Menu
Search
Filters
Close
Menu

A Machine Vision Case Study Of U-Net Networks For Superficial Corrosion And Dirt Image Segmentation On Industrial Coated Steel Structures

A MACHINE VISION CASE STUDY OF U-NET NETWORKS FOR SUPERFICIAL CORROSION AND DIRT IMAGE SEGMENTATION ON INDUSTRIAL COATED STEEL STRUCTURES
Available for download

Convolutional deep neural networks are one of the main machine learning techniques applied to computer vision and object recognition tasks. Currently, they are very popular due to their proven effectiveness in solving image classification tasks and their significant theoretical and practical importance to the advancement of the deep learning field. Examples of successful image classification networks developed are AlexNet, VGG, and GoogLeNet.1,2,3 

Product Number: 51322-17995-SG
Author: Wesley Kaizer, Otavio Correa
Publication Date: 2022
Industry: Corrosion Monitoring and Control
$0.00
$20.00
$20.00

Corrosion detection in industrial assets and components is an important broad problem in the industries since it allows the temporal tracking of possible issues and the execution of preventive maintenance actions, such as protective coating. However, solving this problem using modern machine learning methods usually demands a careful design of artificial intelligence tools, such as neural networks, high computational resources for training and inference, and a large and adequate dataset. In this work we investigate the application of deep convolutional neural networks to the problem of image semantic segmentation of superficial corrosion and dirt present in mining industrial assets, using a set of images collected in place by corrosion inspectors and manually labeled by a data team. We compare two networks based on the popular U-Net model, in which one of them uses the transferred features from a pre-trained VGG-16 image classification model.

Corrosion detection in industrial assets and components is an important broad problem in the industries since it allows the temporal tracking of possible issues and the execution of preventive maintenance actions, such as protective coating. However, solving this problem using modern machine learning methods usually demands a careful design of artificial intelligence tools, such as neural networks, high computational resources for training and inference, and a large and adequate dataset. In this work we investigate the application of deep convolutional neural networks to the problem of image semantic segmentation of superficial corrosion and dirt present in mining industrial assets, using a set of images collected in place by corrosion inspectors and manually labeled by a data team. We compare two networks based on the popular U-Net model, in which one of them uses the transferred features from a pre-trained VGG-16 image classification model.

Product tags
Also Purchased
Picture for A Further Look At The Impacts Of Corrosion Inhibitor On Scale Prevention
Available for download

A Further Look At The Impacts Of Corrosion Inhibitor On Scale Prevention

Product Number: 51322-17676-SG
Author: Qiwei Wang, Tao Chen, Hameed Al-Badairy, Sultan Alsubaie, Yasser Al-Jeshi, Mohammad Alsidan
Publication Date: 2022
$20.00

Scale and corrosion inhibitors are commonly used in many oil and gas production systems to prevent inorganic deposition and to protect asset integrity. Scale inhibitor products are based on organic compounds with phosphate or carboxylic functional groups such as amino phosphonates, phosphate esters, phosphino polymers, polycarboxylate and polysulfonates,1 as shown in Figure 1. These anionic groups have strong affinity to alkaline earth cations and can adsorb on the active growth sites of scale crystal (Figure 2), resulting in stopping or delaying the scale formation process.
Picture for A Case Study-Caustic Gouging Of Boiler Tubes
Available for download

A Case Study-Caustic Gouging Of Boiler Tubes

Product Number: 51322-17780-SG
Author: Suresh Divi, Sri Krishna Chimbli
Publication Date: 2022
$20.00

Caustic corrosion is sometimes referred to as “caustic attack or “caustic gouging.” Corrosion of this type may result from internally fouled heat transfer surfaces and the presence of sodium hydroxide in the boiler water; and concentrated solutions of alkali where the normal washing of the tube metal ID is restricted after Departure from Nucleate Boiling (DNB), i.e., when the steam bubble release exceeds the rinsing rate.
A Case Study On Microbiologically Influenced Corrosion In Oil Producing Well Flowlines
Available for download

A Case Study On Microbiologically Influenced Corrosion In Oil Producing Well Flowlines

Product Number: 51322-17924-SG
Author: Amer Jarragh, Sandip Kuthe, Sharad Londhe, Akhil Jaithlya, Abeer Rashed, Nadine El-Kaouri
Publication Date: 2022
$20.00

Oil field operating company’s (1) flowline network in North and West Kuwait (NWK) has over 3000 wells connected through 6” carbon steel pipelines flowing from wellhead to the nearest Gathering Center (GC). Untreated wet crude is transported through the flowlines to GC’s directly or passing through the Remote Headers and Manifold (RHM) to GCs. In RHM, mixing of the wet crude takes place before it is sent to GC’s via transfer lines for further separation.  The flowlines are laid aboveground except at road crossings where they are buried. 
Categories
Industry
Recently viewed
Popular tags
View all

Association for Materials Protection and Performance

© AMPP All Rights Reserved.
Membership Terms of Content Use
Contact Customer Support