Browsing by Author "Rajput, Amitesh Singh"

Now showing 1 - 20 of 26

2DInpaint: A novel privacy-preserving scheme for image inpainting in an encrypted domain over the cloud
(Elsevier, 2020-10) Rajput, Amitesh Singh
The low cost, agility, and mobility of cloud services for processing and storage data have attracted user’s attention today. Untrusted third parties support these services, and users are always concerned about utilizing them for personal data. Addressing these data-privacy issues for image inpainting over the cloud infrastructure(s), we propose a model, 2DInpaint, to perform image inpainting by protecting image information from an eavesdropping adversary. Inpainting is a technique for modifying an image in an undetectable manner with applications ranging from restoration of damaged photographs to object-removal and replacement of lost blocks in image coding and transmission. It can be accomplished by propagating the information in the isophotes direction of the desired region(s) from the neighborhood. Performing this propagation when the image is in the encrypted domain (ED) is a challenging dilemma. The challenge is addressed by employing a modified version of 2D-bicubic interpolation over the region to be inpainted in ED. The ramp secret sharing scheme is utilized to secure image information and to reduce storage overhead over the cloud server. 2DInpaint is proved to be information-theoretical secure in a probabilistic viewpoint and through various cryptographic attacks. The qualitative and quantitative results of 2DInpaint are analyzed under the scenarios of classical image inpainting, object-removal, and text-removal, and compared with the schemes in the plain domain. Moreover, no limitations related to the topology of the region to be inpainted are required using our approach. To the best of our knowledge, 2DInpaint is the first move towards image inpainting in the ED.
An Alternate Approach for Single Image Haze Removal Using Path Prediction
(Springer, 2023-05) Rajput, Amitesh Singh
Haze removal is an important problem that existing studies have addressed from slight to extreme levels. It finds wide application in landscape photography where the haze causes low contrast and saturation, but it can also be used to improve images taken during rainy and foggy conditions. In this paper, considering the importance of haze removal and possible limitations of a well-known existing method, DehazeNet, we propose an alternate end-to-end method for single image haze removal using simple yet efficient image processing techniques. DehazeNet is among well performing haze removal schemes in the literature, however the problem of coloration and artifacts being produced in the output images has been observed for a certain set of images. Addressing this problem, the proposed solution is devised by taking advantage of the color features of the three color channels of an image to establish a decision criteria. Based on that, a suitable dehazing method for an input hazy image is selected. Removing the problem of poor coloration in output images, we have come up with an alternate method to remove the haze while retaining the visual and perceived quality of the image. The experimental results show that the proposed method yields better structural restoration, reduces haze content significantly, and does not cause any artifacts in the output image.
Blockchain for Privacy-Preserving Data Distribution in Healthcare
(ICISSP, 2024) Rajput, Amitesh Singh
As virtual transformation maintains to reshape healthcare, the security and privacy of health information have become paramount worries. This paper delves into the novel application of blockchain generation as a strategic technique to these urgent issues. In contrast to traditional centralized information control structures, blockchain introduces an intensive alternate with its decentralized, immutable, and transparent nature. This shift gives a robust alternative to comply with sensitive health data. We propose a contemporary, blockchainprimarily based method to seamlessly integrate existing healthcare records into ledgers and share them in a controlled way. The proposed method emphasizes enhanced data integrity, advanced security features, and a patient-centric technique to data governance using customized smart contracts. Experimental results underline the proposed method’s advanced performance for scalability, protection, and general machine performance, making a compelling case for its adoption in healthcare records control
Cloud based image color transfer and storage in encrypted domain
(ACM Digital Library, 2018-08) Rajput, Amitesh Singh
Cloud infrastructures are developed and maintained by third parties and users are always concerned about processing and storing their data over the cloud. Recent technologies such as high definition and 360-degree images/videos require efficient color processing, and current trend towards the cloud computing has initiated a necessity of performing color transfer remotely by untrusted third party servers. Nowadays, this field is emerging fast due to its inherent potential and research work in this direction is highly demanded. To address this necessity, we present a system that addresses the challenge of performing privacy preserving color transfer over third party servers. We use a one-dimensional chaotic logistic map coupled with ramp secret sharing scheme in a manner that secret images can be stored and processed for color transfer in the encrypted domain. Experimental results and security analysis demonstrate effectiveness of the approach against existing techniques of color transfer as well as image encryption.
Color me, store me, know me not: Supporting image color transfer and storage in encrypted domain over cloud
(IEEE, 2017) Rajput, Amitesh Singh
Color transfer is a well known methodology to transfer color between images such that color palette of the resulting image should match as of the reference image. Current trend towards cloud computing has initiated the requirement of performing color transfer remotely by untrusted third party servers. To address this requirement, we present a system that addresses the challenge of performing color transfer between images when the test images are in encrypted form. While standard methods accomplish the task of color transfer in plain domain, our approach is the first known scheme to perform color transfer in encrypted domain. Furthermore, the proposed approach also solves the problem of privacy preserving storage over cloud infrastructures given that the test images can remain in encrypted form without requiring any decryption. Experimental results and security analysis demonstrate the potential and effectiveness of our approach with various application scenarios.
CryptFine: Towards secure cloud based filtering in encrypted domain
(IEEE, 2017) Rajput, Amitesh Singh
Cloud computing is a shift towards increasing the capabilities dynamically without investing in new infrastructure. Enormous storage of data and high-speed computing to customers over the internet can be accomplished using cloud. Security of data in cloud is one of the chief concerns which acts as an obstacle in the realization of cloud computing. Along with quick expansion and application of cloud computing, users concern more and more regarding security and privacy issues involved in these techniques. An efficient method to perform privacy preserving image filtering operations in encrypted domain over cloud is proposed in this paper. The proposed method provides facility for distributed secure processing and storage of the secret image using Shamir's secret sharing method and Chinese remainder theorem. The plain secret image is divided into number of shares in a manner that each share by itself does not reveal any meaningful information about the original image, while collectively, they retain all the information. Two secret keys are used to avail security of image shares and without knowledge of such keys, one cannot restore the secret image back. Hence, the proposed scheme is highly secured and the image can be processed without revealing any information at the cloud data center. Experimental results demonstrate that number of image filtering procedures for enhancement can be carried out efficiently in encrypted domain using the proposed method.
CryptoCT: towards privacy preserving color transfer and storage over cloud
(Springer, 2018-02) Rajput, Amitesh Singh
Current trend toward cloud computing coupled with emerging technologies such as high definition images/videos and 360-degree videos, has led the requirement of performing color transfer remotely by third party servers. However, users are always concerned about storing and processing their personal images over the cloud. Addressing this problem, we propose CryptoCT, a novel approach for privacy preserving color transfer and storage over third party cloud infrastructures. Paillier cryptosystem is employed in a manner that secret images can be processed for color transfer without revealing any information. Unlike the previous methods which involve multiple cloud servers, we use a single cloud server to accomplish the task of encrypted domain color transfer. We show that same color transfer effects as of the existing methods in plain domain are achieved in encrypted domain using our approach. To the best of our knowledge, CryptoCT is among the first known ventures to perform the task of color transfer in encrypted domain. Experimental results and security analysis validates the correctness of our approach.
CryptoLesion: A Privacy-preserving Model for Lesion Segmentation Using Whale Optimization over Cloud
(ACM Digital Library, 2020) Rajput, Amitesh Singh
The low-cost, accessing flexibility, agility, and mobility of cloud infrastructures have attracted medical organizations to store their high-resolution data in encrypted form. Besides storage, these infrastructures provide various image processing services for plain (non-encrypted) images. Meanwhile, the privacy and security of uploaded data depend upon the reliability of the service provider(s). The enforcement of laws towards privacy policies in health-care organizations, for not disclosing their patient’s sensitive and private medical information, restrict them to utilize these services. To address these privacy concerns for melanoma detection, we propose CryptoLesion, a privacy-preserving model for segmenting lesion region using whale optimization algorithm (WOA) over the cloud in the encrypted domain (ED). The user’s image is encrypted using a permutation ordered binary number system and a random stumblematrix. The task of segmentation is accomplished by dividing an encrypted image into a pre-defined number of clusters whose optimal centroids are obtained by WOA in ED, followed by the assignment of each pixel of an encrypted image to the unique centroid. The qualitative and quantitative analysis of CryptoLesion is evaluated over publicly available datasets provided in The International Skin Imaging Collaboration Challenges in 2016, 2017, 2018, and PH2 dataset. The segmented results obtained by CryptoLesion are found to be comparable with the winners of respective challenges. CryptoLesion is proved to be secure from a probabilistic viewpoint and various cryptographic attacks. To the best of our knowledge, CryptoLesion is first moving towards the direction of lesion segmentation in ED.
Distributed information fusion for secure healthcare
(Elsevier, 2024) Rajput, Amitesh Singh
Recent years have seen a significant increase in the demand for cutting-edge healthcare systems. With the rising potential of artificial intelligence and big data technology, all sectors, especially the healthcare sector, have been greatly supported. Huge amounts of privacy-sensitive clinical data are being generated from several sources. While processing these enormous amounts of diverse healthcare data, the problem is that the data are heterogeneous. The data vary with respect to the patient population, environment, data source, size, complexity, medical procedures, and treatment protocols at individual medical centers. This creates the need for a central knowledge base in the healthcare setting. Federated learning-based fusion techniques can turn out to be beneficial to acquire knowledge from these distributed data. This will bring the distributed data together into a single view that can help hospitals and health workers to obtain new insights and helps secure patients' personal information and safeguards them from information leakage. If the distribution of data among the classes is skewed or biased, the distribution is said to be imbalanced. This chapter discusses problems associated with imbalanced and heterogeneous healthcare data and their effects on machine learning models and proposes methods to improve data fairness in a distributed healthcare system using federated learning
Efficient single image haze removal using CLAHE and Dark Channel Prior for Internet of Multimedia Things
(Elsevier, 2022) Rajput, Amitesh Singh
The immersive growth of multimedia sensors in conjunction with IoT in smart cities has increased the requirement of efficient multimedia computing. One such example is the degradation of images during bad weather and their efficient processing to help surveillance, driving assistance, etc. Haze removal is an important procedure to avoid ill-condition visibility of the captured images. An optimized haze removal technique helps the local administrative authorities by integrating multimedia sensors with the IoT to improve the quality of life. In this chapter, we analyze the well-known single image dehazing technique – Dark Channel Prior (DCP) in terms of getting better image quality and optimized computational time. The DCP method is stable and a benchmark for dehazing. However, we found that the haze removal effects obtained by the DCP can be further enhanced by preprocessing the underlying image for contrast enhancement. Also, the computational time is improved with respect to the underlying refinements. Considering the dehazing quality and computational efficiency, we analyze the image quality and demonstrate that it can be further improved with a better preprocessing mechanism. The experiments are conducted over a wide variety of haze images from standard datasets and sufficient improvements over the original DCP method are demonstrated.
EraisNET: An Optical Flow based 3D ConvNET for Erasing Obstructions
(IEEE, 2022) Narang, Pratik; Rajput, Amitesh Singh
Images captured from behind a fence, window, or during rain generally face occlusions. Though prior works have addressed the problems of individually de-raining, reflection, and occlusion removal, a common approach that removes all the obstruction has found little attention in the literature. In this paper, we address the image occlusion problem by proposing a deep learning-based approach wherein the proposed method uses motion differences between two images and extracts important moving features from videos to separate the background and the obstruction. To accomplish this task, a novel 3D-convolution architecture is introduced, which is trained with synthetically blended videos. We have used learned layer-based CNN methods combined with dense-optical flow with generative networks for better output images. Moreover, a dataset for obstruction removal with sequences for reflection and fencing removal is proposed. The proposed approach is well experimented over a different variety of images and is found as a good candidate against state-of-the-art schemes.
Example Based Privacy-Preserving Video Color Grading
(Springer, 2019) Rajput, Amitesh Singh
The integration of cloud computing and smart multimedia gadgets has made an attractive business model today. However, data privacy is one of the major concern when moving to third party driven infrastructures like cloud. Furthermore, due to diverse camera sensors, the captured multimedia may contain insufficient lightning/colors and processing them manually is a painstakingly task. A few schemes have been proposed to address this problem, however they suffered from the major drawback of computational and storage overhead, and becomes non-trivial in case of videos. Considering these challenges, we propose an automatic video color grading approach in this chapter. The proposed approach enables cloud data center to process encrypted multimedia data by transferring its colors as per an example image as the reference. We analyze the correlation between consecutive video sequences and propose to evaluate the color transformation parameters for every alternate video frame. In addition, proxy encryption based Paillier cryptosystem has been used for video encryption. As a result, the computational and storage overheads are drastically reduced with effective video grading results. The feasibility and robustness of the proposed approach are validated through various tests.
Fire Detection Using Dense Trajectories
(Springer, 2018) Rajput, Amitesh Singh
This paper proposes an automatic computer vision-based system for fire detection in videos. There are many previous methods for video-based fire detection but only very few of them have considered the challenge of camera motion or motion of the background scene while finding features based on motion of fire. Our method is divided into two phases. First, we train our system for color characteristics of fire with the help of Gaussian mixture model (GMM), and for texture features which are computed using local binary patterns (LBPs). Next, dense trajectories are computed for motion features which are free from camera motion or challenges of moving scene. Bounding boxes are detected with the help of color and texture models. Subsequently, dense trajectories are projected onto codebooks for feature vector computation, and chi-square kernel-based SVM is employed for classification of fire and non-fire motion representations. Quantitative evaluation of our method indicates the fitness of temporal features for fire detection.
Generation of Multi-Layered QR Codes with Efficient Compression
(Springer, 2023-12) Rajput, Amitesh Singh
In this modern data-centric era, securing information that is transferred over the internet is a serious concern. Various cryptographic and stenographic methods play a vital role in this purpose. In this work, we develop a multi-layered steganographic system capable of hiding payloads in different hidden layers inside QR codes. Furthermore, each layer is compressed and encrypted separately to secure the payload data. Hence, we use the QR (Quick Response) Code as the cover medium to achieve and implement the above principles. In the proposed multi-layered QR Code approach, the top layer works as a standard QR Code while the layers beneath it act as private data. The secret is efficiently compressed to accommodate more information within the same space provided, thereby increasing the holding capacity of the QR code. The data in each layer are retrieved using different methods according to the type of compression used during the encoding process of the specified layer. The proposed model can store up to nine hidden layers of data. Various robustness tests are carried out to validate the efficiency of the hidden layers in the QR code and to check whether the hidden layers can withstand alterations without any data loss. Importantly, the resulting QR code works identically to a normal version upon scanning with any typical QR code scanner.
A Novel Image Encryption and Authentication Scheme Using Chaotic Maps
(Springer, 2014) Rajput, Amitesh Singh
The paper presents an amalgam approach for image encryption and authentication. An ideal image cipher should be such that any adversary cannot modify the image and if any modifications are made, can be detected. The proposed scheme is novel and presents a unique approach to provide two level security to the image. Hashing and two chaotic maps are used in the algorithm where hash of the plain image is computed and the image is encrypted using key dependent masking and diffusion techniques. Initial key length is 132-bits which is extended to 148-bits. Performance and security analysis show that the proposed scheme is secure against different types of attacks and can be adopted for real time applications.
Privacy Preserving Image Scaling Using 2D Bicubic Interpolation Over the Cloud
(IEEE, 2018) Rajput, Amitesh Singh
This paper presents an efficient scheme for privacy preserving image scaling over the cloud. The proposed scheme advances the emerging trend of privacy preserving cloud computing and supports desired operations required for image scaling in the encrypted domain. We use scaling and randomization followed by modulo operation for generation of image shares and employ 2D bicubic interpolation for scaling user images in the encrypted domain. The previous scheme used bilinear interpolation utilizing four neighboring pixel intensity values to approximate intermediate ones while scaling the image. On the contrary, we use sixteen neighboring pixel intensity values utilizing bicubic interpolation by leveraging considerable efforts to support the processing of user images over cloud servers. The proposed scheme is validated through various experiments followed by a comparison with the existing scheme. Additionally, security analysis demonstrates the robustness of the proposed scheme under various attack scenarios.
Privacy-Preserving Distribution and Access Control of Personalized Healthcare Data
(IEEE, 2022) Rajput, Amitesh Singh
The popularity of wearable smart healthcare devices has led to the emergence of a new service paradigm. However, in order to improve service quality, the manufacturers and online service providers collect massive data. This is a big concern as medical data are extremely sensitive. A few schemes have been proposed to overcome this problem. But, they suffer from security risks and overall increased complexity. Also, there is no implicit entity authentication and data integrity involved. We address these problems by allowing rectified data access through a directing authority, known as the transcryptor, using polymorphic encryption. Entity authentication and data integrity are achieved by smartly organizing data access and key information packets. The performance of the proposed approach is tested over different modalities data with varying sizes, whereas the security analysis is demonstrated using a challenge-response game model. The comparison with the state-of-the-art schemes illustrates superiority of the proposed approach.
Privacy-preserving human action recognition as a remote cloud service using RGB-D sensors and deep CNN
(Elsevier, 2020-08) Rajput, Amitesh Singh
Cloud-based expert systems are highly emerging nowadays. However, the data owners and cloud service providers are not in the same trusted domain in practice. For the sake of data privacy, sensitive data usually has to be encrypted before outsourcing which makes effective cloud utilization a challenging task. Taking this concern into account, we propose a novel cloud-based approach to securely recognize human activities. A few schemes exist in the literature for secure recognition. However, they suffer from the problem of constrained data and are vulnerable to re-identification attack, where advanced deep learning models are used to predict an object’s identity. We address these problems by considering color and depth data, and securing them using position based superpixel transformation. The proposed transformation is designed by actively involving additional noise while resizing the underlying image. Due to this, a higher degree of obfuscation is achieved. Further, in spite of securing the complete video, we secure only four images, that is, one motion history image and three depth motion maps which are highly saving the data overhead. The recognition is performed using a four stream deep Convolutional Neural Network (CNN), where each stream is based on pre-trained MobileNet architecture. Experimental results show that the proposed approach is the best suitable candidate in “security-recognition accuracy (%)” trade-off relation among other image obfuscation as well as state-of-the-art schemes. Moreover, a number of security tests and analyses demonstrate robustness of the proposed approach.
A privacy-preserving protocol for efficient nighttime haze removal using cloud based automatic reference image selection and color transfer as a service
(Elsevier, 2020-01) Rajput, Amitesh Singh
The advanced Internet technologies have migrated the people to rejoice a virtual environment known as cloud computing. The user can avail the desired services on a pay-as-you-go model, without worrying about the burden of infrastructure maintenance. However, privacy is one of the major issues in cloud computing. This issue is further widened for highly confidential multimedia data like surveillance images and videos. In the context of cloud based smart multimedia systems, it has been found that due to inconsistent weather conditions, there is a usual requirement of post-processing the captured multimedia for better appearance. However, privacy related concerns are resisting users to move their data to the cloud. One such problem is addressed in this paper, specializing the task of efficient nighttime haze removal using privacy-preserving cloud based automatic reference image selection and color transfer as a service. Different from daytime conditions, nighttime haze image consists of multiple light sources, which makes an ambiguous situation for haze removal. We address this problem by first selecting an appropriate gray image as the reference and then transferring its colors to nighttime haze image. This makes the transformed image a suitable candidate for radiance recovery. The proposed protocol is designed to securely outsource this considerable burden from user end. We accomplish this by first proposing an automatic reference gray image selection method, followed by efficient handling mechanisms for technical challenges arising due to performing color transfer operations securely over cloud. Experimental results and validation demonstrates superiority of the proposed method over state-of-the-art schemes. Security analysis of the proposed protocol is established through a challenge-response game model.
Privacy-preserving smart surveillance using local color correction and optimized ElGamal cryptosystem over cloud
(IEEE, 2019) Rajput, Amitesh Singh
The emergence of cloud computing in integration with smart multimedia devices has created an attractive business model today. However, due to the involvement of third party servers, there is a risk of privacy for highly confidential data like surveillance images/videos. Moreover, due to inconsistent lightning conditions, there is a usual requirement of post-processing the captured multimedia for better appearance. Addressing these problems, we propose a novel cloud based privacy-preserving approach for image color enhancement in this paper. Unlike existing color correction schemes, where colors of the test image are processed in plain domain with visible image contents, we propose to perform color correction operations in the encrypted domain over cloud. As a consequence, superior results are achieved along with complete privacy assurance. In addition, we propose a block-based image encryption method using logistic-tent system and ElGamal cryptosystem. As a result, size of the encrypted image is significantly reduced as compared to the naive approach. Experimental results are performed under various tests and the proposed approach is found to be highly effective as compared to state-of-the-art schemes. Moving ahead, security strength of the proposed approach is demonstrated through a challenge response game model.