Weld Defects Automatic Visual Recognition by Combined Application of Canny Edge Detector and Interval Fusion with Preference Aggregation; Information, Control, and Communication Technologies (ICCT)

Weld Defects Automatic Visual Recognition by Combined Application of Canny Edge Detector and Interval Fusion with Preference Aggregation; Information, Control, and Communication Technologies (ICCT)

Manylion Llyfryddiaeth
Parent link:Information, Control, and Communication Technologies (ICCT).— 2022.— [4 p.]
Prif Awdur: Muravyov (Murav’ev) S. V. Sergey Vasilyevich
Awdur Corfforaethol: Национальный исследовательский Томский политехнический университет Инженерная школа информационных технологий и робототехники
Awduron Eraill: Nguyen Duc C.
Crynodeb:Title screen
Checking the surface quality of the weld is an important step in nondestructive testing. This inspection can be performed visually, but the accuracy of the results is highly operator-dependent and time-consuming. This paper proposes a new approach to automatic visual recognition of welding defect edges from photographic images. The approach is a combination of the Canny edge detector (CED) and the method of interval fusion with preference aggregation (IF&PA). CED uses first-order derivatives to detect pixels with sudden changes in color intensity and to recognize the contours of objects in the image. For this aim, the CED needs a threshold value to recognize real edges and remove false ones. The IF&PA algorithm is proposed to detect this threshold, where the image of the weld surface is divided into equal bands. For each band, a color intensity histogram is constructed, using which the lower and upper bounds, and the average intensity values are defined for an intensity interval, which characterizes the band. All the intervals are represented by rankings forming a preference profile. Kemeny ranking rule is applied to the profile to determine the best color intensity value, which then acts as the CED threshold. The proposed approach has been tested in detecting the edge of some typical weld defects. The test results show that the proposed approach accurately recognizes the edges of weld defects and eliminates most of the noise in the input image.
Iaith:Saesneg
Cyhoeddwyd: 2022
Pynciau:
электронный ресурс
труды учёных ТПУ
weld defects
image segmentation
Canny edge
detector
interval fusion
preference aggregation
дефекты
сварка
детекторы
Mynediad Ar-lein:https://doi.org/10.1109/ICCT56057.2022.9976559
Fformat: Electronig Pennod Llyfr
KOHA link:https://koha.lib.tpu.ru/cgi-bin/koha/opac-detail.pl?biblionumber=668520

Eitemau Tebyg

Eitemau Tebyg