Abstract—In the current educational system, student’s frequent attendance in class is very important for performance evaluation and the quality control. Most institutions still use insecure, time-consuming traditional methods like calling people by the name or having them sign documents. Face recognition technology based on high-definition monitor video and other information technologies is used to solve the problem of the recognizing face for taking attendance. A facial recognition attendance system uses facial technology to automatically record attendance while identifying and verifying a person based on their facial features. This research focusing on a face recognition-based attendance system with getting a less false- positive rate using a threshold to confidence i.e., Euclidean distance value while detecting unknown persons and save their images. Compare to other Euclidean distance-based algorithms like Eigenfaces and Fisher faces, Local Binary Pattern Histogram (LBPH) algorithm is better. We used Haar cascade for face detection because of their robustness and LBPH algorithm for face recognition. It is robust against monotonic grayscale transformations. Scenarios such as face recognition rate, false- positive rate for that and false-positive rate with and without using a threshold in detecting unknown persons are considered to evaluate our system. We got face recognition rate of students is 77% and its false-positive rate is 28%. This system is recognizing students even when students are wearing glasses or grown a beard. Face Recognition of unknown persons is nearly 60% for both with and without applying threshold value. Its false-positive rate is 14% and 30% with and without applying threshold respectively. Keywords—Face recognition, face detection, machine learning, Haar Cascade Algorithm, Local Binary Pattens Histograms (LBPH).

Breast cancer is one of the most prevalent forms of cancer affecting women worldwide. Early detection plays a critical role in improving patient outcomes and survival rates. In recent years, the integration of hybrid models in breast cancer detection has shown promising results by combining the strengths of different machine learning algorithms and techniques. This abstract presents a comprehensive review of the application of hybrid models for breast cancer detection, highlighting their potential advantages and challenges. The proposed hybrid models combine various techniques, such as feature selection, data preprocessing, and classification algorithms, to enhance the accuracy and efficiency of breast cancer detection systems. Feature selection methods, including genetic algorithms, particle swarm optimization, and principal component analysis, are utilized to identify the most informative features from mammographic images or clinical data. These selected features are then fed into classification algorithms, such as support vector machines, random forests, artificial neural networks, or deep learning models, for accurate diagnosis. Furthermore, the hybrid models often incorporate data preprocessing techniques to improve the quality and consistency of the input data. Preprocessing steps may involve image enhancement, normalization, noise removal, or data augmentation, depending on the specific requirements of the model. By optimizing the feature extraction process and improving data quality, hybrid models aim to achieve higher sensitivity, specificity, and overall performance in breast cancer detection. Several studies have reported the successful implementation of hybrid models in breast cancer detection. These models have demonstrated superior performance compared to individual algorithms or traditional approaches. The combination of different algorithms allows for better capturing of complex patterns and subtle variations in breast images or clinical data, leading to more accurate and reliable detection outcomes. Despite the promising results, there are challenges associated with the development and application of hybrid models for breast cancer detection. The selection andfine-tuning of the various components in the hybrid models require careful consideration, as the performance heavily depends on the integration of different algorithms and techniques. Moreover, the interpretability and explainability of hybrid models need to be addressed to gain trust from healthcare professionals and patients. The integration of hybrid models in breast cancer detection has shown great potential in improving accuracy and efficiency. These models leverage the strengths of multiple algorithms and techniques, allowing for enhanced feature selection, data preprocessing, and classification. While challenges remain, further research and development in this area hold promise for advancing breast cancer detection systems, ultimately leading to earlier diagnoses and improved patient outcomes. International Journal of Scientific Research in Engineering and Management (IJSREM) Volume: 07 Issue: 07 | July - 2023 SJIF Rating: 8.176 ISSN: 2582-3930 © 2023, IJSREM | www.ijsrem.com DOI: 10.55041/IJSREM24204 | Page 2 Introduction: Breast cancer is a significant health concern affecting millions of women worldwide. Early detection and accurate diagnosis play a crucial role in improving treatment outcomes and patient survival rates. Deep learning, a subfield of machine learning, has emerged as a powerful approach for breast cancer detection, leveraging its ability to automatically learn intricate patterns and features from complex data. In this article, we delve into the application of deep learning models for breast cancer detection, exploring various architectures,datasets, and challenges associated with their implementation.

-Recent advances in deep learning have led to significant improvements in single image super-resolution (SR) research. However, due to the amplification of noise during the upsampling steps, state-ofthe-art methods often fail at reconstructing high-resolution images from noisy versions of their low-resolution counterparts. However, this is especially important for images from unknown cameras with unseen types of image degradation. In this work, we propose to jointly perform denoising and super-resolution.To this end, we investigate two architectural designs: "in-network" combines both tasks at feature level, while "pre-network" first performs denoising and then super-resolution. Our experiments show that both variants have specific advantages: The in-network design obtains the strongest results when the type of image corruption is aligned in the training and testing dataset, for any choice of denoiser. The prenetwork design exhibits superior performance on unseen types of image corruption, which is a pathological failure case of existing super- resolution models. We hope that these findings help to enable super-resolution also in less constrained scenarios where source camera or imaging conditions are not well controlled. Key Words: Super-resolution, Denoising, Deep learning, Image enhancement

Face recognition software is being used more often in a wide range of applications, but worries about privacy and exploitation have sparked interest in creating methods for "disguised" face identification. The use of neural networks for disguised face identification has recently attracted the attention of academics, who trained their algorithms on pictures of people with masks, hats, and other facial coverings. Yet, there are a number of difficulties in creating efficient disguised facial recognition systems, such as a lack of training data and the requirement to take different lighting and position changes into consideration. The use of adversarial networks, infrared imaging, and deep learning techniques are among the latest advancements in the field of masked facial recognition using neural networks that are examined in this review. The study also covers prospective uses for concealed face recognition technology as well as the moral issues surrounding its implementation. The survey's findings show that neural networks may be used to mask face recognition in order to satisfy privacy concerns while still enabling accurate identification of people in a range of scenarios. Keywords — Disguised Facial Recognition, Neural Networks, Machine Learning, Deep Learning, Adversarial Networks.

In contemporary society, Stress has emerged as a pervasive issue affecting a majority of individuals. Despite their material prosperity, people often find themselves dissatisfied due to pressures of stress. Stress is a difficult phenomenon that can manifest in emotional, physical, and mental forms. The objective of this project is to be identifying indicators of stress in IT professionals over the application of sophisticated image processing and machine learning methods. This technology represents an improvement over previous approaches for stress detection that didn't account for the subjective experiences of employees or real-time detection. The current model incorporates both periodic and live detection of employee emotions. Automated stress detection serves to mitigate health risks and enhance the well- being of both the IT Company and its employees. By understanding the emotional states of IT professionals, businesses can provide appropriate guidance and achieve better outcomes. Key Words: Stress, Machine learning and IT professionals.