採分區式直方圖位移法應用於遠距醫療之低記憶量可逆浮水印技術
No Thumbnail Available
Date
2015
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
在本論文中，我們提出一套用於醫學影像上的改良式直方圖位移浮水印嵌入技術。在先前文獻的方法中，浮水印及數位影像能夠經由傳送端所產生的位置圖資訊進行無損還原。然而，先前的方法並未考量到醫學影像的眾多特性。該特性將會產生大資料量的位置圖資訊，進而影響傳送資料的時間。
為了解決大資料量的問題，我們提出一個新的紀錄位置圖資訊方法。我們利用一個位元的資料量紀錄原始影像及藏入資訊後影像之間的變化量以避免溢位及下溢的問題產生，並利用兩個位元的資料量紀錄區塊的高點資訊避免在接收端發生錯誤的資料判斷。在實驗結果中，本論文提出的方法能夠降低紀錄資訊的資料量達到85%以上。此外，我們移植此技術至Android應用程式來實現遠距醫療系統的概念。
In the paper, we propose an improved lossless data hiding method based on histogram shifting for medical images. In the previous studies, the watermark and cover image can be restored completely by the location map generated in the embedding process. However, the location map will not be considered when the medical image is being used as the cover image. This case will result in large-volume data of location map. To solve this problem, we propose a new method to record the location map. We observe the change of the original pixels and the resulting pixels in the cover image, and record each pixel pair by one bit for avoiding the overflow and underflow. Furthermore, we record each block by two bits to avoid wrong information in the extraction process. In the experiment results, our method can reduce the amount of data of the location map than the previous studies. Besides, we create an app for realizing a telemedicine system on Android OS.
In the paper, we propose an improved lossless data hiding method based on histogram shifting for medical images. In the previous studies, the watermark and cover image can be restored completely by the location map generated in the embedding process. However, the location map will not be considered when the medical image is being used as the cover image. This case will result in large-volume data of location map. To solve this problem, we propose a new method to record the location map. We observe the change of the original pixels and the resulting pixels in the cover image, and record each pixel pair by one bit for avoiding the overflow and underflow. Furthermore, we record each block by two bits to avoid wrong information in the extraction process. In the experiment results, our method can reduce the amount of data of the location map than the previous studies. Besides, we create an app for realizing a telemedicine system on Android OS.
Description
Keywords
無損資料藏入技術, 直方圖位移, 位置圖, 醫學影像, lossless data hiding, histogram shifting, location map, medical images