Аннотация
Image fusion is a process where multiple images (more than one) from different or same images of object are combined to form a single resultant fused image. This fused image is more informative, descriptive and qualitative as compared to its original input images or than individual images. The fusion technique in medical images is useful for resourceful disease diagnosis purpose and for doctors having varying experiences. This paper illustrates multimodality medical image fusion techniques and their results evaluated and analyzed with six quantitative metrics specified in last section of the paper. In this paper, a multimodality image fusion algorithm based on Dual tree discrete wavelet transform and particle swarm optimization (PSO) is proposed. Firstly, the source images are divided into low-frequency coefficients and high-frequency coefficients by the dual-tree discrete wavelet transform (DDWT) as separate parallel branch of band. PSO is used to determine to obtain proper fusion weight parameter from high-frequency coefficients from segmented images by DDWT. Also PSO is used to determine ? parameter called scalar weight. Finally, the fused image is reconstructed by the inverse DDWT. Thus quality of fused image is measured with PSNR, SNR, OCE, MI, Entropy
Пользователи данного ресурса
Пожалуйста,
войдите в систему, чтобы принять участие в дискуссии (добавить собственные рецензию, или комментарий)