A recognition algorithm to detect pipe weld defects

2017 Tehnički Vjesnik  
Preliminary communication Taking magnetic flux leakage (MFL) imaging of pipe weld defects as the research object, a weld defect image recognition algorithm based on greygradient co-occurrence matrix (GGCM) and cluster analysis and mathematical morphology is proposed. Recognition of different types of welding defects was achieved. Firstly, a continuous non-contact scanning MFL system for the pipe weld was used to collect the three-dimensional MFL. Secondly, the three-dimensional MFL signal was
more » ... al MFL signal was converted to a two-dimensional greyscale image. Then the MFL image characteristics of the two-dimensional grayscale image were extracted using GGCM. Based on extracted image features, the characteristic quantity was analysed by using k-means clustering and then through the combination of histogram equalization, Otsu's method of binaryzation, morphologically removing small objects, edge detection, and then structuring a morphologically optimized edge extraction method for edge detection on the grayscale. Through combination of several methods, a new algorithm to improve the detection effect was structured. The results indicated that this algorithm is adaptable and practical. This algorithm solved difficulties associated with the MFL method being used in the weld testing to realize the recognition of pipe weld defects and break through the applicable limitations of traditional signal processing technology. Algoritam raspoznavanja za otkrivanje grešaka u zavarima cijevi Prethodno priopćenje Uzimajući kao objekt istraživanja propuštanje magnetskog toka (MFL) za snimanje grešaka u zavarima cijevi, predložen je algoritam raspoznavanja slika grešaka u zavarima utemeljen na matrici istovremene pojave sivih tonova (GGCM) te klasterskoj analizi i matematičkoj morfologiji. Postignuto je raspoznavanje različitih vrsta grešaka u zavarima. Prvo je korišten sustav kontinuiranog bezdodirnog skeniranja MFL za prikupljanje trodimenzijskog propuštanja magnetskog toka (MFL), a nakon toga je trodimenzijski MFL signal pretvoren u dvodimenzijsku sliku sivih tonova. Zatim su karakteristike MFL slike za dvodimenzijsku sliku u sivim tonovima izdvojene pomoću GGCM. Na temelju izdvojenih značajki slike, analizira se karakteristična količina pomoću particioniranja k-sredine, a zatim kroz kombinaciju izjednačenja histograma, Otsuovu metoda binarizacija, morfološkog uklanjanja malih objekata, otkrivanja rubova, a zatim strukturiranja morfološki optimiziranog algoritma ekstrakcije ruba za otkrivanje rubova na sivim tonovima. Kombiniranjem nekoliko metoda, strukturira se novi algoritam za poboljšanje učinka otkrivanja. Rezultati su pokazali da je ova metoda prilagodljiva i praktična. Ovaj je algoritam riješio poteškoće vezane uz MFL metodu koja se koristi u ispitivanju zavara radi otkrivanja grešaka u zavarima cijevi, a nadilazi granice primjene tradicionalne tehnologije obrade signala. Ključne
doi:10.17559/tv-20170523211205 fatcat:igyhc2spvrd6ranwklulahzmua