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)	影像的稀疏邊緣編碼及解碼方法
專利名稱(英)	Image Sparse Edge Encoding and Decoding Method
專利家族	中華民國：I832721 美國：US2024/0305787A1(公開號)
專利權人	國立清華大學 100.00%
發明人	林秀豪,岑美茵
技術領域	生化醫藥,資訊工程

專利摘要(中)
一種影像的稀疏邊緣編碼及解碼方法，藉由一電腦裝置來執行，該電腦裝置儲存一原始影像，該原始影像包括多筆像素資訊，每一像素資訊具有一像素值及一像素位置。該電腦裝置根據該等像素資訊及一遮罩權重矩陣進行編碼，以獲得多個分別對應該等像素資訊的影像電荷，以提升稀疏度，其中，該遮罩權重矩陣為方矩陣，且所有元素的總和為零。該電腦裝置根據該等影像電荷及該遮罩權重矩陣，利用一相關於求解稀疏線性系統的演算法求多個穩態解，依據該等影像電荷強度優先順序，獲得多筆解碼像素資訊，以根據該等解碼像素資訊獲得具有極高的精度的重建影像。

專利摘要(中)

一種影像的稀疏邊緣編碼及解碼方法，藉由一電腦裝置來執行，該電腦裝置儲存一原始影像，該原始影像包括多筆像素資訊，每一像素資訊具有一像素值及一像素位置。該電腦裝置根據該等像素資訊及一遮罩權重矩陣進行編碼，以獲得多個分別對應該等像素資訊的影像電荷，以提升稀疏度，其中，該遮罩權重矩陣為方矩陣，且所有元素的總和為零。該電腦裝置根據該等影像電荷及該遮罩權重矩陣，利用一相關於求解稀疏線性系統的演算法求多個穩態解，依據該等影像電荷強度優先順序，獲得多筆解碼像素資訊，以根據該等解碼像素資訊獲得具有極高的精度的重建影像。

專利摘要(英)
An image sparse edge encoding and decoding method is implemented by a computer device, and the computer device stores an original image, the original image includes a plurality of pieces of pixel information, and each pixel information has a pixel value and a pixel position. The computer device performs encoding according to the pixel information and a mask weight matrix to obtain multiple image charges correspond to the pixel information respectively to improve the sparseness. The mask weight matrix is a square matrix, and the sum of all elements is zero. The computer device uses an algorithm related to solving a sparse linear system to obtain multiple steady-state solutions according to the image charges and the mask weight matrix. The computer device obtains a plurality of pieces of decoded pixel information according to the priority order of the image charge intensity, so as to obtain a reconstructed image with extremely high precision according to the decoded pixel information.

專利摘要(英)

An image sparse edge encoding and decoding method is implemented by a computer device, and the computer device stores an original image, the original image includes a plurality of pieces of pixel information, and each pixel information has a pixel value and a pixel position. The computer device performs encoding according to the pixel information and a mask weight matrix to obtain multiple image charges correspond to the pixel information respectively to improve the sparseness. The mask weight matrix is a square matrix, and the sum of all elements is zero. The computer device uses an algorithm related to solving a sparse linear system to obtain multiple steady-state solutions according to the image charges and the mask weight matrix. The computer device obtains a plurality of pieces of decoded pixel information according to the priority order of the image charge intensity, so as to obtain a reconstructed image with extremely high precision according to the decoded pixel information.

聯絡資訊
承辦人姓名	李馥如