Implementing a fractional-order operator requires many resources to acquire an accurate response compared to the theoretical response. In this paper, three implementation methods of digital fractional-order operators are exploited. The three implementation methods are based on FIR, IIR, and lattice wave digital filters. The three methods are implemented using different optimization algorithms to optimize the choice of the coefficients of the three filters. This optimization is done to approximate the frequency response of an ideal fractional operator. This comparison aims to determine each implementation method’s accuracy and resource usage level to decide which method is better for different systems. © 2021 IEEE.
Vulnerable Road Users Detection and Tracking using YOLOv4 and Deep SORT
Over the years, The detection and tracking of Vulnerable Road Users (VRUs) have become one of the most critical features of self-driving car components. Because of its processing efficiency and better detection algorithms, tracking-by-detection appears to be the best paradigm. In this paper, a detection-based tracking approach is presented for Multiple VRU Tracking of video from an inside-vehicle camera in real-time. YOLOv4 scans every frame to detect VRUs first, then Simple Online and Realtime Tracking with a Deep Association Metric (Deep SORT) algorithm, which is customized for multiple VRU tracking, is applied. The results of our experiments on both the Joint Attention in Autonomous Driving (JAAD) and Multiple Object Tracking (MOT) datasets exhibit competitive performance. © 2021 IEEE.
COVID-19 Diagnosis from CT-images Using Transfer Learning
In symptomatic patients, a positive COVID-19 test is critical for securing life-saving services such as ICU care and ventilator support; it may cause septic shock, septic pneumonia, respiratory failure, heart difficulties, liver issues, and even death. CAD systems help people in rural places and doctors in the early detection of COVID-19. A diagnostic and severity detection technique utilizing transfer learning and a backpropagation neural network has been developed with the aid of a computer for this purpose. This study aims to compare and analyze multiple deep learning-enhanced strategies for detecting COVID-19 in CT scan medical images. The COVID-19 CT scan binary classification challenge utilized two powerful pretrained CNN models: Inception ResNet V2 and ResNet50. To achieve higher accuracy in the diagnosis of COVID-19 using CT scan images, a new approach called Inception ResNet was employed, and it resulted in 97.3% accuracy and 97.38% specificity. Transfer learning techniques were employed to reduce the training time and get around the shortage of data. The proposed approaches outperformed more than other papers in the literature by 0.2%. © 2023 IEEE.
A Comparative Study of Different Chaotic Systems in Path Planning for Surveillance Applications
This paper compares the performance of four different chaotic systems in path planning for surveillance applications. The four investigated systems are Lorenz, Arneodo, Liu, and Chen. While the Lorenz system was employed in a similar application before, Arneodo, Liu, and Chen systems are newly introduced in this paper. A bounded-grid chaotic path planner is proposed based on the mirror mapping technique, which keeps the robot bounded in the terrain and prevents it from going outside. The effect of using different state variables of each chaotic system to control the motion angle of the robot is discussed and shown to have a significant impact on the robot’s performance. The obtained trajectory and several performance metrics show promising results of the chaotic path planner for the four systems. © 2021 IEEE.
Chaos-Based Image Encryption Using DNA Manipulation and a Modified Arnold Transform
Digital images, which we store and communicate everyday, may contain confidential information that must not be exposed to others. Numerous researches are interested in encryption, which protects the images from ending up in the hands of unauthorized third parties. This paper proposes an image encryption scheme using chaotic systems, DNA manipulation, and a modified Arnold transform. Both DNA manipulation and hyperchaotic Lorenz system are utilized in the substitution of the images’ pixel values. An additional role of hyperchaotic Lorenz system is that it generates the random numbers required within the DNA manipulation steps. DNA cycling is implemented based on simple DNA coding rules and DNA addition and subtraction rules with modulus operation. The modified Arnold transform alters the pixels’ positions, where it guarantees effective pixel permutation that never outputs the same input pixels arrangement again. The proposed design is simple and amenable for hardware realization. Several well established performance evaluation tests including statistical properties of the encrypted image, key space, and differential attack analysis were conducted for several images. The proposed scheme passed the tests and demonstrated good results compared to several recent chaos-based image encryption schemes. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.
A Unified System for Encryption and Multi-Secret Image Sharing Using S-box and CRT
Multi-Secret Image Sharing (MSIS) is used when multiple images need to be shared to multiple participants, but the images can not be recovered without the presence of all shares. In this paper, a unified system for performing encryption and (n,n)-MSIS is proposed. While MSIS is based on the XOR operation, encryption combines the utilization of Chinese Remainder Theorem (CRT), SHA-256, and S-box for improved security. The same designed system is used for the generation of secret shares and the recovery of secret images. In addition, a sensitive system key is designed where three pairwise relatively prime subkeys are automatically generated for utilization in the CRT. The resulting secret shares pass statistical evaluation criteria such as RMSE, correlation, and entropy, and give good results for differential attack measures, and runtime. In addition, the proposed system succeeds in passing the NIST SP-800-22 statistical test suite and key sensitivity measures. © 2022 IEEE.
PRNG Using Primitive Roots of Primes and its Utilization in Chess-based Image Encryption
Recently, number theory has proved its importance in cryptography because of its well-known hard problems. For instance, a primitive root for a prime number shows a special property of uniqueness when raised to different powers mod the prime number. In this paper, a Pseudorandom Number Generator (PRNG) is designed based on this property using a prime number and some of its primitive roots. The PRNG is, first, validated for utilization in cryptography applications using histograms, correlation coefficients, and the National Institute of Standards and Technology (NIST) statistical test suite. Then, the PRNG is utilized in an image encryption system and the system security is tested using statistical measures, differential attack measures, and sensitivity to one-bit change. The results are promising and in the expected good ranges. © 2022 IEEE.
Generic Hardware Realization of K Nearest Neighbors on FPGA
K Nearest Neighbors (KNN) algorithm is a straight-forward yet powerful Machine Learning (ML) tool widely used in classification, clustering, and regression applications. In this work, KNN is applied, with three distance metrics, to classify different datasets, experimentally testing each distance metric effect on the classification performance. A static K is applied for the whole dataset optimally chosen based on a 5-fold cross-validation. A reconfigurable hardware realization on field programmable gate array (FPGA) of each distance metric applying selection sort algorithm is proposed. The FPGA realization reaches a throughput up to 4.44 Gbit/sec while only occupying 1% of the Genesys 2 Kintex-7 board area. The algorithm managed to classify all the tested datasets with above 90% accuracy. © 2022 IEEE.
Registerless Multiplierless YCoCg-R and YCoCg Color Space Converters Hardware Implementation
Multimedia data, e.g., images and videos, are widely used over the internet and on computers. Image processing applications require color space conversion to be able to deal with these types of data more efficiently. This paper investigates three color space conversions and proposes simplified combinational hardware designs and FPGA realizations for RGB to YCoCg-R and YCoCg color spaces encoders and decoders and compares them to their sequential counterparts. The proposed hardware design for the encoders and decoders uses only adders and subtractors without any registers or multipliers. The proposed YCoCg-R converter exhibits better resources utilization compared to implementing the design using shift registers, where it uses 56.3% and 72.1% less LUTs and FFs, respectively. Similarly for the YCoCg color space, the combinational design used 48.1% less LUTs and 67.8% less FFs than its sequential counterpart. © 2022 IEEE.
Generalized ?+?-order Filter Based on Single CCII
Different generalized filters topologies are proposed in the fractional-order domain. Three voltage-mode topologies and one current-mode topology are used to realize several types of fractional-order filters by applying different admittances combinations. The proposed topologies are designed using a single second-generation current conveyor (CCII-) and two fractional-order capacitors, which add more degrees of freedom for the design. The generalized Fractional Transfer Function (FTF) for each proposed topology is investigated where the fractional-order low-pass, band-pass, high-pass, and notch filters with ?+?)-order are realized. The Numerical results are provided where the stability analysis is presented for different cases. Also, the PSPICE simulations are presented to prove the theoretical findings of selected cases. © 2020 IEEE.