Abstract: Fireflies that attract all other fireflies or be attracted to other fireflies which are of higher brightness is the basis of the firefly algorithm. It is an optimization algorithm of calculating objective functional values. In this paper with the use of discrete wavelet transform the implementation of firefly algorithm is done and the experimental results of using certain number of initial and randomly allocated fireflies are used to calculate certain parameters which gives the ratio between the maximum signal power and maximum noise power, bit error rate(BER), coefficient measuring the degree of similarity(Corr), the normalized cross correlation(NC) between the watermarked image and original image.