Scribd
Upload
71 views2 pages

Chapter 10: Conclusion and Future Work

This document discusses techniques for non-destructive testing to detect defects in welding without damaging the internal structure. It presents six novel approaches for classifying weld defects in industrial radiography images using image processing and clustering techniques. The results show these techniques robustly detect different types of weld flaws with a good detection rate. Future work includes simplifying pre-processing, segmentation, and feature extraction techniques to reduce computational complexity and allow faster, more accurate defect recognition.

Uploaded by

murali
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
71 views2 pages

Chapter 10: Conclusion and Future Work

This document discusses techniques for non-destructive testing to detect defects in welding without damaging the internal structure. It presents six novel approaches for classifying weld defects in industrial radiography images using image processing and clustering techniques. The results show these techniques robustly detect different types of weld flaws with a good detection rate. Future work includes simplifying pre-processing, segmentation, and feature extraction techniques to reduce computational complexity and allow faster, more accurate defect recognition.

Uploaded by

murali
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

142

CHAPTER 10: CONCLUSION AND FUTURE WORK

Non Destructive Testing (NDT) is a technique for damage

assessment, disaster prediction and quality control, to detect the defects

without affecting the internal structure. This thesis presents and

proposes some novel techniques for weld flaw classification from

industrial radiography for improving the safety of nuclear power plant,

petrochemical industries etc using image processing and clustering

techniques. Six different novel approaches are conducted and

documented in this thesis for the classification of the weld defect along

with a broad literature survey of various techniques conducted by

numerous researchers in this field. Weld defect classification in

radiographic images using Fuzzy C-Means clustering and Zernike

moments, PCA and K-Means clustering based weld defect identification

from radiographic images, Weld defect recognition in radiography based

on Projection Profile and RST invariant by using LVQ, Detection using

Image reconstruction by Simultaneous Algebraic Reconstruction

Technique (SART), An efficient fast processing Adaptive Median filter

based on enhanced Dijkstras 3-way partitioning is developed in this

thesis, Weld flaw identification from radiographic weld images using

Radon Transform and improved Fuzzy C-Means clustering. This thesis

presents novel approaches for the improvement of automatic


143

classification and judgement of discontinuities or defects in welding. The

result shows that these above techniques are robust and provide a good

detection rate for different types of weld flaws. The future scope of this

research work lies in the field of pre-processing, segmentation and

feature extraction. The segmentation and feature extraction techniques

like Watershed, Hough Transform and Zernike Moments respectively are

mathematically and computationally complex due to the morphological

operations, parametric plane conversions and orthogonal projections

respectively, thereby consuming more execution time. So there is always

a future scope for making the image enhancement, segmentation and

feature extraction techniques simpler and more effective by reducing

computational complexity which will help in faster and more accurate

recognition of weld defects from radiographic images.

You might also like