UAV) verify the effectiveness of the proposed approach, and comprehensive comparisons demonstrate that our matching algorithm outperforms the state-of-art methods in terms of accuracy, distribution and correct ratio. The main contributions of this article are: (i) our proposed MTHardNets can generate high quality descriptors; and (ii) the IHesAffNet can produce substantial affine invariant corresponding features with reliable transform parameters.

Download Full-text

Competency-Industry Relatedness (C-IR) Framework for Sustained Business Growth in Startups during and beyond Pandemic: Myths and Lessons from Publicly Funded Innovative Startups

Sustainability ◽

10.3390/su13094632 ◽

2021 ◽

Vol 13 (9) ◽

pp. 4632

Author(s):

Varun Gupta ◽

Luis Rubalcaba

Keyword(s):

Case Studies ◽

Business Environment ◽

Learning Curves ◽

The European Union ◽

Business Growth ◽

Product Lines ◽

Business Decisions ◽

Multiple Case ◽

Multiple Case Studies ◽

Industry Relatedness

Context: The coronavirus disease 2019 (COVID-19) pandemic led to a turbulent business environment, resulting in market uncertainties, frustrations, and rumors. Wrongly held beliefs—or myths—can hinder startups from turning new market opportunities into their favor (for example, by failing at diversification decisions) or undertaking wrong business decisions, e.g., diversifying in industries that have products of no real market value). Objectives: The objective of the paper is to identify the beliefs that drive the business decisions of startups in a pandemic and to isolate those beliefs that are merely myths. Further, this paper proposes strategic guidelines in the form of a framework to help startups make sound decisions that can lead to market success. Method: The two-step research method involved multiple case studies with five startups based in India, France, Italy, and Switzerland, to identify perceptual beliefs that drove strategic business decisions, followed by a case study of 36 COVID-19-solution focused startups, funded by the European Union (EU). The findings were validated through a survey that involved 102 entrepreneurs. The comparative analysis of two multiple case studies helped identify beliefs that were merely “myths”; myths that drove irrational strategic decisions, resulting in business failures. Results: The results indicate that startups make decisions in pandemic situations that are driven by seven myths, pertaining to human, intellectual, and financial resources. The decision on whether to diversify or continue in the same business operation can be divided into four strategic options of the Competency-Industry Relatedness (C-IR) framework: ignore, delay, phase-in, and diversify. Diversification in the same (or different industry) is less risky for startups if they have the skills, as needed, to diversify in related industries. Diversification in related industries helps startups leverage their experiences and learning curves (those associated with existing product lines) to adapt their existing products in new markets, or utilize their technologies to solve new problems via new products. The desired outcome for these startups should be sustainable business growth—to meet sustainability goals by contributing to the society and the economy. Conclusion: The C-IR framework is a strategic guide for startups to make business decisions based on internal factors, rather than myths. Accurately assessing skill diversity and the nature of new industries (or markets) will help startups leverage their existing resources optimally, without the need for (pricey) external funding. This will foster sustained business growth resulting in a nation economic development. Knowledge transfer from the Innovation ecosystem will further strengthen the C-IR framework effectiveness.

Download Full-text

Automatic Driving of End-to-end Convolutional Neural Network Based on MobileNet-V2 Migration Learning

Proceedings of the 12th International Symposium on Visual Information Communication and Interaction - VINCI'2019 ◽

10.1145/3356422.3356458 ◽

2019 ◽

Author(s):

Minghong Hu ◽

Hui Guo ◽

Xuyuan Ji

Keyword(s):

Neural Network ◽

Convolutional Neural Network ◽

Automatic Driving ◽

End To End

Download Full-text

Service Oriented Architecture (SOA) Implementation

International Journal of Information Systems in the Service Sector ◽

10.4018/ijisss.2018040101 ◽

2018 ◽

Vol 10 (2) ◽

pp. 1-21 ◽

Cited By ~ 3

Author(s):

Tom Yoon ◽

Bong-Keun Jeong

Keyword(s):

Case Studies ◽

Service Oriented Architecture ◽

Successful Implementation ◽

Business Units ◽

Small Project ◽

Service Oriented ◽

Multiple Case ◽

Multiple Case Studies ◽

And Training

Using a multiple case studies and surveys, this article finds that factors essential to successful Service Oriented Architecture (SOA) implementations include establishing effective SOA governance, establishing SOA registries, starting with a small project, collaboration between business and IT units, strengthening trust among business units, and training. This article also explores business and IT motivations for SOA implementation and the benefits realized from this implementation. The findings from this article can provide a guidance for practitioners on the successful implementation of SOA.

Download Full-text