Now days due to advancement of technology & high speed internet, the access and transfer of digital media such as image, text, audio and video becomes easy for everybody. Because of the increased use of internet, the need of security and authenticity has become essential. To secure the multimedia data, digital watermarking techniques have become key tool to tackle the challenges faced by the rapid distribution of digital content over the internet. Digital watermarking is a branch of information hiding which is used to hide proprietary information in digital media like photographs, digital music, digital video etc. In this paper we have developed image watermarking model based on discrete wavelet transform. In this technique, we embed and extract biometric watermark using the alpha blending technique in low frequency band. This paper illustrates a general architecture and modified algorithm for DWT based digital watermarking. To demonstrate working of the aforementioned architecture and algorithm, an experimental system is also developed with different value of alpha at 1,2 and 3 level of DWT. This paper also describes output of the experimental system with their comparison.