A Proposed Text Encryption inside Video Using Harris Corner Detection and Salas20 Encryption Algorithm

Text encryption in video is a vital tool in many industries where protecting sensitive information is of highest concern. Algorithms and automated detection methods commonly struggle to locate hidden text in video frames. This resistance increases the security of the concealed information because it is challenging for uninvited parties to detect the presence of encrypted text. This research suggests a novel method for text encryption inside video utilizing the Salsa20 encryption algorithm and Harris Corner Detection. The goal of this study is to increase the secrecy and security of textual information that is embedded in video content. First step is locateing important corners or spots inside the video frames, the Harris Corner Detection method is first used. These corners act as strong feature points that can be used to incorporate text. Without degrading the video's overall visual quality, the Harris Corner Detection method aids in precisely localizing the places where text may be concealed. Second, the encryption key is generated using the Salsa20 encryption technique. Salsa20 is a popular stream cipher that offers reliable encryption and effective operation. The embedded text is converted into a cipher text by XOR it with the encryption key before embedding it in the video, guaranteeing its confidentiality and protecting it from unauthorized access. Comparing the suggested method to current text encryption methods reveals a number of advantages. The Harris Corner Detection technique integrates text into video frames so that it blends in less noticeably, making it more difficult for adversaries to identify the existence of hidden information. Salsa20's use further guarantees the encrypted text's high level of security and protection, defending it against potential attacks. The proposed system is evaluated using MSE, PSNR, correlation, NPCR, UACI and entropy, these evaluation metrics provide excellent results.