Hybrid Steganographic Authentication (HSA) – Real-Time Pixel Matching, AI-Driven IDS-SIEM Deep Freeze, and Human-Only Recovery
Abstract
This paper presents a new profound way to create a new authentication method that takes really, really long time to crack. Why it takes really, really long time to crack? Unlike traditional method we type password like using special characteristics, even like that, we going to use a new method called image to say. We going to use an image as a password to say. But how you are going to use an image as a password? There’s a method in cloud that called PEM or we can call IAM. It’s kind of like an authentication but PEM is different. When you upload a file, then only we can access the things in it, like that. So, we going to use an image and upload it so we can use that as a password. What do you mean by uploading an image, can you use it as a password? No. There are additional things that I am adding to it. First, there going to be a cryptographic passkey, a 20 words that we going to use as a password. Well, this is what is going to do. It’s going to be the first thing for the password to keep your security. This can be used for companies or like that. For general peoples, you can use without this cryptographic passkey. But it’s going to be on higher level, so I am going to show you the cryptographic. This going to be the first. Second, you will upload an image. Third, you will provide a random word. Not a random word, let’s say “this is my first time using this authentication.” First, there will be a hashed word added to this and randomized and added to the image. Because we are adding a hash, it will be invisible. We cannot see it. Humans cannot see it. Then what happens is the random word that you have given me will be encrypted and added behind the image to say. If you want to understand, let me put it in another way to say. If you’re going to take a family photo, on the front of the photo will be the random words with a hash so it cannot be seen. On the back of the image, it will be white, nothing, but the encrypted words will be embedded in the back. So, it will become a three-layered to say. First will be random words, on the second layer will be the image, on the third layer it will be the encrypted words. Because of this, it will be very, very hard time for quantum computing or the current chatbots to crack it. It will take a real long. But to make it even more secure, there is two monitoring and detection system: AI-powered IDS and SIEM. This will constantly monitor. Why? Well, if the quantum computing is like I am going to find certain pixels to say, like that. It will come and find half pixels, then it will come and try again. So, to prevent this from happening, the AI, which is IDS and SIEM, which is powered by AI, will constantly monitor. Then there’s also another thing also, another thing, there must be 75% of the pixels must match immediately. If that doesn’t happen, it will trigger Deep Freeze. If this deep Freeze is triggered, it can also be triggered by IDS and SIEM. Well, if this happens, there’s only one way to recover, human only recovery method. Well, AI or quantum computing can use forget password and access, but if it’s a human recovery, if it cannot be quantum computing, it cannot do anything or neither do AI. Well, how do you recover to say? When this deep-freeze occurs or triggered, there you will be given an opportunity, not an opportunity, you will have to provide your phone number or Gmail to say. After that, there will be a person calling you. If you give a provider a number, the person will call you. Otherwise, you can use the Gmail, you will get a link. You can use an app like Google Meet or anything to face-to-face. They will have certain questions, why did your account got deep Freeze or you have to answer it. That’s the only way to recover. Due to this, even quantum computing or anything cannot hack to say every chatbot out there or AI will take really long time to crack and I feel sad for them.
Copyright (c) 2026 M.S. Done, K. Sudharsan, C.J Manju

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

