Mobile edge computing (MEC) plays a crucial role in meeting the demands of the future digitized world by providing widespread computational capabilities and seamless integration with emerging technologies such as 6G, Internet of Things (IoT), blockchains, and artificial intelligence (AI). This paper proposes a novel edge and split computing architecture orchestrated by cooperating machine learning elements across network elements. The algorithm optimizes task partitioning, accuracy, and transmission delay in dynamic environments. Evaluations show its superiority in reducing task calculation failures and achieving high precision. Potential applications include smart city domains like autonomous driving, car sharing, robotic delivery, and public transportation.

The paper presents an improved method of network intrusion detection using machine learning based approach. The study trained four types of deep neural network models, including two feed-forward dense models and two models that utilized longshort term memory layers. The study compares the effectiveness of models predictions based on the newly developed dataset UNSW-NB15. A suitable part of the dataset was selected after evaluating its integrity, and the selected data was preprocessed and formatted for neural network training and evaluation. Two models consisting of dense layers were developed, trained, and fine-tuned to optimize their accuracy. The LSTM models were also implemented and fine-tuned accordingly. The accuracy of models’ predictions was evaluated using a portion of the dataset, the validation set, which the model had not seen in training. Without weight balancing, the dense-layer-based model attained a validation accuracy of 78.94% while the LSTM-based model achieved 76.84%. With weight balancing, the accuracy improved to 79.34% for the dense-layer model and 79.21% for the LSTM model. Significant output differences were found when analysing prediction correctness of confusion matrices.

Steganography deals with a concealing the secret message into any type of multimedia information, such as audio, video, images and text. There are several steganographic techniques, among which the cover selection method belongs to. Key point of the cover selection method is to find an optimal cover image from a cover image database. Optimal image is an image in which an embedding process causes minimum substitution changes. The goal of the proposed method is to scan all possible positions to secret message embedding in the cover image and to perform the same scanning for all cover images in the database. It brings much more comparative calculations. Thus, the method achieves better results in number of changes after an embedding process. On the other hand, main drawback is a worse time consumption. Such a method is called cover selection steganography with intra-image scanning.

Image files may contain portions of sensitive or intimate content that should not be visible to everyone. This paper deals with various methods of image processing and coding, to protect privacy through face identification. We proposed an encryption procedure implemented in software that allows face detection and subsequent blurring of selected fragments by DCT image transformation while preserving the details needed for successful reconstruction within the same image. A blind steganographic method was used for information embedding into the source cover image. This protected image is accessible to be viewed for ordinary viewers and a dedicated decoder is required for content restoration. The decoder ensures that only authorized persons are allowed to see the content without censorship. The conclusion of the paper focuses on the quality evaluation of the proposed method based on observed statistical parameters.

Novel image steganalytic method used to detection of secret message in static images is introduced in this paper. This method is based on statistical steganalysis (SS), where statistical vector is composed by 285 statistical features (parameters) extracted from DCT (Discrete Cosine Transformation) domain and 46 features extracted mainly from DWT (Discrete Wavelet Transformation) domain. Classification process was realized by Ensemble classifier that was helpful in reduction of computational and time complexity. Proposed steganalytic method was verified by detection of popular image steganographic methods. Novel method was also compared with existing steganalytic methods by overall detection accuracy of a secret message.

Information hiding represents a constantly growing field mainly consisting of digital watermarking and steganography. Both deal with inserting additional information into multimedia. Basically, watermarking protects the multimedia content against unauthorized manipulation whereas steganography utilizes multimedia in the role of a cover object for unauthorized secret message transmission. This article deals with additional technique to steganography which can decrease an impact to cover image after an embedding process. The goal is to find an optimal image from a cover image database based on bits of the secret message. With 2,000 image database, cover selection method can decrease impact by 10% in comparison to the same embedding method without cover selection. Disadvantage lays in a time consumption. In this article, there is proposed a novel accelerated cover selection method. Acceleration is achieved by optimal shortening of a secret message for comparative process.

The main role of steganalysis is a successful detection of secret communication.This communication is exclusively created by steganography.Steganographic methods deals with hiding a secret information into any type of multimedia data, for example to static images.Among basic requirements to steganographic systems belongs the perceptual transparency.Inserted information is perceptually transparent if an average subject is unable to distinguish any difference between data before and after embedding process.Nevertheless, each steganographic method necessarily causes some change in some statistical parameter.It represents the basis for building a successful steganalyzer.In this article are tested the impact of four steganographic methods to the selected statistical parameters which are usually utilized in the image objective quality assessment.Specifically, peak signal-to-noise ratio, normalized cross correlation, a local histogram of DCT coefficients and sample variance.The contribution of the article consists in the usage of results in the theory of statistical vector creation in building the particular image steganalytic method.

Information hiding represents an essential part of secure communication. Except for cryptography there is another approach, steganography. Methods of steganography deal with an embedding the confidential information into any modality of multimedia information in such a way that third party is not able to recognize caused modification. Thus there are many sophisticated methods how do build the system more secure. One of them is performing an embedding process in transformation domain. This work utilizes domain after discrete wavelet transformation. However, on the other hand there are effective steganalytic tools determined to detect the presence of embedded information by steganography. Nowadays steganalysis utilizes an effective classifiers and well-chosen sets of statistical features thus it is difficult task for steganography to establish new approaches to build as secure system as possible. One of such methods is proposed in this paper. It is based on compering secret message bits with bits of large databases of images. After that comparing is image with the most correlation chosen. It brings reduction of number of changes caused by the embedding process.

This paper presents a robust block-based watermarking scheme for multimedia copyright protection. In this work, we used benefits of domain combination of Singular Value Decomposition (SVD) and Discrete Cosine Transformation (DCT). Our aim was to utilize the favorable properties of both transformations to design of a unique and useful transformation standard algorithm for embedding watermarks into cover data. The proposed watermarking scheme is verified from the point of view of imperceptibility expressed by objective quality measure PSNR (Peak Signal-to-Noise Ratio) and robustness to different types of removal (JPEG compression, Gaussian filtering, Sharpening) and geometric attacks (Scaling and Rotating).

This paper is focused on proposal of image steganographic method that is able to embedding of encoded secret message using Quick Response Code (QR) code into image data. Discrete Wavelet Transformation (DWT) domain is used for the embedding of QR code, while embedding process is additionally protected by Advanced Encryption Standard (AES) cipher algorithm. In addition, typical characteristics of QR code was broken using the encryption, therefore it makes the method more secure. The aim of this paper is design of image steganographic method with high secure level and high non-perceptibility level. The relation between security and capacity of the method was improved by special compression of QR code before the embedding process. Efficiency of the proposed method was measured by Peak Signal-to-Noise Ratio (PSNR) and achieved results were compared with other steganographic tools.

The aim of this work is to present a real and useful application of steganography systems in data communication networks. Different from all the previous work analyzing the traditional steganography system, we introduce the possibility of the TCP protocol mechanism usage in steganography based on intentionally triggered protocol data unit (PDU) retransmission. It describes how to create a subliminal channel in the network and its usage for steganography purposes, so called network-based steganography, which is relatively a new research object in the field of information hiding. We analyzed the way of creating a hidden steganography channel via PDU retransmission in a real network. The achieved results of the proposed method are in practical part and in the end the recovered qualitative benefits of simulations are situated.

Steganography is the process of implanting secret message in a cover data without causing degradation neither to the cover information nor to the secret message implanted in the cover data. Most of the steganographic techniques are applied on images, texts, and protocols. In this paper a novel algorithm of image steganography is proposed to implant a secret text message into a cover image using YCbCr color space model and 2D Haar Discrete Wavelet Transform. In this proposed technique, input text in ASCII code is encrypted by AES what ensures, that the relative letter frequency of a plain text secret message will be disturbed. Proposed technique also solves conversion between color space models RGB and YCbCr in spite of modification in component Cb by the secret message. This step allows extracting of secret message without errors what is a very important characteristic, because change of only one bit causes extracting of different character. Performance of the proposed algorithm has been gauged by the peak signal to noise ratio (PSNR) and mean square error (MSE) for the stego image.

Steganography is the science of hiding secret information in another unsuspicious data. Generally, a steganographic secret message could be a widely useful multimedia: as a picture, an audio file, a video file or a message in clear text - the covertext. The most recent steganography techniques tend to hide a secret message in digital images. We propose and analyze experimentally a blind steganography method based on specific attributes of two dimensional discrete wavelet transform set by Haar mother wavelet. The blind steganography methods do not require an original image in the process of extraction what helps to keep a secret communication undetected to third party user or steganalysis tools. The secret message is encoded by Huffman code in order to achieve a better imperceptibility result. Moreover, this modification also increases the security of the hidden communication.

In this paper, proposed steganalytic method utilized for the detection of secret message is based on extraction of statistical features from cover and stego images in JPEG file format together with calibration technique. The steganalyzer concept uses Support Vector Machines (SVM) classification or Bayes classifier for training a model that is later used by the same steganalyzer in order to identify between a clean (cover) and stego image. The aim of the paper was to compare detection accuracy (ACR) of the trained models for two types of classifiers: Support Vector Machines and Bayes classifier. In this paper, five models created between cover and stego images (images obtained by nsF5, Model Based 1, Model Based 2, Modulo Histogram Fitting with Dead Zone and Pertubed Quantization steganographic method) was tested.

In this paper we propose robust watermarking schema for video files based on spread spectrum theory. The main concept relies on utilization of time as temporal dimension for spreading of watermark bits that are assigned to video frames. The second level spreading that is impartial to first level, uses different set of PN sequences within the video frames what allows to utilize multiple watermarking methods designed for still images simultaneously. Two dimensional spread spectrum watermarking framework allows acquiring a high level of robustness with desired imperceptibility of hidden watermark. Two independent embedding methods LSB and DCT were tested during embedding as representative techniques of different embedding domain. The presented results show that proposed watermarking schema is able to withstand intentional attacks. The proposed system is robust and adaptive to recent and newly developed embedding technique, which could provide a requisite of robustness and protection against copyright infringement.

The Elliptic Curve Cryptosystem is an emerging alternative for traditional Public-Key Cryptosystem like RSA, DSA and DH. It provides the highest strength-per-bit of any cryptosystem known today with smaller key sizes resulting in faster computations, lower power consumption and memory. It also provides a methodology for obtaining high-speed, efficient and scalable implementation of protocols for authentication. The objective is to give the reader an overview on efficient addition and doubling formulas of Edwards curves together with analysis and effective parallel decomposition of these formulas. Practical analysis is provided with implementation consideration.

Since JPEG images have been widely used in our daily life, the steganalysis for JPEG images becomes very important and significant. The Article addresses steganalysis in static images based on DCT transformed region, able to recognize the most popular steganography algorithms occurring on the Internet. We propose a new steganalysis method, where statistical properties of the image are explored, regardless the embedding procedure employed. The feature set used for classification of images consists of 285 statistical features. Experimental results show that in comparing with the universal steganalysis method for JPEG stego images, our method improves detection of widely used steganographic method in detection process, which provides observable differences in investigation performance.

This paper is focused on comparison of two steganalytic methods that are able to detect embedding of secret message using popular and novel steganographic algorithms in JPEG images. First one is based on binary similarity measures and second method exploits 66 or 274 statistical features in transform domain from cover and stego images. The tested universal model was trained using novel steganographic method previously proposed by the same authors. Subsequently, the obtained statistical parameters from cover and stego images are used for training of model that is applied on detection of secret message in testing phase. The aim of this paper was to examine the accuracy detection (ACR) of these novel universal steganalytic models for popular steganographic tools in static images and also to compare time consumption in process of extraction parameters and training of model.

In this paper we propose two dimensional spread spectrum watermarking framework based on Direct Spread Spectrum theory using PN sequences. The presenting schema allows acquiring a high level of robustness with desired imperceptibility. The aim of this paper was to present robust and adaptive watermarking system for multimedia protection, where hidden watermark provides a desired requisition against copyright infringement. Media content delivery system incorporated by cryptographic public-key was proposed in the paper as part of. The framework concept was applied to digital video content to simulate 2D-level spreading technique of hidden watermark data.

This paper presents results of multi-classification and cross-validation of tested steganalysis method in static images in JPEG format. Steganalysis methods are used for revealing a secret communication conducted by different steganographic tools. The steganalysis algorithm analyzes changes in statistical parameters of the images using Feature Based Steganalysis. The multi-classification is ability of proposed system to identify an applied steganography methods and cross-validation is defined as steganalytic model's detection efficiency of steganography methods that were not used in training phase of the model. Testing was performed for different length of statistical features' vector and for different size of embedded secret message. The results are also contributing in design of blind steganography system to detect a new steganography tools.