Fuzzy lattice theory have been widely used in image processing as it allows map functions residing in an original space to functions in a transformed space, resulting in powerful knowledge extraction and image pattern recognition. Despite recognition efficiency and best means of knowledge extraction, the computational complexity and the noise rate involved have been an open problem to be addressed. In this paper, to reduce the computational complexity by optimizing the number of granules between pixels and improving the PSNR through linear fuzzy transform, a method called Euclidean Fuzzy Lattice Orthogonal Image Transform (EFL-OIT) has been presented. © Springer Nature Singapore Pte Ltd 2020.