Fusion for medical image based on discrete wavelet transform coefficient

Zahraa Yaseen Hasan, Rusul Altaie, Hawraa Abd Al-kadum Hassan


More recent digital camera introduced enormous facilities for users from different specifications to take images easily, but the user still wants to improve these images, which it contains different problems like ambiguous and colors is not clear, because not enough light, cloudy weather, bright light, dark region and it's taken from remote distances. This paper aims to use a new approach for fusion images by using a wavelet coefficient based on PSNR and SNR measure (the technical result) instead of using the max, min, average values, and so on in the previous methods. The wavelet coefficient of each sub band is compared between them, the sub band had a value higher of measure is selected for fusion. Firstly, a discrete wavelet transform has been applied to the medical images with 2level. Then, the peak signal to noise ratio and signal to noise ratio measures have been computed for each sub-band. After that PSNR and SNR values have been compared for each sub-band to opposite sub-band and it selected the better value of measures. Secondly, PSNR and SNR values have been gathered for each image. Then select the image that contains value higher PSNR and lower value of SNR for purpose fusion. Finally, perform an inverse discrete wavelet on the fused image to transform it from the frequency to the spatial domain. The results of the work showed that the wavelet coefficient is used to preserve the image details and removed the noise. PSNR value of 1level of dwt is higher than 2level. This paper makes the image more clearer and informative than the original images. 


Discrete wavelet transform; Image fusion; Medical image; Peak signal to noise ratio; Signal to noise ratio

DOI: http://doi.org/10.11591/ijeecs.v21.i3.pp%25p
Total views : 81 times


  • There are currently no refbacks.

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

shopify stats IJEECS visitor statistics