This paper addresses the problem of car detection from aerial images using Convolutional Neural Networks (CNNs). This problem presents additional challenges as compared to car (or any object) detection from ground images because the features of vehicles from aerial images are more difficult to discern. To investigate this issue, we assess the performance of three state-of-the-art CNN algorithms, namely Faster R-CNN, which is the most popular region-based algorithm, as well as YOLOv3 and YOLOv4, which are known to be the fastest detection algorithms. We analyze two datasets with different characteristics to check the impact of various factors, such as the UAV’s (unmanned aerial vehicle) altitude, camera resolution, and object size. A total of 52 training experiments were conducted to account for the effect of different hyperparameter values. The objective of this work is to conduct the most robust and exhaustive comparison between these three cutting-edge algorithms on the specific domain of aerial images. By using a variety of metrics, we show that the difference between YOLOv4 and YOLOv3 on the two datasets is statistically insignificant in terms of Average Precision (AP) (contrary to what was obtained on the COCO dataset). However, both of them yield markedly better performance than Faster R-CNN in most configurations. The only exception is that both of them exhibit a lower recall when object sizes and scales in the testing dataset differ largely from those in the training dataset.
Smart agriculture is an evolving trend in the agriculture industry, where sensors are embedded into plants to collect vital data and help in decision-making to ensure a higher quality of crops and prevent pests, disease, and other possible threats. One of the most critical pests of palms is the red palm weevil, which is an insect that causes much damage to palm trees and can devastate vast areas of palm trees. The most challenging problem is that the effect of the weevil is not visible by humans until the palm reaches an advanced infestation state. For this reason, there is a pressing need to use advanced technology for early detection and prevention of infestation propagation. In this project, we have developed an IoT-based smart palm monitoring prototype as a proof-of-concept that (1) allows monitoring palms remotely using smart agriculture sensors, (2) contribute to the early detection of red palm weevil infestation. Users can use web/mobile applications to interact with their palm farms and help them in getting early detection of possible infestations. We used an industrial-level IoT platform to interface between the sensor layer and the user layer. Moreover, we have collected data using accelerometer sensors, and we applied signal processing and statistical techniques to analyze collected data and determine a fingerprint of the infestation.
The Saudi Arabia government has proposed different frameworks such as the CITC’s Cybersecurity Regulatory Framework (CRF) and the NCA’s Essential Cybersecurity Controls (ECC) to ensure data and infrastructure security in all IT-based systems. However, these frameworks lack a practical, published mechanism that continuously assesses the organizations’ security level, especially in HEI (Higher Education Institutions) systems. This paper proposes a Cybersecurity Maturity Assessment Framework (SCMAF) for HEIs in Saudi Arabia. SCMAF is a comprehensive, customized security maturity assessment framework for Saudi organizations aligned with local and international security standards. The framework can be used as a self-assessment method to establish the security level and highlight the weaknesses and mitigation plans that need to be implemented. SCMAF is a mapping and codification model for all regulations that the Saudi organizations must comply with. The framework uses different levels of maturity against which the security performance of each organization can be measured. SCMAF is implemented as a lightweight assessment tool that could be provided online through a web-based service or offline by downloading the tool to ensure the organizations’ data privacy. Organizations that apply this framework can assess the security level of their systems, conduct a gap analysis and create a mitigation plan. The assessment results are communicated to the organization using visual score charts per security requirement per level attached with an evaluation report.
Android ransomware is a threatening malware that is targeting individuals and enterprises. Many existing approaches suggested different ransomware detection solutions to protect users' devices and data. These solutions used mainly static-based or dynamic-based analysis systems. However, the current solutions have considered only the old versions of Android platforms. In this paper, an efficient machine learning-based ransomware detection approach is proposed. This approach has studied deeply the latest version of Android (Version 11, API Level 30) to include the updated list of features including permissions and API packages calls that might be utilized by ransomware attacks. A new dataset was created after parsing 1000 apps to extract these features. Afterwards, different machine learning techniques were executed to generate different predictive models for Andoird ransomware. Some predictive models reached 98.3% of detection accuracy even after reducing around 26% of the overall features set.
Jamming is a terrifying attack that could harm 802.11p-based vehicular communications by occupying the communication channels by overwhelming the network with jamming packets, especially for self-driving cars, as it is essential to send/receive messages without any interruptions to control the vehicles remotely. In wireless vehicular ad hoc networks (VANET), the attacker’s mission is more accessible due to the network’s open nature, way of communication, and lack of security measures. Most of the existing studies have focused on jamming detection approaches. However, few of them have addressed the jammer localization challenge. Moreover, even in these limited studies, the solutions’ assumptions, the proposed countermeasures, and their complexity were also missing. Therefore, this paper introduces a new approach to detecting, localizing, and avoiding jamming attacks in VANETs with high efficiency in terms of accuracy, implementation and complexity. The proposed approach uses the signal strength of the jammer for estimating only the distance between jammer and receiver, while then a less complex algorithm is proposed for localizing the jammer and then redirecting the vehicles away from the roads the attacker is using. This approach was simulated using real-life maps and specialized network environments. Additionally, the performance of the new approach was evaluated using different metrics. These evaluation metrics include (1) the estimated position of the jammer, (2) the handling of the jammer by announcing its location to normal vehicles (3) the avoidance of the jammed routes by increasing their weight, which forces the cars to reroute and evade the jamming area. The high localization accuracy, measured by the Euclidean distance, and the successful communication of the attacker’s position and its avoidance have highly increased the packet delivery ratio (PDR) and the signal-to-interference-plus-noise ratio (SINR). This was noticed significantly before and after avoiding the jamming area when for example, the PDR increased from 0% to 100% before and after bypassing the jammer’s routes.
Cybersecurity attacks are still causing significant threats to individuals and organizations, affecting almost all aspects of life. Therefore, many countries worldwide try to overcome this by introducing and applying cybersecurity regularity frameworks to maintain organizations’ information and digital resources. Saudi Arabia has taken practical steps in this direction by developing the essential cybersecurity control (ECC) as a national cybersecurity regulation reference. Generally, the compliance assessment processes of different international cybersecurity standards and controls (ISO2700x, PCI, and NIST) are generic for all organizations with different scopes, business functionality, and criticality level, where the overall compliance score is absent with no consideration of the security control risk. Therefore, to address all of these shortcomings, this research takes the ECC as a baseline to build a comprehensive and customized risk-based cybersecurity compliance assessment system (RC2AS). ECC has been chosen because it is well-defined and inspired by many international standards. Another motive for this choice is the limited related works that have deeply studied ECC. RC2AS is developed to be compatible with the current ECC tool. It offers an offline self-assessment tool that helps the organization expedite the assessment process, identify current weaknesses, and provide better planning to enhance its level based on its priorities. Additionally, RC2AS proposes four methods to calculate the overall compliance score with ECC. Several scenarios are conducted to assess these methods and compare their performance. The goal is to reflect the accurate compliance score of an organization while considering its domain, needs, resources, and risk level of its security controls. Finally, the outputs of the assessment process are displayed through rich dashboards that comprehensively present the organization’s cybersecurity maturity and suggest an improvement plan for its level of compliance.