A method and system for encryption of a visual media for a secure output generation

Abstract

The present disclosure relates to a method and a system for encryption of a visual media for a secure output generation. The method encompasses selecting the visual media. The method further comprises calculating a number of Most Significant Bits (MSBs) for the visual media. Further, the method comprises encrypting the visual media based on the number of MSBs. Furthermore, the method comprises training a sub-system [204a] based on the encrypted visual media. Thereafter, the method comprises utilizing the trained sub-system [204a] for the secure output generation.

Description

Description Title of Invention :A METHOD AND SYSTEM FOR ENCRYPTION OF A VISUAL MEDIA FOR A SECURE OUTPUT GENERATIONTechnical Field

[0001] Embodiment of the present disclosure generally relates to multimedia security technology. More particularly, embodiments of the present disclosure relate to a method and a system for encrypting of a visual media for a secure output generation.Background Art

[0002] The following description of the related art is intended to provide background information pertaining to the field of the disclosure. This section may include certain aspects of the art that may be related to various features of the present disclosure. However, it should be appreciated that this section is used only to enhance the understanding of the reader with respect to the present disclosure, and not as admissions of the prior art.

[0003] In today's digital age, cloud computing has become an integral part of our daily lives, offering the ability to store, process, and analyse vast amounts of information over the internet, eliminating the need for extensive on-premises infrastructure and enabling scalable, flexible solutions. The users can save various information on cloud, making it possible to access the files from anywhere with internet connectivity.

[0004] With the rapid growth of the technology, the cloud computing has been advanced by integrating Artificial intelligence (Al) which significantly enhanced the capabilities and efficiency. Many organizations have started creating various Al based applications using cloudbased development services. It enables real-time data processing and provides immediate insights which facilitate rapid decision making. However, the integration of Al with cloud computing also raises important considerations regarding data security, privacy and trust. As Al technologies increasingly rely on cloud platforms, the risk of unauthorized access and misuse of sensitive information increases that further raises the concern related to the privacy and security of the information of the end user in the visual media.

[0005] Currently, for protecting privacy of the user in the visual media such as, but not limited to, an image, a video, etc. the existing solutions uses a method that fully encrypts a portion of the visual media that may include sensitive information of the end user. Further, encryption of the information is a process that transforms the information that may be meaningful to a human into an unreadable format i.e., cypher text that may not be meaningful to a human but may ormay not remain meaningful for the Al. In addition, encryption is a security approach where data is encoded and can only be accessed or decrypted by a user using a decryption key. Even if someone intercepts the encrypted data, they can’t understand it without the proper decryption key. Such approach helps in maintaining data privacy and security of the end user across various digital platforms and applications. Further, in the existing solution, only sensitive portion of the individual is encrypted, and non-encrypted portion of the information is available for processing on cloud or on other such data processing platforms.

[0006] Furthermore, the existing solutions uses various algorithms and / or encryptions techniques such as, but not limited to, homomorphic encryption, etc. Homomorphic encryption is an advanced technique over conventional encryption technique that allows various operation / computations on encrypted data without decrypting said encrypted data. Homographic encryption enables secure computations in a situation where sensitive data needs to be processed or shared such as in healthcare or financial services. However, there are several challenges associated with homomorphic encryption that limits its applicability in the existing solutions. In the existing solutions it is limited to a few types of computations / operations, more specifically, currently homomorphic encryption can perform only simple polynomial operations on the data. Further, it involves complex mathematical operations, which can be computationally intensive. This results in higher processing time compared to traditional encryption methods, making it less efficient for real-time applications. Furthermore, some homomorphic encryption techniques support only basic operations, and supporting complex operations or functions may be challenging. This limitation can restrict the applicability of homomorphic encryption to certain types of computations such as to apply on visual media. For instance, some of the existing solutions encompasses providing a full homomorphic encryption of images, which has various limitations such as the full homomorphic encryption is compute intensive, and visual media encrypted using the full homomorphic encryption cannot be efficiently used by Al based systems. Moreover, processing techniques for the visual media encrypted using the full homomorphic encryption are limited and therefore the full homomorphic encryption as provided by the existing solutions is not efficient.

[0007] The above-mentioned limitations of the existing solutions restrict the performance for the intensive tasks. Due to complex operations, it requires extensive data processing which limit its practicality. Moreover, use of the full homomorphic encryption for visual media in Al based solutions is not efficient. Thus, there exist a need for a technical solution that can overcome at least the above-mentioned technical limitations of the existing solutions. Morespecifically there is a need in the art to provide a method that can encrypt the visual media for a secure and efficient output generation.Solution to Problem

[0008] This section is provided to introduce certain aspects of the present disclosure in a simplified form that are further described below in the detailed description. This summary is not intended to identify the key features or the scope of the claimed subject matter.

[0009] An aspect of the present disclosure may relate to a method for encryption of a visual media for a secure output generation. The method comprises selecting the visual media. Further, the method comprises calculating a number of Most Significant Bits (MSBs) for the visual media. Further, the method comprises encrypting the visual media based on the number of MSBs. Further, the method comprises training a sub-system based on the encrypted visual media. Further, the method comprises utilizing the trained sub-system for the secure output generation.

[0010] In an exemplary aspect of the present disclosure, the visual media comprises at least one of a set of images and a set of videos.

[0011] In an exemplary aspect of the present disclosure, the utilizing the trained sub-system for the secure output generation comprises: (a) receiving a target visual media at the trained sub-system in an encrypted form, (b) analyzing the target visual media by the trained subsystem, and (c) generating an output by the trained sub-system based on the analysis.

[0012] In an exemplary aspect of the present disclosure, the encryption of the target visual media is based on a target number of MSBs for the target visual media.

[0013] In an exemplary aspect of the present disclosure, the target number is one of a number less than the number of MSBs and a number equal to the number of MSBs.

[0014] In an exemplary aspect of the present disclosure, the encrypting the visual media further comprises encrypting the number of MSBs for each pixel in the visual media.

[0015] In an exemplary aspect of the present disclosure, the number of MSBs is calculated based on at least one of a set of properties of the visual media and a pre-defined privacy level.

[0016] In an exemplary aspect of the present disclosure, the pre-defined privacy level is user defined, and the pre-defined privacy level is defined based on a required degree of privacy protection.

[0017] In an exemplary aspect of the present disclosure, prior to calculating the number of MSBs for the visual media, the visual media is resized in at least one resolution supported by the sub-system.

[0018] In an exemplary aspect of the present disclosure, the sub-system is an artificial intelligence-based module.

[0019] Another aspect of the present disclosure may relate to a system for encryption of a visual media for a secure output generation. The system comprises a processing unit, and a memory unit connected to the processing unit. The processing unit is configured to select the visual media. Further, the processing unit is configured to calculate a number of Most Significant Bits (MSBs) for the visual media. Further, the processing unit is configured to encrypt the visual media based on the number of MSBs. Further, the processing unit is configured to train a sub-system based on the encrypted visual media. Further, the processing unit is configured to utilize the trained sub-system for the secure output generation.

[0020] Yet, another aspect of the present disclosure relates to a non-transitory computer readable storage medium storing one or more instructions for encryption of a visual media for a secure output generation, the instructions include executable code which, when executed by one or more units of a system, causes a processing unit of a system to select the visual media. Further, the executable code which, when executed by the one or more units of the system, causes the processing unit of the system to calculate a number of Most Significant Bits (MSBs) for the visual media. Further, the executable code which, when executed by the one or more units of the system, causes the processing unit of the system to encrypt the visual media based on the number of MSBs. Furthermore, the executable code which, when executed by the one or more units of the system, causes the processing unit of the system to train a sub-system based on the encrypted visual media. Thereafter, the executable code which, when executed by the one or more units of the system, causes the processing unit of the system to utilize the trained sub-system for the secure output generation.Advantageous Effects of Invention

[0021] Some of the objects of the present disclosure, which at least one embodiment disclosed herein satisfies are listed herein below.

[0022] It is an object of the present disclosure to provide a solution for approximate homomorphic encryption of a visual media.

[0023] It is also an object of the present disclosure to provide a solution for generating an encrypted visual media on which one or more types of processing (including but not limited to an artificial intelligence (Al) based processing) can be done in an efficient manner.

[0024] It is an object of the present disclosure to provide a solution for encrypting one or more most significant bits of each pixel in the visual media.

[0025] It is another object of the present disclosure to provide a solution for training a subsystem (e.g., an artificial intelligence-based model) on partially encrypted visual media.

[0026] It is another object of the present disclosure to provide a solution for fast and efficient encryption of the visual media.

[0027] It is another object of the present disclosure to provide a solution to protect the privacy and security of the end user.

[0028] It is yet another object of the present disclosure to provide a solution for generating an output from the trained sub-system using an encrypted visual media as an input.Brief Description of Drawings

[0029] The accompanying drawings, which are incorporated herein, and constitute a part of this disclosure, illustrate exemplary embodiments of the disclosed methods and systems in which like reference numerals refer to the same parts throughout the different drawings. Components in the drawings are not necessarily to scale, emphasis instead being placed upon clearly illustrating the principles of the present disclosure. Also, the embodiments shown in the figures are not to be construed as limiting the disclosure, but the possible variants of the method and system according to the disclosure are illustrated herein to highlight the advantages of the disclosure. It will be appreciated by those skilled in the art that disclosure of such drawings includes disclosure of electrical components or circuitry commonly used to implement such components.

[0030] FIG. 1 illustrates an exemplary block diagram of a system for encryption of a visual media for a secure output generation, in accordance with exemplary embodiments of the present disclosure.

[0031] FIG. 2 illustrates an exemplary functional flow diagram of a system architecture in accordance with exemplary embodiment of the present disclosure.

[0032] FIG. 3 illustrates an exemplary flow diagram depicting functionalities of an encryption module for generating the encrypted visual media, in accordance with exemplary embodiments of the present disclosure.

[0033] FIG. 4 illustrates an exemplary sequence flow diagram depicting functionalities of the encryption module for generating and utilising the encrypted visual media, in accordance with exemplary embodiments of the present disclosure.

[0034] FIG. 5 illustrates an exemplary flow diagram depicting a method for encryption of a visual media for a secure output generation, in accordance with exemplary embodiments of the present disclosure.

[0035] FIG. 6 illustrates an exemplary use case of training of a sub-system and generating a secured output, in accordance with exemplary embodiments of the present disclosure.

[0036] FIG. 7 illustrates an exemplary use case of generating text from the encrypted visual media, in accordance with exemplary embodiments of the present disclosure.

[0037] The foregoing shall be more apparent from the following more detailed description of the disclosure.Description of Embodiments

[0038] In the following description, for the purposes of explanation, various specific details are set forth in order to provide a thorough understanding of embodiments of the present disclosure. It will be apparent, however, that embodiments of the present disclosure may be practiced without these specific details. Several features described hereafter may each be used independently of one another or with any combination of other features. An individual feature may not address any of the problems discussed above or might address only some of the problems discussed above.

[0039] The ensuing description provides exemplary embodiments only, and is not intended to limit the scope, applicability, or configuration of the disclosure. Rather, the ensuing description of the exemplary embodiments will provide those skilled in the art with an enabling description for implementing an exemplary embodiment. It should be understood that various changes may be made in the function and arrangement of elements without departing from the scope of the disclosure as set forth.

[0040] Specific details are given in the following description to provide a thorough understanding of the embodiments. However, it will be understood by one of ordinary skill in the art that the embodiments may be practiced without these specific details. For example, circuits, systems, processes, and other components may be shown as components in block diagram form in order not to obscure the embodiments in unnecessary detail.

[0041] Also, it is noted that individual embodiments may be described as a process which is depicted as a flowchart, a flow diagram, a data flow diagram, a structure diagram, or a block diagram. Although a flowchart may describe the operations as a sequential process, many of the operations may be performed in parallel or concurrently. In addition, the order of the operations may be re-arranged. A process is terminated when its operations are completed but could have additional steps not included in a figure.

[0042] The word “exemplary” and / or “demonstrative” is used herein to mean serving as an example, instance, or illustration. For the avoidance of doubt, the subject matter disclosed herein is not limited by such examples. In addition, any aspect or design described herein as “exemplary” and / or “demonstrative” is not necessarily to be construed as preferred or advantageous over other aspects or designs, nor is it meant to preclude equivalent exemplary structures and techniques known to those of ordinary skill in the art. Furthermore, to the extent that the terms “includes,” “has,” “contains,” and other similar words are used in either the detailed description or the claims, such terms are intended to be inclusive — in a manner similar to the term “comprising” as an open transition word — without precluding any additional or other elements.

[0043] As used herein, “a user equipment”, “a user device”, “a smart-user-device”, “a smartdevice”, “an electronic device”, “a mobile device”, and “a device” may be any electrical, electronic and / or computing device or equipment, capable of implementing at least some of the features of the present disclosure. The user equipment / device may include, but is not limited to, a mobile phone, a smart phone, a laptop, a general-purpose computer, a desktop, a personal digital assistant, a tablet computer, a wearable device or any other computing device which is capable of implementing at least some of the features of the present disclosure.

[0044] As used herein, “storage unit” or “memory unit” refers to a machine or computer- readable medium including any mechanism for storing information in a form readable by a computer or similar machine. For example, a computer-readable medium includes read-only memory (“ROM”), random access memory (“RAM”), magnetic disk storage media, optical storage media, flash memory devices or other types of machine-accessible storage media. The storage unit stores at least the data that may be required by one or more units of the system to perform their respective functions.

[0045] All modules, units, components used herein, unless explicitly excluded herein, may be software modules or hardware processors, the processors being a general-purpose processor, a special purpose processor, a conventional processor, a digital signal processor (DSP), a plurality of microprocessors, one or more microprocessors in association with a DSP core, a controller, a microcontroller, Application Specific Integrated Circuits (ASIC), Field Programmable Gate Array circuits (FPGA), any other type of integrated circuits, etc.

[0046] One or more of the plurality of modules may be implemented through an Al model. A function associated with Al may be performed through the non-volatile memory, the volatile memory, and the processor. The processor may include one or a plurality of processors. Forimplementing the one or the plurality of modules through an Al model, the one or the plurality of processors may be a general purpose processor(s), such as a central processing unit (CPU), an application processor (AP), or the like, a graphics-only processing unit such as a graphics processing unit (GPU), a visual processing unit (VPU), and / or an Al-dedicated processor such as a neural processing unit (NPU).

[0047] The one or the plurality of processors control the processing of the input data in accordance with a predefined operating rule or artificial intelligence (Al) model stored in the non-volatile memory and the volatile memory. The predefined operating rule or artificial intelligence model is provided through training or learning.

[0048] Here, being provided through learning means that, by applying a learning algorithm(s) to a plurality of learning data, a predefined operating rule or Al model of a desired characteristic is made. The learning may be performed in a device itself in which Al according to an embodiment is performed, and / or may be implemented through a separate server / system.

[0049] The Al model may consist of a plurality of neural network layers, such as long shortterm memory (LSTM) layers. Each layer has a plurality of weight values and performs a layer operation through calculation of a previous layer and an operation of a plurality of weights. Examples of neural networks include, but are not limited to, convolutional neural network (CNN), deep neural network (DNN), recurrent neural network (RNN), restricted Boltzmann Machine (RBM), deep belief network (DBN), bidirectional recurrent deep neural network (BRDNN), generative adversarial networks (GAN), and deep Q-networks.

[0050] Also, a learning algorithm refers to a method for training a device (for example, a robot) using a plurality of learning data to cause, allow, or control the device to make a determination or prediction. Examples of learning algorithms include, but are not limited to, supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning.

[0051] As discussed in the background section, the increase in the dependency of an Artificial Intelligence (Al) on the cloud platform has raised a concern related to the privacy and security of the information of an individual in the visual media such as, but not limited to, an image, a video etc. wherein the information may also include sensitive information of the individual. Further, to protect the privacy and security of the information of the individual various technical solutions have been developed. The existing solutions includes one or more method that fully encrypts a portion of the visual media that may include sensitive information of the end user. Further, the existing technologies use a homomorphic encryption technique which is used for a full homomorphic encryption of visual media and is limited to perform only simplepolynomial operations. Further, the full homomorphic encryption involves complex mathematical operations, which can be computationally intensive in nature. This results in higher processing time compared to traditional encryption methods, making the full homomorphic encryption less efficient for real-time applications.

[0052] In order to overcome the at least some of the limitations of the prior known solutions, the present disclosure provides a solution for partially encrypting the entire visual media for a secure output generation. In an implementation the visual media comprises at least one of a set of images and a set of videos. In order to apply the method for partially encrypting the visual media for secure output generation, the visual media is selected first. Further, one or more Most Significant Bits (MSBs) of the selected visual medial are calculated and encrypted. Furthermore, a sub-system is trained based on the encrypted visual media and the sub-system is then utilized for the secure output generation. Thereby, achieving the privacy and security of an information present in a visual media that is encrypted using an approximate homomorphic method.

[0053] Hereinafter, exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily carry out the present disclosure.

[0054] Referring to FIG. 1, an exemplary block diagram of a system

[0100] for encryption of a visual media for a secure output generation, in accordance with exemplary embodiments of the present disclosure is shown. The system

[0100] comprises at least one processing unit

[0102] and at least one memory unit

[0104] , Also, all of the components / units of the system

[0100] are assumed to be connected to each other unless otherwise indicated below. Also, in FIG. 1 only a few units are shown, however, the system

[0100] may comprise multiple such units, or the system

[0100] may comprise any such numbers of said units, as required to implement the features of the present disclosure.

[0055] Particularly, for encryption of the visual media for the secure output generation, the processing unit

[0102] is configured to select the visual media. The visual media comprises at least one of a set of images and a set of videos. In an implementation, the set of images and the set of videos may be associated with interest of a user. In an example, the set of images may include a selfie image of the user, an identity card of the user, and the like.

[0056] Further, the processing unit

[0102] is configured to calculate a number of Most Significant Bits (MSBs) for the visual media. The most significant bit (MSB) is a bit in a binary number, a byte, and a binary-coded data element that has the highest positional value. Itrepresents the leftmost or highest-order bit in a pixel of the visual media. Further, the number of MSBs is calculated based on at least one of a set of properties of the visual media and a predefined privacy level. Also, the pre-defined privacy level is user defined, and the pre-defined privacy level is defined based on a required degree of privacy protection. Further, the detailed description regarding the calculation of the MSBs is provided in the detailed description of FIG.2. Further, prior to calculating the number of MSBs for the visual media, the visual media is resized in at least one resolution supported by a sub-system. Furthermore, the sub-system is an artificial intelligence-based module.

[0057] Further, the processing unit

[0102] is further configured to encrypt the visual media based on the number of MSBs. The encryption of the visual media further comprises encrypting the number of MSBs for each pixel in the visual media. Further, the processing unit

[0102] is configured to train the sub-system based on the encrypted visual media. Furthermore, the processing unit

[0102] is configured to utilize the trained sub-system for the secure output generation. Thereafter, the processing unit

[0102] to utilize the trained sub-system for the secure output generation, receives a target visual media at the trained sub-system in an encrypted form. Once the sub-system receives the target visual media, the sub-system analyses the target visual media and generates an output based on the analysis. Moreover, the encryption of the target visual media is based on a target number of MSBs for the target visual media and the target number is one of a number less than the number of MSBs calculated for the visual media and a number equal to the number of MSBs calculated for the visual media.

[0058] Referring to FIG. 2, an exemplary functional flow diagram of a system architecture (comprising a system

[0202] and a server

[0204] ) is shown, in accordance with exemplary embodiment of the present disclosure. The system

[0202] comprises at least one camera [2021a], at least one memory [2022a], at least one image resize module [202b], at least one Most Significant Bit (MSB) digits calculator [202c], at least one encryption module [202d] and at least one output application [202e], Also, all of the components / units of the system

[0202] are assumed to be connected to each other unless otherwise indicated below. In an implementation, the system

[0202] may encompasses or may be in communication with the system

[0100] of FIG. 1. Also, the system

[0202] may reside in or may be in communication with a user device. In an implementation, the system

[0202] performs the technical functions as disclosed in the present disclosure in conjunction with one or more components of the system

[0100] , Yet, in another implementation the system

[0202] itself (via its one or more components) performs the technical functions as disclosed in the present disclosure.

[0059] Further, the server

[0204] comprises at least one sub-system [204a], The sub-system [204a] is an artificial intelligence-based module. The system

[0202] and the server

[0204] are connected to each other to perform the one or more functions as disclosed in the present disclosure.

[0060] The camera [2021a] is configured to select a visual media comprising a set of images and a set of videos. In an embodiment, the camera [2021a] may be configured to capture the set of images and the set of videos. The memory [2022a] is configured to store the visual media in the system

[0202] , The camera [2021a] and the memory [2022a] transmits the visual media to the image resize module [202b] of the system

[0202] , In an example, the visual media is an input image Io. The image resize module [202b] is configured to resize the input image To into at least one resolution supported by the sub-system [204a], The resolution may be set based on a manual input to generate image I. The image resize module [202b] is an implementation is configured to resize the input image Io by using one or more image resizing techniques such as by using an average of pixel value of the input image To, if a required resolution of the generated image I is smaller than the input image Jo. The image resize module [202b] in another implementation is configured to resize the input image To by using an interpolation of pixel values, if a required resolution of the generated image I is larger than the input image Jo. In another implementation of the preset disclosure, the image resize module [202b] may resize the image by using at least one of Artificial Intelligence (Al) based resizing technique, a Generative Adversaial Network (GAN) based resizing technique, and a Generative Al (Gen Al) based resizing technique.

[0061] In an implementation, Al based resizing technique involves using one or more artificial intelligence based methods to resize visual medias (e.g., images) by adjusting their dimensions while attempting to preserve visual quality and important details. The Al based resizing technique often employs techniques like interpolation and deep learning models to maintain clarity and avoid distortion.

[0062] In another implementation, the Generative Adversarial Network (GAN) based resizing technique involves utilizing a pair of neural networks, a generator and a discriminator to create high-quality resized visual media. The generator creates visual media of new sizes, while the discriminator evaluates their quality, refining the resizing process to produce more realistic results.

[0063] In yet another implementation, the Generative Al (Gen Al) based resizing technique leverages advanced generative models to enhance resolution of the visual media or adjustdimensions. The Gen Al based resizing technique focuses on generating new visual media content that aligns with the original visual media's style and context, potentially adding new details along with scaling of the visual media.

[0064] Also, the system

[0202] comprises the Most Significant Bits (MSB) digits calculator [202c], The MSB digits calculator [202c] is configured to calculate a number of Most Significant Bits (MSBs) for the visual media to be encrypted. The calculation of the number of MSBs is based on at least one of a set of properties of the visual media and a pre-defined privacy level. The pre-defined privacy level may be one of a medium privacy level, a high privacy level and an absolute privacy level. The pre-defined privacy level is defined based on a manual input. In addition, the pre-defined privacy level is defined based on a degree of privacy protection set based on the manual input. In an exemplary embodiment, for medium privacy level, per pixel encrypted bit may be 1. Similarly, for high privacy level, per pixel encrypted bit may be 2 and for absolute privacy level, per pixel encrypted bit may be 3. However, the disclosure is not limited thereto and a person skilled in the art would appreciate that the per pixel encrypted bit may vary based on a use case. In an exemplary embodiment, a higher number of MSBs may be chosen based on a manual input for an image containing sensitive information to ensure higher security, while less sensitive images might use fewer MSBs.

[0065] In an implementation, the MSB digits calculator [202c] is configured to calculate the number of bits to be encrypted, wherein the number of bits may be calculated as:

[0066] where, r0x c0is the resolution of a sample image,

[0067] r x c is the resolution of the input image,

[0068] n0is the number of bits encrypted in the sample image, and

[0069] n is the number of bits to be encrypted for the input image.

[0070] In an implementation, the sample image is an image having a resolution supported by the sub-system [204a],

[0071] Also, prior to calculating the number of MSBs for the visual media, the visual media is resized in at least one resolution supported by the sub-system.

[0072] Further, the encryption module [202d] is configured to encrypt the number of Most Significant Bits (MSBs) for each pixel in the visual media as calculated by the MSB digits calculator [202c], In general, encryption is a form of data security in which information is converted to ciphertext. Only authorized people who have the key may decrypt the ciphertext and access the corresponding original plaintext information. A detailed explanation regarding the encryption of the visual module in the encryption module [202d] is provided in the detailed description of Fig. 3 and Fig. 4.

[0073] Further, the sub-system [204a] in the server

[0204] is trained based on the encrypted visual media (i.e., the visual media encrypted by the encryption module [202d]). The trained sub-system [204a] is utilized to generate the secure output.

[0074] In an implementation, the training of the sub-system [204a] includes three steps. At first step, the training involves feeding data (i.e., encrypted visual media) into the sub-system [204a], The data is fed to the sub-system [204a] to identify one or more features that may be present in the encrypted visual media such as, skin colour, hair colour, etc. in an event the encrypted visual media is an image of a person. Further, a massive amount of data (i.e., a plurality of encrypted visual media) is fed to the sub-system [204a] to identify the one or more features more accurately.

[0075] Next, at second step, the of training of the sub-system [204a] involves validation testing for which the performance of the sub-system [204a] is evaluated based on a data that may be unseen to the sub-system [204a], The validation testing is performed to determine if the training needs to be continued or needs to be modified. In an embodiment, validation testing evaluates how a model performs on a new data.

[0076] Finally, at third step, the training of the sub-system [204a] involves applying the subsystem [204a] in the real world.

[0077] In an implementation, the trained sub-system [204a] may receive a target visual media in an encrypted form. The encryption of the target visual media is based on a target number of MSBs for the target visual media. The target number is one of a number less than the number of MSBs calculated for the visual media and a number equal to the number of MSBs. For example, if a sub-system [204a] is trained based on a plurality of visual media, wherein each visual media from said plurality of visual media has 2 encrypted MSBs for each pixel. In such example a target visual media may have 1 or 2 encrypted MSBs for each pixel, i.e., the target number of MSBs in such example is less than or equal to 2.

[0078] The trained sub-system [204a] may analyse the received encrypted target visual media. Thereafter, the trained sub-system [204a] may generate an output based of the analysis of the received encrypted target visual media. The analysis of the received encrypted target visual is performed based on a training or a learning of the sub-system [204a] using the plurality of encrypted visual media.

[0079] Furthermore, the system

[0202] comprises the output application [202e] configured to receive the output generated by the trained sub-system [204a] configured in the server

[0204] , In an implementation, the sub-system [204a] may generate the output in the form of at least one of: a text, an image, a video and an audio, however the disclosure is not limited thereto.

[0080] Referring FIG. 3, an exemplary flow diagram

[0300] depicting functionalities of encryption module [202d] for generating the encrypted visual media, in accordance with exemplary embodiments of the present disclosure is shown. In an implementation, the encryption module [202d] performs a step of MSB bits separation [SI], MSB bits stream encryption [S2] and encrypted visual media (say image) generation [S3],

[0081] As shown in FIG. 3, the encryption starts at step SI.

[0082] At step SI, the encryption module [202d] separates one or more Most Significant Bits (MSBs) from every pixel for each colour of a visual media. In general, a Most Significant Bit is a bit in a binary number, byte, or binary-coded data element that has the highest positional value. The MSB represents the leftmost or highest-order bit in a binary representation.

[0083] At step S2, the encryption module [202d] encrypts a block of plaintext bits corresponding to the one or more MSBs to create a block of ciphertext of equal length using a stream cipher. The stream cipher provides the ciphertext with length same as that of the plaintext.

[0084] At step S3, the encryption module [202d] generates the encrypted image by replacing each plaintext bit in an input image with the mapped encrypted bit.

[0085] The encryption terminates after step S3.

[0086] Further, the functionalities of the encryption module [202d] for generating the encrypted visual media, is explained in detail with respect to FIG. 4.

[0087]

[0088] Referring to FIG. 4, an exemplary sequence flow diagram

[0400] depicting functionalities of the encryption module [202d] for generating and utilising the encrypted visual media is shown, in accordance with exemplary embodiments of the present disclosure. The sequence flow diagram

[0400] depicts a method for generating the encrypted visual media.

[0089] At step

[0402] , the method comprises separating Most Significant Bit(s) (MSB(s)) from each pixel of a visual media (say an image) based on a level of privacy protection set based on a manual input. Generally, each pixel of the visual media comprises 8-bits for each primary colour red, green, and blue, thus making a total of 24 bits for each pixel. In an implementation of the present disclosure, a first bit from each row of pixel Pl and P2 depicted in Fig. 4 is assumed to be a Most Significant Bit for the primary colour corresponding to the row, r denoting red, g denoting green, and b denoting blue. Further, the first bit from each pixel is separated to form a plain text. Also, the method comprises concatenating each separated bit of the pixel Pl and the pixel P2 to form a plaintext. Thereafter as indicated in Fig. 4, a block of plaintext is created for example, Irl lgl lbl2r22g22b2 by concatenating each separated bit from each row of the pixels Pl and P2.

[0090] At step

[0404] , the process comprises the step of MSB bit stream encryption. The MSB bit stream encryption converts the plain text into the ciphertext of equal length to form an encrypted block, for instance, X1X2X3X4X5X6, as indicated in Fig. 4. In an implementation, the stream cipher is used to generate ciphertext of same length as that of the plaintext. In general, cyphertext is the encrypted output of plaintext data which is transformed using a stream cipher.

[0091] At step

[0406] , the process comprises the step of generation of an encrypted image. The encrypted block, generated at the step

[0404] , is mapped bit by bit with the block of plain text. Thereafter as depicted in Fig. 4, the plaintext bit in each row of the pixel Pl and the pixel P2 is replaced by corresponding encrypted bit. Based on the mapped encrypted bit in each row of the pixel Pl and the pixel P2, the encrypted image is generated.

[0092] Referring to FIG. 5, an exemplary flow diagram depicting a method

[0500] for encryption of a visual media for a secure output generation, in accordance with exemplary embodiments of the present disclosure is shown. In an implementation, the method

[0500] may be performed by the system

[0100] , Also, in an implementation the system

[0100] works in conjunction with the system

[0202] connected to the server

[0204] , The method

[0500] starts at step

[0502] ,

[0093] At step

[0504] , the method comprises selecting the visual media. The visual media comprises at least one of a set of images and a set of videos. Further, the set of images comprise one or more images either captured by the camera [2021a] of the system

[0202] or one or more images stored in the memory [2022a] of the system

[0202] , Further, the set of videos comprises one or more videos either shot by the camera [2021a] of the system

[0202] or one or more videos stored in the memory [2022a] of the system

[0202] ,

[0094] Next, at step

[0506] , the method comprises calculating a number of Most Significant Bits (MSBs) for the visual media. In an exemplary embodiment, the MSBs are the bits in a digital representation of the visual media that contribute most to the visual media's quality and details. For example, for an image with 24-bit colour depth, the MSBs may include the top 8 bits of each colour channel. The number of MSBs may be calculated based on a required privacy level. Prior to calculating the number of MSBs for the visual media, the visual media is resized in at least one resolution supported by the sub-system [204a], Further, the most significant bit (MSB) is a bit in a binary number, a byte, and a binary-coded data element that has the highest positional value. The MSB represents the leftmost or highest-order bit in the visual media. Further, the number of MSBs is calculated based on at least one of a set of properties of the visual media and a pre-defined privacy level. Further, the pre-defined privacy level is user defined, and the pre-defined privacy level is defined based on a required degree of privacy protection. Further, the required degree of privacy protection is based on the number of bits to be encrypted such as, but not limited to, if 1 bit is encrypted the degree of privacy protection may be medium. If 2 bits are encrypted, the degree of the privacy protection may be high and if 3 bits are encrypted the degree of the privacy protection may be relatively higher. Furthermore, the process of calculating the number of MSBs is explained in the detailed description of FIG. 2.

[0095] Further, at step

[0508] , the method

[0500] comprises encrypting the visual media based on the number of MSBs. The encrypting of the visual media further comprises encrypting the number of MSBs for each pixel in the visual media. Further, the number of the MSBs to be encrypted in each pixel is based on the degree of the privacy protection set based on a manual input. Generally, encryption is a form of data security in which information is converted to ciphertext and only authorized user who have the key can decipher the ciphertext and access the corresponding original plaintext information.

[0096] The method of encryption of the visual media further comprises separating one or more Most Significant Bits (MSBs) from every pixel for each colour of the visual media. In an example, in a coloured image, the encryption might focus on one or more MSBs of each pixel's red, green, and blue colour channels separately. Further, the method comprises encrypting a block of plaintext bits to create a block of ciphertext of equal length using a stream cipher. The block of plaintext includes the one or more MSBs separated by the encryption module [202d], Further, the stream cipher provides the ciphertext with length same as that of the plaintext.Furthermore, the method comprises generating the encrypted image by replacing each plaintext bit in the original image with the mapped encrypted bit.

[0097] Furthermore, at step

[0510] , the method

[0500] comprises training the sub-system [204a] based on the encrypted visual media. The sub-system [204a] is an artificial intelligence-based module. Further, the method of training the sub-system [204a] comprises training the subsystem [204a] by feeding data (i.e., encrypted visual media) into the sub-system [204a], The data is feed to the sub-system [204a] to identify one or more features that may be present in the encrypted visual media such as, but not limited to, skin colour, hair colour, etc. in an event the encrypted visual media is an image of a person. Further, a massive amount of data (i.e., a plurality of encrypted visual media) is feed to the sub-system [204a] to identify the one or more features more accurately. Further, the method comprises validation testing, in which the performance of the sub-system [204a] is evaluated based on a data that may be unseen to the sub-system [204a], results in determining at least one of a continuation the training and a modification in the training. The method further comprises testing the sub-system [204a] in the real world.

[0098] Moreover, at step

[0512] , the method comprises utilizing the trained sub-system for the secure output generation. The method for generating secure output by utilizing the sub-system [204a] further comprises receiving a target visual media at the trained sub-system in an encrypted form. Further, the encryption of the target visual media is based on a target number of MSBs for the target visual media. The target number is one of a number less than the number of MSBs calculated for the visual media and a number equal to the number of MSBs. For instance, if a visual media that is used for training the sub-system [204a] encompasses 8 encrypted MSBs, the target media may comprise either 8 encrypted MSBs or fewer encrypted MSBs, depending on the required security and protection for the target media. The method further comprises analysing the target visual media by the trained sub-system [204a] and generating an output by the trained sub-system [204a] based on the analysis. In an implementation, the sub-system [204a] may generate the output in the form of at least one of a text, an image, a video and an audio, however the disclosure is not limited thereto.

[0099] Thereafter, the method terminates at step

[0514] ,

[0100] Referring FIG. 6, an exemplary use case

[0600] of training of a sub-system [204a] and generating the secured output by the sub-system [204a] based on the partially encrypted one or more visual media, in accordance with exemplary embodiments of the present disclosure is shown. As depicted at

[0602] each of the one or more visual media is encrypted using theencryption module [202d], Further, as depicted at

[0604] the one or more encrypted visual media is used for the training of the sub-system [204a], Also, as depicted at

[0606] the trained subsystem [204a] may be configured at a cloud platform to provide an output in the form of at least one of: a text, an image (with recognised face etc.) and a video (with recognised face etc.).

[0101] In an implementation as depicted at

[0608] for utilising the trained sub-system [204a], an encrypted visual media (say an encrypted image) is provided to the trained sub-system [204a] for face recognition. Next

[0610] depicts that the trained sub-system [204a] is trained for face recognition and is configured to generate an output (as depicted at

[0612] ) based on the encrypted image provided to the sub-system [204a] as the input. Thus, the privacy is protected by generating and utilising the encrypted visual media.

[0102] Referring FIG. 7, an exemplary use case

[0700] of generating text from the encrypted visual media, in accordance with exemplary embodiments of the present disclosure is shown. As depicted at [SI] in the Fig 7, an encrypted visual media (say an image) is provided to a trained sub-system [204a] configured at an image server. Next, the trained sub-system [204a] at [S2] is configured to generate a text based on the encrypted image shared on the trained subsystem [204a], Therefore, at [S2] a generative text “Fun with local village kid” is generated by the trained sub-system [204a] without decrypting the image. Thus, at [S3] it is depicted that a privacy of visual media is protected by utilising the encrypted visual media.

[0103] The present disclosure further discloses a non-transitory computer readable storage medium storing one or more instructions for encryption of a visual media for a secure output generation, the instructions include executable code which, when executed by one or more units of a system, causes a processing unit

[0102] of a system

[0100] to select the visual media. Further, the executable code which, when executed by the one or more units of the system, causes the processing unit

[0102] of the system

[0100] to calculate a number of Most Significant Bits (MSBs) for the visual media. Further, the executable code which, when executed by the one or more units of the system, causes the processing unit

[0102] of the system

[0100] to encrypt the visual media based on the number of MSBs. Furthermore, the executable code which, when executed by the one or more units of the system, causes the processing unit

[0102] of the system

[0100] to train a sub-system based on the encrypted visual media. Thereafter, the executable code which, when executed by the one or more units of the system, causes the processing unit

[0102] of the system

[0100] to utilize the trained sub-system for the secure output generation.

[0104] Thus, the present invention provides a novel solution for encryption of a visual media for a secure output generation. The present solution facilitates providing an encryption of themost significant bits of each pixel in the visual media. Further, the present solution facilitates providing a training of a sub-system based on a plurality of partially encrypted visual media. Further, the present solution is a fast and an efficient method for encrypting the visual media and utilising the encrypted visual media. Furthermore, the present solution protects the privacy and security of an end user by providing the encrypted visual media. Moreover, the present solution generates a secured output from the trained sub-subsystem using an encrypted visual media.

[0105] While considerable emphasis has been placed herein on the preferred embodiments, it will be appreciated that many embodiments can be made and that many changes can be made in the preferred embodiments without departing from the principles of the invention. These and other changes in the preferred embodiments of the invention will be apparent to those skilled in the art from the disclosure herein, whereby it is to be distinctly understood that the foregoing descriptive matter to be implemented merely as illustrative of the invention and not as limitation.

Claims

Claims

1. A method for encryption of a visual media for a secure output generation, the method comprises: selecting the visual media; calculating a number of Most Significant Bits (MSBs) for the visual media; encrypting the visual media based on the number of MSBs; training a sub-system based [204a] on the encrypted visual media; and utilizing the trained sub-system [204a] for the secure output generation.

2. The method as claimed in claim 1, wherein the visual media comprises at least one of a set of images and a set of videos.

3. The method as claimed in claim 1, wherein the utilizing the trained sub-system [204a] for the secure output generation comprises: receiving a target visual media at the trained sub-system [204a] in an encrypted form, analyzing the target visual media by the trained sub-system [204a], and generating an output by the trained sub-system [204a] based on the analysis.

4. The method as claimed in claim 3, wherein the encryption of the target visual media is based on a target number of MSB s for the target visual media.

5. The method as claimed in claim 4, wherein the target number is one of a number less than the number of MSBs and a number equal to the number of MSBs.

6. The method as claimed in claim 1, wherein the encrypting the visual media further comprises encrypting the number of MSBs for each pixel in the visual media.

7. The method as claimed in claim 1, wherein the number of MSBs is calculated based on at least one of a set of properties of the visual media and a pre-defined privacy level.

8. The method as claimed in claim 7, wherein the pre-defined privacy level is user defined, and the pre-defined privacy level is defined based on a required degree of privacy protection.

9. The method as claimed in claim 1, wherein prior to calculating the number of MSBs for the visual media, the visual media is resized in at least one resolution supported by the subsystem [204a],

10. The method as claimed in claim 1, wherein the sub-system is an artificial intelligencebased module.

11. A system for encryption of a visual media for a secure output generation, the system comprises: a processing unit [102]; and a memory unit [104] connected to the processing unit [102], wherein the processing unit [102] is configured to: select the visual media, calculate a number of Most Significant Bits (MSBs) for the visual media, encrypt the visual media based on the number of MSBs, train a sub-system [204a] based on the encrypted visual media, and utilize the trained sub-system [204a] for the secure output generation.

Images