Multi-Focus Image Fusion (MFIF) plays an important role in the field of computer vision. It aims to merge multiple images that possess different focus depths, resulting in a single image with a focused appearance. Though deep learning based methods have demonstrated development in the MFIF field, they vary significantly with regard to fusion quality and robustness to different focus changes. This paper presents the performance analysis of three deep learning-based MFIF methods specifically ECNN (Ensemble based Convolutional Neural Network), DRPL (Deep Regression Pair Learning) and SESF-Fuse. These techniques have been selected due to their publicly availability of training and testing source code, facilitating a thorough and reproducible analysis along with their diverse architectural approaches to MFIF. For training, three datasets were used ILSVRC2012, COCO2017, and DIV2K. The performance of the techniques was evaluated on two publicly available MFIF datasets: Lytro and RealMFF datasets using four objective evaluation metrics viz. Mutual Information, Gradient based metric, Piella metric and Chen-Varshney metric. Extensive experiments were conducted both qualitatively and quantitatively to analyze the effectiveness of each technique in terms of preserving details, artifacts reduction, consistency at the boundary region, texture fidelity etc. which jointly determine the feasibility of these methods for real-world applications. Ultimately, the findings illuminate the strengths and limitations of these deep learning approaches, providing valuable insights for future research and development in methodologies for MFIF.

Due to an aggressive development and covert transmission of the computer users over the web, the steganography is acquiring its vogue day by day. It is the method to encode the covert data in a transmission medium in a fashion that the actuality of data must remain hidden. To protect the integrity of the data over the internet becomes necessary for the high sensitivity communications or the data transfers. The steganography carries a number of options for the embedding of the secret data into the cover data. In this manuscript, we have proposed the hybrid representation for the embedding of the data, which utilizes the random pattern embedding along with the cryptography for the higher steganography security levels. The proposed model has been designed to enhance the security level by utilizing the sequential data encoding and decoding in the video data. The video embedding creates the higher level of security for the embedded data. The experimental results have been obtained in the form of the embedding capacity, mean squared error (MSE) and Bits per Pixel (BPP). The proposed model has been found to be efficient enough for the video steganography.