Multi-Lingual Character Recognition and Extraction using Recurrent Neural Networks
A Novel Approach to Smart Autonomous Monitoring of Indoor Plant Health Based on Leaf Color
Edge Preserving Image Denoising with Deep Networks and Ensemble Convolutional Neural Network Architectures
Neural Network Approach to Detect Renal Calculi
Enhancing Fabric Quality Control: Implementing Real-Time Defect Detection with Image Processing Techniques and Arduino
Identification of Volcano Hotspots by using Resilient Back Propagation (RBP) Algorithm Via Satellite Images
Data Hiding in Encrypted Compressed Videos for Privacy Information Protection
Improved Video Watermarking using Discrete Cosine Transform
Contrast Enhancement based Brain Tumour MRI Image Segmentation and Detection with Low Power Consumption
Denoising of Images by Wavelets and Contourlets using Bi-Shrink Filter
The COVID-19 pandemic has compelled radical and innovative reforms and Education and academia have been identified as sectors most adversely affected by the pandemic. Disrupting the age-old classroom setup, the pandemic has forced educational institutions such as schools and universities to implement 'online classes.' However, the evaluation aspect of education remains to be desired. Many automatic online exam proctoring systems have been proposed for online examinations during this COVID-19 pandemic, but they have certain limitations, including fewer and inaccurate functionalities. In this paper, a smart exam monitoring system is presented that addresses many of the problems with past systems, aiming to help institutions prevent malpractices during exams. This smart exam monitoring system leverages advanced AI algorithms to monitor online exams in a more precise and comprehensive manner. It can detect various forms of cheating, such as screen sharing and unauthorized resource access, while also ensuring a fair evaluation process. By integrating cutting-edge technology into the education sector, the aim is to uphold the integrity of online examinations and adapt to the challenges posed by the ongoing COVID-19 pandemic.
Color images can be encrypted using a hyperchaotic system in the Discrete Cosine Transform (DCT) and gyrator transform domains. These encryption techniques provide an added layer of security to protect sensitive image data from unauthorized access or tampering. The Red, Green, And Blue (RGB) components of the original color image are encoded into two-dimensional streams, respectively. These two-dimensional streams, representing the RGB components, are then processed by the display or printing device to recreate the full-color image. This paper proposes a new image encryption scheme that integrates DCT and the gyrator transform simultaneously. This innovative approach enhances the security and robustness of image encryption, making it suitable for a wide range of applications in data protection and transmission. Specifically, four chaotic sequences generated by a hyperchaotic system are blended into streams to synthesize complex sequences. Subsequently, the two-dimensional complex streams are scrambled using the discrete cosine transform and encoded back into an image format. Finally, the parameters in the hyperchaotic system using DCT and the gyrator transform improve the security of the proposed scheme. Theoretical and experimental analyses both confirm the security and validity of the proposed work.
Driver drowsiness is a critical factor contributing to road accidents worldwide, with potentially devastating consequences. To mitigate this problem, numerous driver drowsiness detection systems have been developed, employing various computerized devices and technologies. The development of technologies for detecting drowsiness is a major challenge in the field of accident-avoidance systems. Since the dangers of drowsiness exist on the road, techniques need to be advanced to prevent its consequences. The most crucial aspect of this work is to develop a drowsiness detection system by tracking the driver's eyes. It is believed that signs of driver exhaustion can be detected early enough to prevent a car accident. This technology aims to enhance road safety by alerting drivers or triggering safety measures when drowsiness is detected, ultimately saving lives and preventing accidents. Detecting fatigue involves observing eye movements and blink patterns. This review aims to provide an in-depth analysis of the effectiveness of these systems in identifying and alerting drowsy drivers, ultimately enhancing road safety. The analysis of facial images is a popular research area with applications such as face recognition and human identification security systems. Furthermore, this review delves into the various methodologies and algorithms used in driver drowsiness detection, including machine learning, image analysis, and signal processing techniques. It also highlights the need for further research to address the challenges and limitations faced by these systems and suggests potential avenues for improvement.
Face identification becomes an extremely challenging task, while wearing masks since some aspects of the face are obscured, making it crucial to develop advanced facial recognition technologies that can handle these complex scenarios effectively. The purpose of this paper is to offer an overview of several different approaches and algorithms that are utilized for human recognition when a face mask is applied. In this study, a number of distinct methodologies, including the Haar cascade, Adaboost, VGG-16 CNN Model, and others, are discussed. These approaches are compared side by side in order to determine which one has the best chance of success. It is possible that in the not too distant future, with the continued development of technology and the passage of time, more trustworthy methods for human recognition while wearing a face mask will come into use. It describes some of the practical uses of face detection and highlights the potential for advancements in reliable methods for human recognition even when individuals are wearing face masks. This technology has a variety of uses in public areas, schools, and other settings where persons need to be recognized despite wearing face masks in order to assist society. Facial recognition technology can also enhance security and streamline access control processes in a wide range of applications, from airports to corporate offices.
With the development of multimedia technology and the popularity of smartphones, the collection of multimedia information from various applications has rapidly grown. This information encompasses text, images, video, and audio compositions. This multimedia information has not only transformed the way we communicate and share content but has also posed new challenges and opportunities in fields such as data analysis, content creation, and information management. Consequently, multimedia education has become the focal point for research institutions and businesses. The modern solution faces two main challenges: expensive uptime and comparative speed. Therefore, this study first introduces a washing method called Permutation (OPH) to reduce expensive working time. Furthermore, the study explores innovative techniques to enhance the comparative speed of the modern solution, ultimately aiming to achieve a more efficient and cost-effective system. With the OPH, a permutation wash (GOPH) has been developed, along with several useful techniques to address pressure changes. This development is based on the fact that most multimedia files differ from one another. A new hashing method, known as Hierarchical Permutation Hash (HOPH), has been created to enhance productivity. Extensive testing with real multimedia files clearly demonstrates that HOPH is five to seven times faster than Minis' while maintaining the same level of accuracy. HOPH achieves this remarkable speed improvement by efficiently leveraging hierarchical structures in data, making it a promising solution for a wide range of multimedia applications.
