專利授權區國立清華大學國際產學營運總中心 Operations Center for Industry Collaboration

搜尋專利授權區

關鍵字

» 新增關鍵字

選單

專利授權區

專利授權區
專利名稱(中)	基於時域光譜的斷層攝影方法、系統及裝置
專利名稱(英)	TOMOGRAPHY METHOD, SYSTEM AND APPARATUS BASED ON TIME-DOMAIN SPECTROSCOPY
專利家族	中華民國：I731689 美國：11,300,504
專利權人	國立清華大學 100.00%
發明人	楊尚樺,洪繹峻
技術領域	光電光學

專利摘要(中)
一種基於時域光譜的斷層攝影方法、系統及裝置。所述方法控制光發射器發出脈衝光束掃描待測物體的一切面，同時利用光接收器偵測穿過待測物體的脈衝光束，以取得掃描路徑上多個位置的時域脈衝訊號，以及改變掃描角度並重複執行掃描及偵測步驟，以收集切面的多個角度的時域脈衝訊號作為時間資料集；利用經訓練的機器學習模型的多個核心從各角度的時域脈衝訊號中擷取多個特徵，其中機器學習模型包括以多個切面的時間資訊集及對應的真實影像進行訓練，以學習用以擷取特徵的多個核心；以及轉換所擷取的特徵至空間域以重建切面的斷層影像。

專利摘要(中)

一種基於時域光譜的斷層攝影方法、系統及裝置。所述方法控制光發射器發出脈衝光束掃描待測物體的一切面，同時利用光接收器偵測穿過待測物體的脈衝光束，以取得掃描路徑上多個位置的時域脈衝訊號，以及改變掃描角度並重複執行掃描及偵測步驟，以收集切面的多個角度的時域脈衝訊號作為時間資料集；利用經訓練的機器學習模型的多個核心從各角度的時域脈衝訊號中擷取多個特徵，其中機器學習模型包括以多個切面的時間資訊集及對應的真實影像進行訓練，以學習用以擷取特徵的多個核心；以及轉換所擷取的特徵至空間域以重建切面的斷層影像。

專利摘要(英)
A tomography method, system and apparatus based on time-domain spectroscopy are provided. In the method, a light emitter is controlled to emit a pulse beam to scan a cross-section of an object to be measured and simultaneously a light sensor is used to detect the pulse beam passing through the object to be measured, so as to obtain time-domain pulse signals of locations in a scan path. The scanning and detecting steps are repeated after changing the scan angle, and the time-domain pulse signals of multiple angles of the cross-section are collected as a time data set. Multiple features are retrieved from the time-domain pulse signals by using kernels of a trained machine learning model, in which the machine learning model is trained with data sets and corresponding true images of multiple cross-sections, so as to learn the kernels for retrieving the features. The retrieved features are transformed into a spatial domain to reconstruct a tomographic image of the cross-section.

專利摘要(英)

A tomography method, system and apparatus based on time-domain spectroscopy are provided. In the method, a light emitter is controlled to emit a pulse beam to scan a cross-section of an object to be measured and simultaneously a light sensor is used to detect the pulse beam passing through the object to be measured, so as to obtain time-domain pulse signals of locations in a scan path. The scanning and detecting steps are repeated after changing the scan angle, and the time-domain pulse signals of multiple angles of the cross-section are collected as a time data set. Multiple features are retrieved from the time-domain pulse signals by using kernels of a trained machine learning model, in which the machine learning model is trained with data sets and corresponding true images of multiple cross-sections, so as to learn the kernels for retrieving the features. The retrieved features are transformed into a spatial domain to reconstruct a tomographic image of the cross-section.

聯絡資訊
承辦人姓名	李曉琪
承辦人電話	03-5715131 #31061
承辦人Email	hsiaochi@mx.nthu.edu.tw