Steganography is the art of hiding data in images. It has been well hidden when large amounts of data are hidden in the image and it is difficult to detect with the human eye. With the advent of digital media, it is imperative that robust ways to embed information are developed making it difficult for attackers to intercept. In this paper, a combination of Fuzzy Edge Detection and Cohort Intelligence (CI) is used to make the process of image hiding more efficient. Threshold ranging is used to make extraction simpler; the edge image is divided into 5 images based on threshold values and one is chosen for the embedding. During extraction, the edges of that image are chosen. Comparison with Multi Random Start Local Search (MRSLS) has been performed to check optimality of CI in terms of time and PSNR value. This algorithm is also tested on four images of different sizes and the results have been noted proving the efficiency of using the combination.