การสกัดกลุ่มแนวรอยเลื่อนพะเยาแบบกึ่งอัตโนมัติโดยใช้ภาพถ่ายดาวเทียม

• Main Author: ณฏั ฐ์สุขพรสวรรค์
• Other Authors: ผู้ช่วยศาสตราจารย์ดร.ชนิดาสุวรรณประสิทธ์
• Format: Theses and Dissertations
• Language: other
• Published: เชียงใหม่ : บัณฑิตวิทยาลัย มหาวิทยาลัยเชียงใหม่ 2020
• Online Access: http://cmuir.cmu.ac.th/jspui/handle/6653943832/69551
• Institution: Chiang Mai University

The objective of the study of semi-automatic Phayao-Fault zone extraction by satellite imagery are 1) to select the suitable method for lineament extraction, 2) to select of the lineaments by comparison and 3) to analyze relationship between fault line and physical characteristics in the Phayao-Fault zone are using Landsat 8 OLI which acquired on 15 February 2016 and 22 February 2016 The process of the study included satellite images improvement comparison using 3*3 pixels windows size filtering between Sobel Filter, Laplacian Filter, Directional Filter, Principal Component Analysis, and LD_B7 which was a method developed in this study and lineaments extraction using Canny Algorithm the previous step. The process of lineament classification from digital elevation model, slope, 3Dof digital elevation model, satellite images, improved satellite image, Google Map, geological structure, and surveys dataset. Finally, the relationship analyzing between extracted fault lines and physical characteristics of the study area. The result of study showed that LD_B7 did the better result than other techniques in study for satellite images, improvement and lineament extraction with 3*3 pixels window size filtering in SWIR2 band. It was proved by accuracy assessment using 4 method including visual interpretation, classified with overlay dataset, classified with lineament sampling collection, and sensitivity assessment using different date extraction. The result of classified lineament from LD_B7 found the fault lines appeared on the extracted lineaments. There were extracted fault lines in each direction including 1) in north-south direction 408 lines with 17.65% of north-south direction, 2) in east-west direction 480 lines with 20.99% of east-west direction, 3) in northeast-southwest direction 330 lines with 14.31% of northeast-southwest direction and 4) in northwest-southeast direction 313 lines with 13.67% of northwest-southeast direction. The most fault lines found in north-south direction. Along with other lineaments from agriculture field boundary, water body boundary, mountain ridge, and topographic shadow ware found from LD_B7 extraction method. For the relationship between lineaments and physical characteristics of the study area, the most of extracted fault lines were normal fault. Additionally, the extracted fault lines were checked with 18 survey points. And found that 10 fault evidence from field survey data with 55.66% of the total survey point were correlated to extracted fault lines. On the other hand, 6 points with 33.33% were not correlated to the extracted fault lines and there were 2 points with 11.11% that had no extracted fault line overlap but the neighbor fault lines with similar direction were found. The study of semi-automatic Phayao-Fault zone extraction by satellite imagery, was carried out the guideline of specific lineament extraction, especially fault line, which are useful for investigating fault line. For getting more accuracy and precision, the appropriate dataset and methodology should be concerned for the further study.