Research on Automatic Crack Detection for Concrete Infrastructures Using Image Processing and Deep Learning | Chapter 6 | Current Approaches in Science and Technology Research Vol. 3

 Automatic crack detection is a critical task in the generation of a crack map for existing concrete infrastructure inspection. This paper describes an automatic crack detection and classification method based on a genetic algorithm (GA) for optimizing image processing technique parameters (IPTs). Under various complex photometric conditions, the crack detection results of concrete infrastructure surface images remain noise pixels. Following that, a deep convolution neural network (CNN) method is used to automatically classify crack candidates and non-crack candidates. Furthermore, the proposed method is compared to state-of-the-art crack detection methods. The experimental results validate the reasonable accuracy in practice. The final goal was to create a crack map, which necessitated automatic pixel-level accuracy.

Author(s) Details

Cuong Nguyen Kim
Faculty of Highway & Bridge, Mien Trung of Civil Engineering, Vietnam.

Kei Kawamura
Graduate School of Science & Technology for Innovation, Yamaguchi University, Japan.

Hideaki Nakamura
Graduate School of Science & Technology for Innovation, Yamaguchi University, Japan.

Amir Tarighat
Department of Civil Engineering, Shahid Rajaee Teacher Training University, Iran.

View Book :- https://stm.bookpi.org/CASTR-V3/article/view/1411

Deep Neural Networks for Multilingual Machine Translation | Chapter 7 | Advanced Aspects of Engineering Research Vol. 4

 We are facing an informational flood of various languages as a result of the advent of information and communication technology and the democratisation of Internet content creation. Newspapers and news portals both contribute significantly to this material, making it difficult to understand the sheer volume of daily circulating data on the web or in print. Manually managing and translating is admittedly difficult, and even traditional IT tools fail to produce results in all languages. To address this problem, we propose experimenting with deep learning for multilingual machine translation. This paper describes and evaluates our newly developed neural text-to-text translation method. The corpora elaboration and two deep neural processing modules for machine translation were used to create this framework. The device provides users with an ergonomic interface that displays the corresponding translated sentences to the sentences they fed it.

Author (s) Details

Meryem Boukrissa
Laboratory LRIT, Faculty of Science Rabat, University Mohammed V, Rabat, Morocco.

Dr. Fadoua Ataa Allah
Computer Science Studies, Information Systems and Communication Center, Royal Institute of the Amazigh Culture, Rabat, Morocco.

View Book :- https://stm.bookpi.org/AAER-V4/article/view/640