Normal view MARC view ISBD view

Reversible watermarking with tamper localization for medical images / Lim Hui Ching

By:

Lim, Hui Ching

Material type: Text

TextPublication details: Kuantan, Pahang : UMP, 2016Description: xvi, 62 p. : ill. (some col.) ; 30 cm. + 1 CD-ROMISBN:

THE0001174(Local)

Subject(s):

Dissertation note: Project paper (Bachelor of Computer Science (Software Engineering) With Honours) -- Universiti Malaysia Pahang – 2016 Abstract: This research focused on the application of medical image watermarking on ultrasound images. Reversible watermarking was used to authenticate ultrasound images to prevent unauthorized manipulation on the images and allow the retrieval of the region embedded with watermark. Region of interest (ROI) was embedded as watermark in other area to produce a watermarked image. The watermarked image should visually similar to the original version of the image. The watermarking scheme proposed in this research is reversible tamper localization watermarking scheme. The watermarking scheme was tested with four different sample ultrasound images and a storage image. The ultrasound image to be tested is divided into three types of region namely region of interest (ROI), region of non-interest (RONI) and untouchable region. During embedding process, the RONI bits will be stored into the storage image known as z. Next, ROI pixel values were hashed and been stored in RONI so that it can be retrieved during extraction process to authenticate the image. After that, the ROI bits were stored into the RONI. In extraction process, ROI of the image to be tested was hashed to obtain its hash value in order to detect tampering. The new hash value and the original hash value were being compared to detect tampering. If tampering occurred, image recovery was performed on the ROI. After that, the RONI was reversed to original bits to retrieve original image. Peak signal-to-noise ratio (PSNR) value of the watermarked image was used to measure the similarity of watermarked image with the original image. The higher the PSNR value the higher the similarity thus high PSNR value is preferred. The value of PSNR of the watermarked image of the four selected sample images is between 48.20 dB to 51.13 dB. In conclusion, medical image watermarking can be effective in protecting medical image from unauthorized manipulation.

Tags from this library: No tags from this library for this title. Log in to add tags.

Average rating: 0.0 (0 votes)

Holdings ( 2 )
Title notes ( 4 )

Holdings
Item type	Current library	Call number	Copy number	Status	Date due	Barcode
Final Year Report	UMPLIB PEKAN	FSKKP .L56 2016 r Bc. (Browse shelf(Opens below))	1	Not for loan		0000117588
Final Year Report	UMPLIB PEKAN	CD 10697 \| FSKKP .L56 2016 r Bc. (Browse shelf(Opens below))	1	Not for loan		0000117589

Faculty of Computer Systems and Software Engineering

Project paper (Bachelor of Computer Science (Software Engineering) With Honours) -- Universiti Malaysia Pahang – 2016

Bibliography : p. 57-58

This research focused on the application of medical image watermarking on ultrasound images. Reversible watermarking was used to authenticate ultrasound images to prevent unauthorized manipulation on the images and allow the retrieval of the region embedded with watermark. Region of interest (ROI) was embedded as watermark in other area to produce a watermarked image. The watermarked image should visually similar to the original version of the image. The watermarking scheme proposed in this research is reversible tamper localization watermarking scheme. The watermarking scheme was tested with four different sample ultrasound images and a storage image. The ultrasound image to be tested is divided into three types of region namely region of interest (ROI), region of non-interest (RONI) and untouchable region. During embedding process, the RONI bits will be stored into the storage image known as z. Next, ROI pixel values were hashed and been stored in RONI so that it can be retrieved during extraction process to authenticate the image. After that, the ROI bits were stored into the RONI. In extraction process, ROI of the image to be tested was hashed to obtain its hash value in order to detect tampering. The new hash value and the original hash value were being compared to detect tampering. If tampering occurred, image recovery was performed on the ROI. After that, the RONI was reversed to original bits to retrieve original image. Peak signal-to-noise ratio (PSNR) value of the watermarked image was used to measure the similarity of watermarked image with the original image. The higher the PSNR value the higher the similarity thus high PSNR value is preferred. The value of PSNR of the watermarked image of the four selected sample images is between 48.20 dB to 51.13 dB. In conclusion, medical image watermarking can be effective in protecting medical image from unauthorized manipulation.