Description:Field of the Invention
This invention relates to a system of Image Authentication using Artificial Intelligence.
Background of the Invention
It is of great importance to check the authenticity of photographs for educational, archival and research purposes. High-quality visual documentation is necessary but the advent of digital photo editing tools has made it easier to tamper with photos thereby compromising their credibility. The genuineness of historic pictures can be hard to ascertain because presently employed ways usually involve time consuming examination by eye that may contain faults or simple forensic techniques not fit for large scale application.Existing methods are inadequate because they can't efficiently handle large collections of images or detect subtle alterations. Manual verification by experts is slow and prone to human error, while basic digital forensic methods lack the sophistication to capture complex visual and contextual details.To deal with such challenges, an automated system is proposed that will be able to confirm whether the historical images are authenticated or not. It can identify fake photos by comparing visual features of uploaded photos against a database of known authentic historical photographs. This method guarantees the dependability and scalability of the process through its reliance on sophisticated image processing techniques. The preservation of visual historicity for educational, archival or research purposes would be enhanced in this way.
US7940965B2 discloses that the face of a subject person can be rapidly detected. An image input unit inputs an image to be processed. A photographing-position input unit inputs photographing-position information attached to the processed image. An angle-range information determination unit determines an angle range, where face detection should be performed to the processed image, on the basis of the information obtained by the photographing-position information input unit. On the basis of information indicating the determined angle range, under the control of a process control unit, a face detection unit performs face detection to the processed image input by the image input unit in predetermined angle increments. A face integration unit integrates all of face data detected by a basic-angle-range face detection unit into face information and then outputs the information.
RESEARCH GAP: This invention decreases human labour intensiveness and expertise requirements thereby reducing expenses associated with manual identification procedures.
High Precision: The advanced feature extraction methods used in similarity analysis enable detection of even minute differences and similarities between pictures thus enhancing the precision of the results.
US11835702B2 A medical observation system including a medical imaging device that captures a plurality of images of a living body while changing a focus position, and circuitry that generates a composite image by compositing the plurality of images captured by the medical imaging device, and switches output between the generated composite image and one of the plurality of images based on a result of analysis performed on at least one of the plurality of images.
RESEARCH GAP: Adaptability: It is a versatile tool for verifying images, which can be used in different places with various conditions for instance color, size or resolution.
Scalability: It is therefore appropriate for large-scale activities like museums; archives and online platforms as it can handle vast amounts of images without overloading.
None of the prior art indicate above either alone or in combination with one another disclose what the present invention has disclosed. This invention relates to Image Authentication using Artificial Intelligence.
SUMMARY OF THE INVENTION
This summary is provided to introduce a selection of concepts, in a simplified format, that are further described in the detailed description of the invention.
This summary is neither intended to identify key or essential inventive concepts of the invention and nor is it intended for determining the scope of the invention.
To further clarify advantages and features of the present invention, a more particular description of the invention will be rendered by reference to specific embodiments thereof, which is illustrated in the appended drawings. It is appreciated that these drawings depict only typical embodiments of the invention and are therefore not to be considered limiting of its scope. The invention will be described and explained with additional specificity and detail with the accompanying drawings.
The traditional ways of verifying whether images are authenticated or not are becoming more and more difficult because the methods these ways use often take much time and rely on human judgment, hence making it impossible to process a large number of photographs efficiently. To solve this problem, our innovation will automate image authenticity verification without using any artificial intelligence. Rather, our tool uses advanced image processing techniques like conventional feature extraction methods and similarity analysis. We aim at ensuring that we have a simple and clear methodology in order to be able to provide a dependable answer for upholding the sanctity of old pictures.
BRIEF DESCRIPTION OF THE DRAWINGS
The illustrated embodiments of the subject matter will be understood by reference to the drawings, wherein like parts are designated by like numerals throughout. The following description is intended only by way of example, and simply illustrates certain selected embodiments of devices, systems, and methods that are consistent with the subject matter as claimed herein, wherein:
FIGURE1. FRAMEWORK OF AUTHENTICATION PROCESS
FIGURE 2: DATA FLOW PROCESS OF IMAGE AUTHENTICATION PROCESS
The figures depict embodiments of the present subject matter for the purposes of illustration only. A person skilled in the art will easily recognize from the following description that alternative embodiments of the structures and methods illustrated herein may be employed without departing from the principles of the disclosure described herein.
DETAILED DESCRIPTION OF THE INVENTION
The detailed description of various exemplary embodiments of the disclosure is described herein with reference to the accompanying drawings. It should be noted that the embodiments are described herein in such details as to clearly communicate the disclosure. However, the amount of details provided herein is not intended to limit the anticipated variations of embodiments; on the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the scope of the present disclosure as defined by the appended claims.
It is also to be understood that various arrangements may be devised that, although not explicitly described or shown herein, embody the principles of the present disclosure. Moreover, all statements herein reciting principles, aspects, and embodiments of the present disclosure, as well as specific examples, are intended to encompass equivalents thereof.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments. As used herein, the singular forms “a",” “an” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms “comprises,” “comprising,” “includes” and/or “including,” when used herein, specify the presence of stated features, integers, steps, operations, elements and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and/or groups thereof.
It should also be noted that in some alternative implementations, the functions/acts noted may occur out of the order noted in the figures. For example, two figures shown in succession may, in fact, be executed concurrently or may sometimes be executed in the reverse order, depending upon the functionality/acts involved.
In addition, the descriptions of "first", "second", “third”, and the like in the present invention are used for the purpose of description only, and are not to be construed as indicating or implying their relative importance or implicitly indicating the number of technical features indicated. Thus, features defining "first" and "second" may include at least one of the features, either explicitly or implicitly.
Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which example embodiments belong. It will be further understood that terms, e.g., those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
The traditional ways of verifying whether images are authenticated or not are becoming more and more difficult because the methods these ways use often take much time and rely on human judgment, hence making it impossible to process a large number of photographs efficiently. To solve this problem, our innovation will automate image authenticity verification without using any artificial intelligence. Rather, our tool uses advanced image processing techniques like conventional feature extraction methods and similarity analysis. We aim at ensuring that we have a simple and clear methodology in order to be able to provide a dependable answer for upholding the sanctity of old pictures. To calculate the authentication score of the image the following steps were involved as shown in figure 1. The first steps are dataset collection. This step includes gathering a dataset of authenticated images, typically around 50,000 images of different monuments, places etc from different sources. The dataset serves as foundation for training the model and for our invention. A diverse dataset is important so that the model works well across different situations and locations.
After collecting the dataset, feature extraction of these images using pre-trained Convolutional Neural Network (CNN) i.e,ResNet101v2 takes place. These are CNNs trained on large-scale image datasets (such as ImageNet), and thus can extract significant features from images.The CNN converts each image into a high-dimensional feature vector which compactly and effectively stores important characteristics of such images as edges, textures or patterns.
The extracted feature vectors are stored in High-dimensional Databases via FAISS (Facebook AI Similarity Search).FAISS is a high dimensional vector similarity search and clustering library.Storing the feature vectors in a database allows for fast retrieval and matching of images based on their visual similarities. For this reason, FAISS applies indexing to optimize search performance, making it well-suited to large-scale image retrieval tasks.
User Interaction:
Users upload an image they want to identify through user UI. The system interface allows them to pick an image from the phone’s gallery or take a new photo. This way, all users can easily access the system for image authentication and submit images that need identification services. The uploaded image feature extraction is done through the same pre-trained CNN as dataset images. This ensures that all images processed by the system share similar feature representations. The ResNet101v2 extract the feature vector of any uploaded picture, portraying its visual characteristics in a manner that is compatible with the stored characteristic of vectors.Comparisons with stored authenticated images in FAISS database are then made with uploaded image feature. The process for similarity search is fast and efficient, which helps to find the closest matches from the authenticated images to the uploaded image. This search allows us to identify among all of them which images have similar visual features as the uploaded one since it shows just how alike they look and based on the similarity between the uploaded image’s feature vector and the nearest neighbors retrieved from the FAISS database,the system calculates an authentication score. With this score, those who are interested can know if a particular upload is likely to be from a given location in their system. If higher scores indicate stronger matches, then lower scores probably mean less confidence.If the authentication score is above the threshold, the system displays the image is authenticated. The process concludes once the user receives the results on their interface. The system then resets, ready to process the next image.This step ensures that the system maintains responsiveness and readiness to handle subsequent image authentication requests from users.
It has a user interface (UI) for mobile and web applications in its image authentication system. In this regard, users can either pick photos from their already existing gallery or simply take new ones with their device camera. This intuitive UI design allows the user to work with the system without difficulty. At the point of image upload, API Gateway serves as an entry point which receives these requests and then routes them to backend services that are appropriate for such uploads. Next is Image Processing Service, which is responsible for handling feature vector extraction from images through a pre-trained CNN model that guarantees consistency in similarity search results. It is built on tools like TensorFlow Serving or PyTorch Serve so as to enable real-time inference model and ensure scalability and availability.
The processed feature vectors are kept in a database that also contains metadata about images and uses FAISS for storage and retrieval of high-dimensional vectors efficiently. The Similarity Search Service integrates FAISS to perform fast searches, finding the closest matches to the uploaded image's feature vector. Authentication score is calculated considering the similarity metrics after ending a similarity search process. After that, the Authentication Service determines if this score exceeds a predetermined threshold for positive identification and generates responses accordingly. In case the image is verified, location information is given to the user. If not, there are feedbacks from the user. This feedback is processed by the User Feedback Service, which collects user insights to improve over time on its model and database.
To ensure continuous improvement, periodically retrain the CNN model using new data and user feedback through Model Retraining Pipeline which enhances accuracy and performance. The whole communication flow is so simplified that users upload images via mobile or web application; these images are forwarded to Image Processing service by an API Gateway where feature vector is extracted. The Similarity Search Service sends this vector into FAISS database to look for similar images in it. Based upon these results, authentication score is calculated and whether the score reaches a certain threshold of acceptance or not is determined by Authentication service who then send this back through API Gateway back to User. User Feedback can be utilized in Model Retraining Pipeline as requested in order to keep updating system efficiency.
An image Authentication using Artificial Intelligence comprises a data collection module (101), an image processing module (102), a Feature extraction module (103), a Convolutional Neural Network model (CNN) (104), a FAISS (Facebook AI Similarity Search) (105), a mobile application (106), and a Model retraining Pipeline (107), wherein the data collection module is gathering a dataset of authenticated images, typically around 50,000 images of different monuments, places etc from different sources and The dataset serves as foundation for training the model and for our invention.
In another embodiment the image processing module is responsible for handling feature vector extraction from images through a pre-trained the CNN model that guarantees consistency in similarity search results.
In another embodiment the Feature extraction module ensuring that we have a simple and clear methodology in order to be able to provide a dependable answer for upholding the sanctity of old pictures.
In another embodiment the FAISS (Facebook AI Similarity Search) applies indexing to optimize search performance, making it well-suited to large-scale image retrieval tasks.
In another embodiment the mobile application is used for users that can either pick photos from their already existing gallery or simply take new ones with their device camera.
In another embodiment To ensure continuous improvement, periodically retrain the CNN model using new data and user feedback through Model Retraining Pipeline which enhances accuracy and performance.
ADVANTAGES OF THE INVENTION
1. Speed and Efficiency: In the blink of an eye, the automated system processes a great many of pictures, thus providing authentication results in real time, quicker than manual verification.
2. Consistency: While human judgment may vary greatly, this system provides constant results through using the same processing and comparison techniques every time which ensures authentication uniformity.
3. Accessibility: It is friendly to users who can easily upload images using it; either from their phone memory or by taking a new photo hence it’s open to people with different degrees of IT literacy.
4. Reduced Human Error: Therefore, the more accurate verification will be achieved since the manual control is excluded what might lead to mistakes due to overfatigue, negligence or personal prejudice.
5. Data Utilization: The system leverages a large dataset of authenticated images to improve the accuracy of its results, utilizing existing data effectively to enhance performance.
6. Reliability: The system gives assurance to users because of its strength in marking the image as not fake.
, Claims:1. An image Authentication System using Artificial Intelligence comprises a data collection module (101), an image processing module (102), a Feature extraction module (103), a Convolutional Neural Network model (CNN) (104), a FAISS (Facebook AI Similarity Search) (105), a mobile application (106), and a Model retraining Pipeline (107), wherein the data collection module is gathering a dataset of authenticated images, typically around 50,000 images of different monuments, places etc from different sources and The dataset serves as foundation for training the model and for our invention.
2. The system as claimed in claim 1, wherein the image processing module is responsible for handling feature vector extraction from images through a pre-trained the CNN model that guarantees consistency in similarity search results.
3. The system as claimed in claim 1, wherein the Feature extraction module ensuring a simple and clear methodology in order to be able to provide a dependable answer for upholding the sanctity of old pictures.
4. The system as claimed in claim 1, wherein the FAISS (Facebook AI Similarity Search) applies indexing to optimize search performance, making it well-suited to large-scale image retrieval tasks.
5. The system as claimed in claim 1, wherein the mobile application is used for users that can either pick photos from their already existing gallery or simply take new ones with their device camera.

6. The system as claimed in claim 1, wherein to ensure continuous improvement, periodically retrain the CNN model using new data and user feedback through Model Retraining Pipeline which enhances accuracy and performance.