A forensic analysis method for estimating vehicle speed is introduced by comparing two video sources from the same device (e.g., smartphone, security camera) in the same field of view. An initial reference video, featuring a single vehicle with a known speed determined by GPS data, is required for the proposed algorithm. Using this reference, the speed of a vehicle in a second video-where the speed is unknown or under question-can be estimated. Speed calculation in the second video relies on a user-defined region of interest (ROI), which is specified by free-hand drawing two lines. The algorithm combines the YOLO method and some optimization techniques in order to reduce processing time-by up to 97%. This approach offers a significant improvement in forensic video analysis, providing a faster and more reliable method for estimating vehicle speeds over a defined distance, which is crucial for law enforcement and investigations.
It is known that experts widely use digital forensics tools, in order to investigate criminal cases. The main scope of this research is considered as the performance comparison of digital forensic tools in various Operating Systems, such as “Windows 10”, “Windows 11”, “Ubuntu”, and “Windowsfx”. This is done for optimization reasons and more specifically for the minimization of processing time in forensic data analysis. The research was divided into two main parts by using exactly the same hardware as well as the default Operating Systems' configuration. The first part presents the performance comparison of the above Operating Systems by using two benchmarking tools such as “Phoronix Test Suite” and “Indigobench”. The second part focuses on the performance measurements of the corresponding digital forensic tools “FTK Imager”, “Photorec”, “Quickhash” and “Peautils”. It should be noted that every tool is tested ten times in order to increase the accuracy of the averaged total execution time of every iteration. The results indicate that the “Windowsfx” Operating System has the best overall performance after exhausting research of testing all the aforementioned tools.
Digital forensic experts often encounter the challenge of managing a significant volume of data during their investigations, which can lead to time and efficiency constraints. To address this issue, a tool has been developed to analyze provided images and specifically target the identification of pornographic content, with a particular emphasis on detecting child pornography as a future work, which is a criminal offense in many countries. This tool incorporates various optimization techniques to achieve its objective. First of all, it examines each image to determine the presence of a face or a person in order to extract the corresponding color palette. Subsequently, it conducts an in-depth analysis of the entire image for detecting skin regions by using a precise chromatic spectrum of pixels and extracting the relevant color palette. Finally, a comparison is performed between the two extracted palettes to identify common and contiguous areas. The proposed tool allows the user to set a specific threshold, the system provides a prompt response indicating the presence or absence of pornography. This ensures an efficient workflow for digital forensic professionals who are focused on combating child pornography and contribute to the global official activities by judicial authorities against this heinous crime.
The determinacy question, concerning whether or not there is convergence to a unique equilibrium path, in an explicitly dynamic setting, has been addressed in closed economies but only scantily in an open one. We undertake this task, using a model that leads to a 4th -degree characteristic equation which cannot be handled analytically; therefore, we develop a particular algorithm to solve it. We show that the determinacy-consistent specification of Taylor’s rule depends, critically, on the greater or lower openness in the trade sector. Greater openness supports a novel specification of Taylor’s rule with stronger emphasis on output gaps and weaker emphasis on the price stability than in earlier studies. Under this rule, determinate equilibrium dynamics is shown to be possible. When trade openness decreases significantly, determinacy is still possible, but it is compatible with a far more conventional formulation Taylor’s rule where price stability has absolute priority.
VMI (Visual-Motor Integration) is a psychometric test that is being used by mental health specialists to determine more precise clinical examination through the ability in handwriting and copying accuracy. According to the literature, psychometric tests are categorized as objectively or subjectively scored. For that reason, the proposed work increases the accuracy of VMI results by eliminating possible errors of any subjective human evaluation. So, an optimization algorithm is developed for analyzing high-resolution digital images of VMI sheets. The algorithm simulates Braille method, which is used by visually impaired people to decrease processing data by utilizing three different fingerprint patterns and locate the required ROIs (Regions of Interest). Finally, a comparison procedure indicates possible differences between the pre-drawn shape and the hand-drawing replica for computing two score values with respect to size and morphology.
In this study, the main idea is the development of an autonomous mobile system for dictating text documents via image processing algorithm for blind people. The system is constituted by the Raspberry Pi 2B - the mobile processing unit- and a pair of specially designed glasses with an HD camera and Bluetooth headset. The blind user should hold the book open (two pages) with his hands stretched straight at the level of his eyes; then a calibration procedure takes place in order to capture the best image. Therefore, 1-D signal transformation of the above image is produced in order to filter every text line. Finally, every word of each text line is identified via an OCR (Optical Character Recognition) method and the user hears it via TTS (Text To Speech) procedure.
Third generation Night Vision devices produce images in very dark conditions amplifying thousands of times the ambient light. The goal is to develop an algorithm which automatically colorizes the produced green noisy images and offers better verification (through human eye) for target identification. Certainly, the aim of the proposed method is to improve the received signals of the optic nerve, so distinguishing shapes and colors will be more accurate. In the first phase of the overall process, a night vision image is converted to a grayscale digital image and a histogram analysis takes place. In the second phase, a grayscale image is automatically selected from a small database. The database contains grayscale images and the corresponding color images (the conversion procedure is the same with the night vision gray scaling procedure). The selection of the grayscale database image is based on the histogram comparison to the gray scaled night vision image. In the final phase, both images (night vision and color) are converted to the YCbCr color encoding and luminance comparison between the two images takes place for choosing the nearest correct color for every pixel in the night vision image, with respect to the source color image.
Here is a short explanation of an algorithm, which introduces a new approach for detecting and measuring elliptical regions from a binary image. These regions appear in the form of highly irregular white shapes in a black and white image. The algorithm computes three ellipses for every shape in the image. The inscribed ellipse Ei which is the largest ellipse contained in the shape, the circumscribed ellipse Ec which is the smallest ellipse that contains the shape and the approximation ellipse Ea which is the ellipse with area equal to the area of the shape. The goal is to fit the Ea ellipse to the elliptical region (region of interest-ROI) by processing the minimum possible number of pixels, with maximum efficiency. Of course, the algorithm is capable of dealing with the individual case of circular regions. Some computational results are presented on a set of benchmark images from actual data. Finally, a comparison with the NASA's Goddard IDL program library takes place in order to confirm the performance of the proposed algorithm.
We introduce an innovative approach for detecting the region of interest in palmprint identification, from a highly noisy image, using a combinatorial algorithm. The existing research faces some critical issues such as noise, shadows, illumination variance, scars, rings, hand disorientation, disability (missing fingers) and different age group samples. All above inconvenient points are overcome by the proposed technique, in the preprocessing phase of palmprint verification. This is done by filling the hand shape with non overlap disks and by locating the disk with maximum possible radius, which contains the palm with biometric features, such as principal lines and wrinkles. Additionally, a quick process computes the hand orientation. The results in a wide range of test bed images lead to the correct computed disk, even in the worst cases. Finally, a comparison with prior work takes place in order to confirm the performance of the proposed algorithm.
In this study a combinatorial algorithm is developed for the detection of regions that could be approximated by circular disks. The figure of a region is a section where the boundary is inordinately asymmetrical and jagged. Three circular disks are computed for each region. The indisk which is the largest disk contained in the region, the outdisk which is the smallest disk that contains the region and has the same center as the indisk, and the approximation disk with area equal to the area of the region and has the maximum intersection with the region. Reasonably there are some problems described as a considerable remark for efficient use of the proposed algorithm: in meteorology, where hail pads have to be analyzed in order to determine the number and characteristics of the hailstones that collided with the pad, in biology especially in cytology where cells have to be identified for the reason that contains useful information that should be extracted. Computational results on a set of benchmark images from actual data are presented.
In this paper a combinatorial algorithm is presented and implemented for the identification and measurement of apertures from a noisy image. These apertures appear in the form of highly irregular shapes, and are approximated by circular disks. Three circular disks are defined for each aperture. The indisk which the largest disk contained in the aperture, the outdisk which is the smallest disk that contains the aperture and with center that of the indisk, and the approximation disk with center that of the indisk and area equal to the area of the aperture. The problem of identifying jagged apertures from a noisy image that could be approximated by circular disks has many applications in diverse areas. In this paper we will present how this problem emerges in meteorology, where hail pads have to be analyzed in order to determine the number and characteristics of the hail stones that collided with the pad. Computational results on a set of benchmark images from actual data are presented
