Sub-pixel channel boundary delineation based on partial area effect
عنوان دوره: ششمین همایش ژئوفیزیک اکتشافی نفت
کد مقاله : 2150-NIGS
نویسندگان
1دانشکده مهندسی نفت، دانشگاه صنعتی امیرکبیر، تهران، ایران
2استاد دانشکده مهندسی نفت دانشگاه صنعتی امیرکبیر و استاد بازنشسته موسسه ژئوفیزیک دانشگاه تهران
3استاد گروه آموزشی آمار، دانشکده ریاضی و علوم کامپیوتر، دانشگاه صنعتی امیرکبیر
4استاد گروه آموزشی مخابرات دانشکده مهندسی برق دانشگاه امیرکبیر
چکیده
Considering buried channels as prevalent stratigraphic traps is indispensable in petroleum exploration and drilling. This article presents a novel seismic attribute that employs the partial area effect strategy to detect individual discontinuities, especially channel boundaries. We tested the recommended method on a simulated time slice with meandering and branching sinusoidal channels of different sinuosity and thickness and on field data sets from southwest Iran. We evaluated the effectiveness of the partial area effect algorithm against well-known edge detectors and seismic attributes. The results revealed that the designated algorithm could produce a more accurate map of channel boundaries with reduced influence from random noise. Quantitative validation tests, including localization error, precision, sensitivity, and error rate, likewise confirmed the superiority of the proposed sub-pixel procedure.
کلیدواژه ها
Channel boundaries؛ Classic edge detectors؛ seismic attributes؛ Partial area effect؛ Sub-pixel resolution
Title
Sub-pixel channel boundary delineation based on partial area effect
Authors
Masoume Lotfi, Abdolrahim Javaherian, Saeid Rezakhah Varnousfaderani, Hamidreza Amindavar
Abstract
Considering buried channels as prevalent stratigraphic traps is indispensable in petroleum exploration and drilling. This article presents a novel seismic attribute that employs the partial area effect strategy to detect individual discontinuities, especially channel boundaries. We tested the recommended method on a simulated time slice with meandering and branching sinusoidal channels of different sinuosity and thickness and on field data sets from southwest Iran. We evaluated the effectiveness of the partial area effect algorithm against well-known edge detectors and seismic attributes. The results revealed that the designated algorithm could produce a more accurate map of channel boundaries with reduced influence from random noise. Quantitative validation tests, including localization error, precision, sensitivity, and error rate, likewise confirmed the superiority of the proposed sub-pixel procedure.
Keywords
Channel boundaries, Classic edge detectors, seismic attributes, Partial area effect, Sub-pixel resolution