Deepfake is a type of artificial intelligence used to create convincing images, audio, and video hoaxes and it concerns celebrities and everyone because they are easy to manufacture. Deepfake are hard to recognize by people and current approaches, especially high-quality ones. As a defense against Deepfake techniques, various methods to detect Deepfake in images have been suggested. Most of them had limitations, like only working with one face in an image. The face has to be facing forward, with both eyes and the mouth open, depending on what part of the face they worked on. Other than that, a few focus on the impact of pre-processing steps on the detection accuracy of the models. This paper introduces a framework design focused on this aspect of the Deepfake detection task and proposes pre-processing steps to improve accuracy and close the gap between training and validation results with simple operations. Additionally, it differed from others by dealing with the positions of the face in various directions within the image, distinguishing the concerned face in an image containing multiple faces, and segmentation the face using facial landmarks points. All these were done using face detection, face box attributes, facial landmarks, and key points from the MediaPipe tool with the pre-trained model (DenseNet121). Lastly, the proposed model was evaluated using Deepfake Detection Challenge datasets, and after training for a few epochs, it achieved an accuracy of 97% in detecting the Deepfake
Use of Surveillance cameras in houses and markets became common, that resulted to minimize theft and make it a difficult task because it let recording and viewing what is going around. The wide application of these cameras, pushed thieves to seek new ways for abolition of the surveillance system and digital recording of events, such as cutting the signal wire between the camera and Digital video recorder or changing the direction of the camera away from the focus spot or damaging the camera or steal the device which means the loss of the recorded media. This paper focuses on such abolitions and fixed it by suggesting a way to notify the administrator immediately and automatically by Email about any violation of the system using MATLAB, whic
... Show MoreCopper zinc tin sulfide selenide, Cu2 ZnSn(S1−x Se x)4 , absorbers are promising earth-abundant and environmentally benign materials for low-cost photovoltaic applications. This study investigates the structural and optical properties of Cu 2 ZnSn(S1−x Se x)4 nanostructured thin films prepared by pulsed laser deposition using melt-quenched targets with selenium compositions x = 0.0–1.0. X-ray diffraction revealed that films with low selenium content remained amorphous, whereas higher selenium incorporation promoted the formation of polycrystalline kesterite–stannite phases with preferred orientations along (112), (200), (220), and (312). The crystallite size increased from 12.3 to 17.9 nm as selenium reached x = 1.0, indicating enha
... Show MoreThis research deals with the use of a number of statistical methods, such as the kernel method, watershed, histogram, and cubic spline, to improve the contrast of digital images. The results obtained according to the RSME and NCC standards have proven that the spline method is the most accurate in the results compared to other statistical methods.
The complexity and variety of language included in policy and academic documents make the automatic classification of research papers based on the United Nations Sustainable Development Goals (SDGs) somewhat difficult. Using both pre-trained and contextual word embeddings to increase semantic understanding, this study presents a complete deep learning pipeline combining Bidirectional Long Short-Term Memory (BiLSTM) and Convolutional Neural Network (CNN) architectures which aims primarily to improve the comprehensibility and accuracy of SDG text classification, thereby enabling more effective policy monitoring and research evaluation. Successful document representation via Global Vector (GloVe), Bidirectional Encoder Representations from Tra
... Show MoreAl-Dalmaj marsh and the near surrounding area is a very promising area for energy resources, tourism, agricultural and industrial activities. Over the past century, the Al-Dalmaje marsh and near surroundings area endrous from a number of changes. The current study highlights the spatial and temporal changes detection in land cover for Al-Dalmaj marsh and near surroundings area using different analyses methods the supervised maximum likelihood classification method, the Normalized Difference Vegetation Index (NDVI), Geographic Information Systems(GIS), and Remote Sensing (RS). Techniques spectral indices were used in this study to determine the change of wetlands and drylands area and of other land classes, th
... Show MoreDeep learning convolution neural network has been widely used to recognize or classify voice. Various techniques have been used together with convolution neural network to prepare voice data before the training process in developing the classification model. However, not all model can produce good classification accuracy as there are many types of voice or speech. Classification of Arabic alphabet pronunciation is a one of the types of voice and accurate pronunciation is required in the learning of the Qur’an reading. Thus, the technique to process the pronunciation and training of the processed data requires specific approach. To overcome this issue, a method based on padding and deep learning convolution neural network is proposed to
... Show MoreIn recent years, the field of research around the congestion problem of 4G and 5G networks has grown, especially those based on artificial intelligence (AI). Although 4G with LTE is seen as a mature technology, there is a continuous improvement in the infrastructure that led to the emergence of 5G networks. As a result of the large services provided in industries, Internet of Things (IoT) applications and smart cities, which have a large amount of exchanged data, a large number of connected devices per area, and high data rates, have brought their own problems and challenges, especially the problem of congestion. In this context, artificial intelligence (AI) models can be considered as one of the main techniques that can be used to solve ne
... Show MoreMachine learning is considered a powerful technique in many applications such as classification, clustering, recognition and prediction. Deep learning is a modern, vital and superior machine learning that gives stunning performance, especially with huge data. Stock market price prediction is the process of determining the future value of a prospect of a financial instrument traded in the market, to gain a great profit a successful prediction must be conducted, in order to achieve that machine learning is used, in this article, two approaches are proposed to predict the stock market prices and movement using two datasets, the first approach employs two machine learning models (J48 & logistic regression) while the second approach based on rec
... Show MoreThe digital revolution had greatly affected the methods through which we communicate, starting from the basic concepts of the internet technology and the web content in addition to the important issues that concern the culture of the digital media, the internet governance and the variation in the digital age in general and the graphic and internal design in particular.
This research addresses an important topic that goes along with the scientific development in the field of the digital design, especially in the internal and graphic designs. This study consists of two sections: the first includes the problem of the study and the need for it. Starting from the problem of the research, there is no clear perception of the formal characte