Exposing Manipulated Photos and Videos in Digital Forensics Analysis

Tampered multimedia content is being increasingly used in a broad range of cybercrime activities. The spread of fake news, misinformation, digital kidnapping, and ransomware-related crimes are amongst the most recurrent crimes in which manipulated digital photos and videos are the perpetrating and disseminating medium. Criminal investigation has been challenged in applying machine learning techniques to automatically distinguish between fake and genuine seized photos and videos. Despite the pertinent need for manual validation, easy-to-use platforms for digital forensics are essential to automate and facilitate the detection of tampered content and to help criminal investigators with their work. This paper presents a machine learning Support Vector Machines (SVM) based method to distinguish between genuine and fake multimedia files, namely digital photos and videos, which may indicate the presence of deepfake content. The method was implemented in Python and integrated as new modules in the widely used digital forensics application Autopsy. The implemented approach extracts a set of simple features resulting from the application of a Discrete Fourier Transform (DFT) to digital photos and video frames. The model was evaluated with a large dataset of classified multimedia files containing both legitimate and fake photos and frames extracted from videos. Regarding deepfake detection in videos, the Celeb-DFv1 dataset was used, featuring 590 original videos collected from YouTube, and covering different subjects. The results obtained with the 5-fold cross-validation outperformed those SVM-based methods documented in the literature, by achieving an average F1-score of 99.53%, 79.55%, and 89.10%, respectively for photos, videos, and a mixture of both types of content. A benchmark with state-of-the-art methods was also done, by comparing the proposed SVM method with deep learning approaches, namely Convolutional Neural Networks (CNN). Despite CNN having outperformed the proposed DFT-SVM compound method, the competitiveness of the results attained by DFT-SVM and the substantially reduced processing time make it appropriate to be implemented and embedded into Autopsy modules, by predicting the level of fakeness calculated for each analyzed multimedia file.

Micro Electrochemical Milling of Micro Metal Parts with Rotating Ultrasonic Electrode

With the rapid development of MEMS, the demand for metal microstructure is increasing. Micro electrochemical milling technology (MECM) is capable of manufacturing micro metallic devices or components based on the principle of electrochemical anode dissolution. To improve the capacity of MECM, this paper presents a compound method named ultrasonic vibration-assisted micro electrochemical milling technology (UA-MECM). Firstly, the simulation and mathematical model of UA-MECM process is established to explain the mechanism of ultrasonic vibration on micro electrochemical milling. Then, the effects of ultrasonic parameters, electrical parameters and feedrate on machining localization and surface quality are discussed considering sets of experiments. The surface roughness was effectively reduced from Ra 0.83 to Ra 0.26 µm with the addition of ultrasonic vibration. It turns out that ultrasonic vibration can obviously improve machining precision, efficiency and quality. Finally, two- and three-dimensional microstructures with good surface quality were successful fabricated. It shows that ultrasonic vibration-assisted electrochemical milling technology has excellent machining performance, which has potential and broad industrial application prospects.

Ultra- structure of Toombak; smokeless tobacco of Sudan and its effects on oral and systemic health

Introduction: Toombak is a smokeless tobacco used by the Sudanese. Nicotiana Rustica leaves are fermented, and sodium bicarbonate is added, increasing Ph. This causes high absorption spikes of unprotonated nicotine through mucosal epithelium, and into the blood, distinguishing Toombak as an addictive product. Furthermore,Toombak hashighmicrobial contamination due to its open field, rural production. Materials and methods: 21 tobacco samples were collected from 3 main towns in the capital Khartoum, Sudan. Analysis included microbiome (16S rRNA sequencing), metabolome (Liquid and Gas Chromotography- volatile organic compound method), heavy metal content and scanning electron microscopy (SEM). Results: Compared to limit of detection values (LOD), Tobacco specific nitrosamines (TSNA’s), in particular, 4(methylnitrosamino)-1-(3-pyridyl)–butanone (NNK) were markedly increased [NNK; 1.2-4.7 mg/g; LOD 0.02]. Free nicotine ranged from 16-31 mg/g; [LOD 0.01]. High choline and carnitine, volatile aldehydes; and benzyl alcohol were also detected. 8 cations, of which iron [1465 mg/kg] and copper [3.76 mg/kg] were strikingly observed, compared to average levels from other products; [Iron 20-60mg/kg, Copper 1.3mg/kg]. 8 Phyla, including but not limited to; Actinobacteria; cornyebacterium(contain nitrate reductase genes) and Firmicutes; staphylococcus and facklamia, were sequenced. SEM highlighted a non-homogenous product with elevated sodium spectrum. Conclusions: TSNA’s were potent in Toombak due to high starting nicotine and rich nitrate reductase bacteria. High choline and carnitine can promote cardiovascular disease through their conversion to trimethylamine-N-oxide. Diphtheria and infective endocarditis are associated with cornyebacterium and facklamia respectively while staphylococcus can lead to numerous systemic opportunistic diseases.

An Extension-Based Classification System of Cloud Computing Patents

Owing to the large number of professional glossaries and unknown patent classification, analysts usually fail to collect and analyze patents efficiently. One solution to this problem is to conduct patent analysis using a patent classification system. However, in a corpus such as cloud patents, many keywords are common among different classes, making it difficult to classify the unknown class documents using the machine learning techniques proposed by previous studies. To remedy this problem, this study aims to establish an efficient classification system with a special focus on features extraction and application of extension theory. We first propose a compound method to determine the features, and then, we propose an extension-based classification method to develop an efficient patent classification system. Using cloud computing patents as the database, the experimental results show that our proposed scheme can outperform the classification quality of the traditional classifiers.

LC–MS-MS Method for the Determination of Antidepressants and Benzodiazepines in Meconium

An LC–MS-MS method for the determination of 14 benzodiazepines (BZDs) (alprazolam, α-hydroxyalprazolam, clonazepam, bromazepam, diazepam, nordiazepam, lorazepam, lormetazepam, oxazepam, flunitrazepam, 7-aminoflunitrazepam, triazolam, midazolam and zolpidem) and 15 antidepressants (ADs) (amitriptyline, nortriptyline, imipramine, desipramine, clomipramine, norclomipramine, fluoxetine, norfluoxetine, sertraline, norsertraline, paroxetine, venlafaxine, desmethylvenlafaxine, citalopram and desmethylcitalopram) in meconium was developed and validated. Meconium samples (0.25 ± 0.02 g) were homogenized in methanol and subjected to mixed-mode cation exchange solid-phase extraction. Chromatographic separation was performed in reversed phase, with a gradient of 0.1% formic acid in 2 mM ammonium formate and acetonitrile. Two different chromatographic gradient methods were employed, one for the separation of ADs and another for BZDs. Analytes were monitored by tandem mass spectrometry employing electrospray positive mode in MRM mode (2 transitions per compound). Method validation included: linearity [n = 5, limit of quantification (LOQ) to 400 ng/g], limits of detection (n = 6, 1–20 ng/g), LOQ (n = 9, 5–20 ng/g), selectivity (no endogenous or exogenous interferences), accuracy (n = 15, 90.6–111.5%), imprecision (n = 15, 0–14.6%), matrix effect (n = 10, −73 to 194.9%), extraction efficiency (n = 6, 35.9–91.2%), process efficiency (n = 6, 20.1–188.2%), stability 72 h in the autosampler (n = 3, −8.5 to 9%) and freeze/thaw stability (n = 3, −1.2 to −47%). The method was applied to four meconium specimens, which were analyzed with and without hydrolysis (enzymatic and alkaline). The authentic meconium samples tested positive for alprazolam, α-hydroxyalprazolam, clonazepam, diazepam, nordiazepam, fluoxetine, norfluoxetine, clomipramine and norclomipramine. Therefore, the present LC–MS-MS method allows a high throughput determination of the most common BZDs and ADs in meconium, which could be useful in clinical and forensic settings.