فراس منعم بجاي الابراهيمي

Safety in chemical and biological laboratories requires face masks. Nevertheless, the masked face images present obstacles to accommodate the wide pose variations, low resolution, changes in size and incomplete cropping, and varying emotive expressions, making feature extraction and recognition robust dicult. To decern between faces with proper, improper, and no facemask, Deep Learning (the use of Convolutional Neural Networks (CNNs)) and classiers such as SoftMax, Support Vector Machines (SVM) and K-Nearest Neighbors (KNN) were used. The methodology utilized the 2075 AR face images that contain various resolutions, orientations, and occlusions. The paper develops a Deep Learning-based model utilizing CNNs and classiers to detect proper, improper and no facemask usage. The results were promising, especially with occlusions in the moderated level: SoftMax with an accuracy of 88.41%, SVM with 80.19% and KNN of 47.58%. This methodology will assist in addressing challenges in detecting masked faces especially the wrong mask, which could help monitor compliance and ensure safety in other environments as well. The signicance of the problem addressed is of utmost importance as it helps develop and ensure accurate detections of proper usage of facemasks. This issue is critical for safety in the chemical and biological laboratories and has implications for compliant enforcement, for safety, and for other industries especially those reliant on facial recognition technologies.

Iraq’s primary crop, crucial for both domestic consumption and exports, is rice. The prevalence of rice infections poses a significant challenge to farmers, impacting crop yield and resulting in substantial losses. Human identification of diseases relies on expertise, making early diagnosis crucial for sustaining rice plant health. To address the limited number of rice leaf images in the database, our approach incorporates augmentation and dilation rate. Integrating drone technology and machine learning algorithms offers a promising solution to efficiently diagnose rice leaf diseases. However, existing methods face challenges such as picture backgrounds, insufficient field image data, and symptom variations. This work introduces a robust methodology, leveraging a specialized Convolutional Neural Network (CNN) model for rice leaf photos, effectively enhancing disease classification accuracy. The proposed approach successfully identifies and diagnoses three distinct classes: leaf smut, brown spot, and bacterial leaf blight.

Facial recognition is a popular biometric identification method that analyzes face images and extracts information that may be used to identify people. It is frequently used in a variety of fields, such as surveillance, access control, attendance control, and facial recognition for potential offenders. However, it's really difficult to create an automated system that recognize faces similarly to how a human can, especially when dealing with complicated databases that present difficulties like varying posture angles reach to 180 degrees. Additionally, it can be difficult to maintain the image's high resolution without drastically reducing its size. Working on images without cropping out the face presents another difficulty because of the posing angle. In addition, the depth of a face's features can have a big impact on how accurately it is recognized. In this paper, deep learning is proposed using a convolutional neural network as a solution to address all the challenges mentioned, which is able to extract deep features and correctly recognize those features. The FEI face database was used to implement all experiments, and the acquired experimental results were presented and assessed. The testing outcomes on the FEI face database proved the efficiency of the proposed methodology, and we were able to achieve a best recognition accuracy of 98.67% and F1_score of 98.53%. The suggested strategy, which is based on CNN, performs better than current approaches.

A palmprint is a small part of the palm flat that contains additional characteristics that can be used in authentication systems. It also has the property of permanence, which indicates that it will not alter through time. However, extracting the deepest and useful features from palmprint is a critical point. Most of the recently developed methods use principal lines, wrinkles, and creases, which is not enough to distinguish two people due to closeness. Recently, deep learning methods have been considered as an important key point for these kinds of tasks in order to extract deep features like texture features. We present a deep convolutional neural networks (CNN) that is specifically designed to suit palmprint images in order to achieve secure authentication processes. The COEP palmprint database was used in the experiments, and the accuracy measure as well as the F1-score were used in the evaluation process. The proposed model had a high level of accuracy, with a score of 97.55 percent. Palmprint authentication is performed efficiently using the described method.

Palmprint authentication has received a lot of attention as one of the most prevalent biometric methods. A palmprint is a portion of the palm’s surface that has special characteristics that could be used for authentication. Getting the most valuable features out of a palmprint, on the other hand, is a major challenge. Another challenge is coming up with an efficient strategy to authentication that uses fewer images, especially with approaches that require a high number of images in the training phase, which is a major issue. The majority of recently developed approaches rely on primary lines, wrinkles, and creases, which in some situations are insufficient to separate two people due to resemblance. Deep learning methods have recently been viewed as a critical component for extracting deep features such as texture features in these types of tasks. We concentrated on the palmprint authentication challenge in this work by creating a hybrid model called AE+CNN, which is based on the autoencoder (AE) model and a convolutional neural network (CNN) model. The proposed model comprises three phases: pre-processing, region of interest extraction (ROI), and feature extraction and matching using hybrid AE+CNN. The experiments used the COEP palmprint database, which has a limited number of palmprint images, posing a significant challenge for deep learning models that require a large number of images for training. The F1-score and the accuracy metric were both employed in the evaluation. with a score of 97.85 % for accuracy and 96.81 % for F1-score. Gradientweighted class activation mapping (Grad-CAM) was also applied to figure out which parts of the palmprint are the most discriminative for class classification.

Face recognition systems are considered to be the most user-friendly systems in the biometric world. Also, they are one of the active research fields in which the researchers work to improve the efficiency and efficacy of biometric applications. The major challenges reside on these applications are to extract unique and differentiable features that are not disturbed by pose, illuminations and other environmental and physical variations. Active Appearance Model (AAM) is one of the population solutions that extract features by precise modeling of human faces under various physical and environmental circumstances. In this paper, an Enhanced AAM based 1 to N face recognition system is proposed. The experimental results show the performance of the proposed face recognition system