A multi-focus image fusion technique for RGB color images using discrete cosine transform (DCT)

Abstract

Image fusion is a process that combines information from multiple images of the same scene into a single image. In this study, a new image fusion technique is proposed using the Discrete Cosine Transform (DCT). Each of the source images may represent a partial view of the scene, and contains both ‘relevant’ and ‘irrelevant’ data. Sincethe target ofimage fusion technique is to combine the source images into a single composite image which contains a more accurate description of the scene than any of the individual sources. Moreover, the fused image has gotten the best possible quality without introduction of distortion or loss of information. DCT algorithm is considered efficient in image fusion. The proposed scheme is performed in five steps: (1) RGB colour image (input image) is split into three channels R, G, and B for both images.(2) DCT algorithm is applied to each channel (R, G and B) for image1 and image2 respectively. (3) The variance values are computed for the corresponding 8x8 blocks of each channel. (4) Each block of R (image 1) is compared with each block of R (image2) based on the variance value and then the block having maximum variance value is selected to be the block in the new image. This process is repeated for all channels of image1 and image 2. (5) Inverse discrete cosine transform (IDCT) is applied on each fused channel to convert it from coefficient values to pixel values, and then combined all the channels to generate the fused image. The proposed technique can potentially solve the problem of undesirable side effects like blurring or blocking artifacts from reducing the quality of the resultant image in image fusion.The proposed approach is evaluated using three measurement units: the average of Qabf, standard deviation and peak Signal Noise Rate (PSNR). The experimental results of this proposed technique achieved good results as compared to previous studies.