**Author details**

*Digital Forensic Science*

manipulated.

can be ignored.

watermark.

**11. Conclusion and future works**

the area being tampered in the video frames.

been changed or manipulated.

find the video frame portion that has been changed.

test performance of the algorithm. If there is no manipulation done to the watermarked video (no attack case), then the extracted watermark is same exactly to the original watermark. Therefore, we conclude that the video is still original, has not

Common manipulations of video have been done to test authentication and localize altering in the watermarked video. These manipulations are adding a text label into video frames, inserting a new object into the video, changing contrast, and cropping some pixels. In the case of adding text and inserting an object into the video frames, we got the extracted watermarks that contain silhouette of the object or text. The silhouettes can be seen visually. When we compared to the original watermark, the extracted watermark is not the same. Therefore we conclude that the video has been manipulated. By subtracting the original watermark from the extracted watermark and adjusting the results on the watermarked video, we can

Common manipulation of video is changing the contrast or brightness of the video. By changing the contrast or brightness of the video, it means changing all pixel values in the video frame. When the watermarks are extracted from the video, we found that extracted watermarks also change entirely. The extracted watermarks are totally damaged; therefore we can conclude that the video has been

When a block area in the watermarked video frame is cropped, the extracted watermark is also cropped in the correspondence block. The extracted watermark

This proposed algorithm has some weakness. It cannot detect manipulation of the watermarked video if one or more fames are removed. However, if some video frames are inserted to the watermarked video, the algorithm can still detect this manipulation, because the new frames do not contain the embedded watermarks. Other weakness is LSB modification method itself. Bits of the watermark are only embedded to one least significant bit of pixel values. If manipulation of the watermarked video is performed on other than the least significant bit, the algorithm cannot detect it. However, this manipulation is considered uncommon so it

We have proposed a fragile video watermarking based on the chaotic map. In order to increase security, the watermark is encrypted using XOR operation with a random image. The random image is generated by using Cross-Coupled Chaotic random Bit Generator (CCCBG). The encrypted watermark is embedded to every RGB component of each frame. In the extraction process, the encrypted watermark is extracted from the watermarked video and compared to the original

Some experiments have been done to test capability of the algorithm to detect tampering to the watermarked video. We have tried some common attacks to the watermarked video. The experiment results showed that the algorithm could detect tampering on the watermarked video. This algorithm has also capability to localize

The algorithm can only be used to the uncompressed videos. It can be developed for the compressed video format such as MPEG-4. Embedding of watermarks is performed in encoding and decoding must be operated in transform domain. Some transform methods such as Fourier Transform, DCT, or wavelet transform can

has a black region in the cropped area of the correspondence frame.

**50**

be used.

Rinaldi Munir\* and Harlili Harlili School of Electrical Engineering and Informatics, Institut Teknologi Bandung, Bandung, Indonesia

\*Address all correspondence to: rinaldi@informatika.org

© 2020 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/ by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
