The major goal of this study is to embed watermarks into a video file. The proposed scheme embeds watermarks into image block which contains features with high intensity, high texture, and fast motion, and improves the robustness of the embedded watermarks based on spatial domain. Since human visual system can?t sense the variations of these feature regions within the video frames, an adaptive watermarking technique was developed to embed watermark into these regions. To this end, a digital video is first divided into various frames consisting of several blocks which are transformed by discrete wavelet transform. Spread spectrum incorporated with just notice difference is utilized to embed the watermark into the feature blocks or non-feature blocks. The proposed method can resist attacks performed by linear transformations including average, frame reduction, and frame shuttle, and obtain better performance in comparison with traditional schemes.