TÁI TẠO BIÊN DẠNG BỀ MẶT CHI TIẾT QUANG-CƠ VỚI GIAO THOA ÁNH SÁNG TRẮNG BẰNG MÔ HÌNH HỌC SÂU U-NET=RECONSTRUCTION OF THE SURFACE PROFILE OF OPTICAL-MECHANICAL STRUCTURE WITH WHITE LIGHT INTERFERENCE BY DEEP LEARNING U-NET MODEL

0 Người đánh giá. Xếp hạng trung bình 0

Tác giả: Hoàng Hải Lê, Kim Thư Lê, Văn Nhu Lê, Doãn Thông Nguyễn, Linh Khương Trần

Ngôn ngữ: vie

Ký hiệu phân loại:

Thông tin xuất bản: Journal of Science and Technique: Section on Physics and Chemical Engineering, 2024

Mô tả vật lý: tr.49

Bộ sưu tập: Metadata

ID: 483666

Thêm vào giỏ

Nghiên cứu này đề xuất một phương pháp để xác định chính xác vị trí đỉnh của tín hiệu giao thoa trong giao thoa ánh sáng trắng. Qua đó nâng cao độ chính xác cũng như độ phân giải của phép đo, ứng dụng trong đo lường chính xác và tái tạo vi cấu trúc ba chiều bề mặt chi tiết quang cơ. Bằng việc sử dụng thuật toán học sâu với kiến trúc mạng nơron U-net để đưa ra tín hiệu dự đoán sát với tín hiệu chuẩn nhất, kết hợp với biến đổi nhanh Fourier để lọc nhiễu và làm khớp với hàm số đề xuất, vị trí chính xác của đỉnh đường bao tín hiệu được xác định từ đó đưa ra thông tin về độ cao của điểm. Phương pháp này có thể đạt độ chính xác cỡ 0,9 nanomet ở mức nhiễu 50 dB, độ chính xác tăng hơn 40% ở các mức nhiễu so với phương pháp biến đổi Fourier truyền thống. Các vấn đề còn tồn tại của phương pháp biến đổi Fourier truyền thống trước đây như việc xác định vị trí đỉnh của tín hiệu giao thoa tại một vị trí, độ chính xác còn bị ảnh hưởng nhiều bởinhiễu và tín hiệu làm khớp được khắc phục một cách hiệu quả và rõ rệt.This research proposes a method to accurately determine the location of the interference signal peak in white light interferometry. This improves the accuracy and resolution of the measurement, enabling precise measurement and reconstruction of three-dimensional microstructures on optical surfaces. By using a deep learning algorithm with a U-net neural network architecture to generate a predicted signal that closely matches the reference signal, combined with Fast Fourier Transform to filter noise and fit the proposed function, the exact location of the envelope signal peak is determined, providing information about the height of the point. This method can achieve an accuracy of around 0.9 nanometers at a noise level of 50 dB, which is over 40% more accurate than the traditional Fourier transform method at various noise levels. The persistent issues of the traditional Fourier transform method, such as determining the location of the interference signal peak at a signal position and the accuracy being heavily influenced by noise and the fitting signal, are effectively addressed.

1. Đọc trực tuyến

1. Đọc trực tuyến

Tạo bộ sưu tập với mã QR