With the growing demand for digitalization, multimedia data transmission through wireless networks has become more prominent. These multimedia data include text, images, audio, and video. Therefore, a secure method is needed to modify them so that such images, even if intercepted, will not be interpreted accurately. Such encryption is proposed with a two-layer image encryption scheme involving bit-level encryption in the time-frequency domain. The top layer consists of a bit of plane slicing the image, and each plane is then scrambled using a chaotic map and encrypted with a key generated from the same chaotic map. Next, image segmentation, followed by a Lifting Wavelet Transform, is used to scramble and encrypt each segment’s low-frequency components. Then, a chaotic hybrid map is used to scramble and encrypt the final layer. Multiple analyses were performed on the algorithm, and this proposed work achieved a maximum entropy of 7.99 and near zero correlation, evidencing the resistance towards statistical attacks. Further, the keyspace of the cryptosystem is greater than 2¹²⁸, which can effectively resist a brute force attack. In addition, this algorithm requires only 2.1743 s to perform the encryption of a 256 × 256 sized 8-bit image on a host system with a Windows 10 operating system of 64-bit Intel(R) Core(TM) i5-7200U CPU at 2.5 GHz with 8 GB RAM.

Retinal blood vessel segmentation influences a lot of blood vessel-related disorders such as diabetic retinopathy, hypertension, cardiovascular and cerebrovascular disorders, etc. It is found that vessel segmentation using a convolutional neural network (CNN) showed increased accuracy in feature extraction and vessel segmentation compared to the classical segmentation algorithms. CNN does not need any artificial handcrafted features to train the network. In the proposed deep neural network (DNN), a better pre-processing technique and multilevel/multiscale deep supervision (DS) layers are being incorporated for proper segmentation of retinal blood vessels. From the first four layers of the VGG-16 model, multilevel/multiscale deep supervision layers are formed by convolving vessel-specific Gaussian convolutions with two different scale initializations. These layers output the activation maps that are capable to learn vessel-specific features at multiple scales, levels, and depth. Furthermore, the receptive field of these maps is increased to obtain the symmetric feature maps that provide the refined blood vessel probability map. This map is completely free from the optic disc, boundaries, and non-vessel background. The segmented results are tested on Digital Retinal Images for Vessel Extraction (DRIVE), STructured Analysis of the Retina (STARE), High-Resolution Fundus (HRF), and real-world retinal datasets to evaluate its performance. This proposed model achieves better sensitivity values of 0.8282, 0.8979 and 0.8655 in DRIVE, STARE and HRF datasets with acceptable specificity and accuracy performance metrics.

With the growing internet technology, science and requirement of concealing defense research work, there is a need for highly secured information exchange, which is the essence of steganography. Classic secret information could be undermined or even faked thereby, creating a menace to the core of secrecy. Steganography camouflages the hidden information into an unsuspicious digital file i.e. image, video or audio, thereby covering the existence of it. Using modified LSB substitution and readjustment procedure the mean square error has been reduced. Pixel indicator is used for increasing the embedding capacity of secret data. The security of secret data embedded into a cover file can be increased using random walk inside the file. In this paper we have used knight's tour for random walk, by not affecting the image quality. For high security while embedding, we randomize the three planes of RGB cover image using row vector, divide the image into four pixel blocks and then use Pixel value differencing (PVD) to embed data adaptively followed by knight's tour to select next block for embedding. This method gives highly secured and high capacity steganography