People focus on the intelligent healthcare system expecting sufficient medical facilities even from a remote location. However, to make the healthcare system more trustworthy, secure access control plays a vital role in resisting several cyber-attacks. Security threat on medical data is highly sensitive since it is associated with life risks. When a user accesses a healthcare system through the Internet, the main concern is preventing unauthorized access to vital healthcare-related data. Traditional authentication (based on a password, smartcard, and/or biometric) provides a solution to allow such kind of application only to the authorized users. But once a user is authenticated, and he/she is idle for a while, the chances of security threat are reasonably high. This article has designed an intelligent user recognization mechanism for continuous monitoring of the user’s activity throughout the session. We have made the system user-friendly through single sign-on and eliminated trusted third-party dependence to avoid data breaches. Once the authentication is performed, then in a later stage, the continuous monitoring mechanism based on machine learning techniques helps to decide the user’s unique behavior allowing the application server to make better decisions regarding the user’s authenticity in the current session. The security analysis shows that the proposed system is more secure than the traditional authentication mechanisms.

Steganography is an area where researchers focus on how to develop novel hiding techniques that are aided by the mathematical or computational model. In order to conceal the secret bits in a cover image, the imperceptible property and embedding payload capacity must be maintained. In most of the developed methods, some selective least significant bits (LSBs) of each pixel of the cover image is modified to hide the data which may lead to substantial distortion in the stego-image especially when the payload size is large. Therefore, to attain the trade-off between payload capacity and image quality, the authors have devised a binary lower triangular matrix-based secret message embedding process in which data can be hidden in the color pixel. In this secret message embedding process, the secret data is embedded aided by the binary lower triangular matrix. The formulated mathematical model for the secret message embedding process has ensured a secure embedding technique with reduced aberration on the stego-image. The proposed steganography scheme is preserving significant visual quality while emphasizing the security aspect by adopting an indirect data-hiding mechanism in contrast to the conventional mechanism. Furthermore, for the betterment of the visual properties of the stego-image, we have generically extended the formulation and tested it to check the visual quality. The empirical results of the proposed steganography scheme depict outcomes satisfactorily for the RGB color images. Researchers can explore and apply this novel data-hiding scheme in different data communication applications.

Medical images have become an integral part of healthcare systems to provide quick and accurate treatment of diseases. The challenge, however, is to store these data, as they are huge in volume. DNA provides an efficient medium for the storage of bulk data. This data can be secured against malicious use using DNA based encryption algorithms. This paper presents a novel DNA based compression-encryption algorithm, specially designed for medical images. A significant part of medical images contains homogeneous pixels, which are even more prominent when decomposed into their bit-planes. This redundancy can be removed while compression, based on which the proposed DNA-based QuadTree Decomposition algorithm has been designed. A sequence of four DNA nucleotides represents the entire image, and these can be used to recover back the original image using the decoding algorithm. To secure the obtained sequences, DNA-based Advanced Encryption Standard (AES) algorithm is applied in Cipher Block Chaining (CBC) mode which encrypts the sequences using the symmetric key and initial vector parameters. The image cannot be recovered from these encrypted sequences unless decrypted using the correct key, and the algorithm for AES is secured against cryptographic attacks.

Content‐based image retrieval (CBIR) uses primitive image features for retrieval of similarimages from a dataset. Generally, researchers extract these visual features fromthe whole image. Therefore, the extracted features contain overlappedinformation of texture, colour, and shape features, and it is a criticalchallenge in the field of CBIR. This problem can be overcome by extracting thecolour features from the colour as well as shape and texture features from theintensity dominant part only. In this study, the authors have proposed aniterative algorithm to separate colour and texture dominant part of the imageinto two different images. Here, a combination of edge maps and gradients hasbeen used to achieve separate colour and texture images. Further,scale‐invariant feature transform and 2D dual‐tree complex wavelet transform hasbeen realised to extract unique shape and texture features from the textureimage. Simultaneously, a probability‐based semantic centred annular histogramhas been suggested to extract unique colour features from the colour image.Finally, a novel weighted distance‐based feature comparison scheme has beenproposed for similarity matching and retrieval. All the image retrievalexperiments have been carried out on seven standard datasets and demonstratedsignificant improvements over other state‐of‐arts CBIR systems

In this paper, the authors have presented a robust and blind watermarking scheme based on discrete cosine transform (DCT) for protecting the copyright ownership of digital images. Initially, the image is decomposed into non-overlapping blocks and subsequently DCT is employed on each block to embed a binary bit of watermark into each transformed block by modifying some middle significant AC coefficients using repetition code. During the embedding phase of the proposed method, DC and some higher AC coefficients are kept intact after zigzag scanning of each DCT block to ensure the high visual quality of watermarked image. This scheme is suitable to protect the copyright information even in compressed form of the watermarked image since it exploits middle bands of DCT coefficients for embedding the watermark bits. The proposed scheme is also tested to verify the withstand capability against several image processing attacks and satisfactory results are achieved.

Among the various watermarking schemes, the reversible watermarking scheme has drawn extensive attention in the recent years for its application in sensitive issues like medical, military and typical law-enforcement images. Cover image dependent embedding capacity and lack of robustness are the most crucial concerns of the reversible watermarking methods. To overcome these issues, a discrete cosine transform (DCT) and histogram shifting based robust reversible image watermarking scheme using Arnold scrambling is presented in this paper. Initially, the image is decomposed into non-overlapping blocks and consequently DCT are applied to each block to embed a binary bit of watermark into each transformed block by modifying one pair of middle significant AC coefficients. In this initial step, location map is also generated for the cover image restoration purpose in the extracting side. Then, this location map is embedded in the cover image using histogram modification technique. In the receiver side, at first location map is generated from an image using histogram modification method and watermark is recovered from the corresponding image. Using location map reversible image is reversed in the following extracting phase of the proposed method. The proposed reversible watermarking scheme has also been experimented to verify the robustness property against several image processing attacks and satisfactory results are achieved.

Content-based image retrieval (CBIR) scheme has gained popularity in the field of information retrieval for retrieving some relevant images from the image database based on the visual descriptors such as color, texture and/or shape of a given query image. In this paper, color features have been exploited from each color component of an RGB color image by using multi-resolution approach since most of the information of an image is undetected at one resolution level while some other undetectable information is visualized in other multi-resolution levels. Initially, Gaussian image pyramid is employed on each color component of the color image and subsequent DCT is computed directly on the obtained multi-resolution image planes. Then some significant DCT coefficients are selected according to the zigzag scanning order. For formation of the feature vector, we have derived some statistical values from AC coefficients and all other DC coefficients are included entirely. Finally, a similarity measure is suggested during image retrieval process and it is found that the overall computation overhead is reduced due to consideration of the proposed similarity measure. The proposed CBIR scheme is validated on a two standard Corel-1K and GHIM-10K image databases and satisfactory results are achieved in terms of precision, recall and F-score. The retrieved results show that the proposed scheme outperforms significantly over other related CBIR schemes.