FIGURE 6 - uploaded by Master Prince
Content may be subject to copyright.
Source publication
Human face and facial features gain a lot of attention from researchers and are considered as one of the most popular topics recently. Features and information extracted from a person are known as soft biometric, they have been used to improve the recognition performance and enhance the search engine for face images, which can be further applied in...
Contexts in source publication
Context 1
... types of shortcuts are used in ResNet-50 layers; Identity shortcuts are used when input/ output have the same dimensions, while projection shortcuts are used to match dimensions [42]. Figure 6 shows more details about ResNet-50 layers architecture. Downsampling is performed between blocks with a stride of 2 [42]. ...
Context 2
... Optimizer: SGD and Adam • Freezing layers (blocks): there are 4 blocks in ResNet-50 explained in Figure 6. Switch freezing between blocks in different experiments and sometimes made all the network trainable. ...
Context 3
... types of shortcuts are used in ResNet-50 layers; Identity shortcuts are used when input/ output have the same dimensions, while projection shortcuts are used to match dimensions [42]. Figure 6 shows more details about ResNet-50 layers architecture. Downsampling is performed between blocks with a stride of 2 [42]. ...
Similar publications
Background
As Arab countries seek to implement the 'Guideline on Good Pharmacovigilance Practice (GVP) for Arab countries', understanding policy implementation mechanisms and the factors impacting it can inform best implementation practice. This study aimed to explore the mechanisms of and factors influencing pharmacovigilance policy implementation...
Citations
... ResNet50 Model Architecture[6] ...
In nations such as Bangladesh, agriculture plays a vital role in providing livelihoods for a significant portion of the population. Identifying and classifying plant diseases early is critical to prevent their spread and minimize their impact on crop yield and quality. Various computer vision techniques can be used for such detection and classification. While CNNs have been dominant on such image classification tasks, vision transformers has become equally good in recent time also. In this paper we study the various computer vision techniques for Bangladeshi rice leaf disease detection. We use the Dhan-Shomadhan -- a Bangladeshi rice leaf disease dataset, to experiment with various CNN and ViT models. We also compared the performance of such deep neural network architecture with traditional machine learning architecture like Support Vector Machine(SVM). We leveraged transfer learning for better generalization with lower amount of training data. Among the models tested, ResNet50 exhibited the best performance over other CNN and transformer-based models making it the optimal choice for this task.
... Graph Convolutional Networks are effectively used to learn discriminant spatial features from the input face images and is proven to outperform for occluded face detection [7]. The authors in [8] use a deep convolutional neural network (DCNN) with a SoftMax classifier to perform the classification task on Arab ethnicities. A coupled attribute learning framework that leverages the relationships between facial attributes and face recognition is proposed in [9]. ...
Security concerns are rampant in our world today. No location is immune to crime, whether it be public spaces, commercial areas, offices, or private property. While law enforcement agencies work to address crimes after they have taken place, there is no guarantee that they can prevent them from occurring. Communities and businesses often turn to security personnel and surveillance cameras to keep watch, but this only provides passive protection. Criminals may act while security is absent and cameras only record events, they do not prevent them. Although, there are few systems with face recognition and alerting mechanisms are existing, they are too expensive and not affordable to a common man. In response to this challenge, this work proposes a most economic, hassle free proactive home security system which can be used as a plug and play device for the already installed surveillance system employing CCTV cameras. This system features facial recognition technology, enabling it to identify authorized individuals by storing their images in a database, thus swiftly distinguishing between intruders and residents. In cases of unauthorized intrusions, the system promptly activates alerts and alarms. Additionally, it incorporates supplementary security measures utilizing sensors and actuators for heightened protection. Moreover, the system utilizes an IP camera, permitting remote monitoring via live streaming. An array of machine learning algorithms is assessed to ascertain the most effective method for facial recognition, integrated into a monitoring dashboard using RTSP streaming.
... Some have blue eyes and red hair, while others are dark-skinned (Awad et al., 2021;Wang, 2022). This diversity reflects the complex history of the Arab world, where various ethnic groups, including Persians, Turks, and Africans, have migrated and intermixed with the Arab population over time (Al-Humaidan & Prince, 2021;Eid, 2007). Despite this diversity, traditional Arabic and Islamic references tend to consider Arab as a race, particularly for those who inhabit the Arabian Peninsula (Ibn Khaldun, 1958;Tabari, 1987). ...
This research explored the impact of religious, cultural, and traditional beliefs on Arab Muslims' understanding of mental disorders and their treatment, with a particular emphasis on the role of Islamic theology. Employing an exploratory sequential mixed-methods design, the study first conducted qualitative semi-structured interviews with 12 Arab Muslim participants (6 men and 6 women) to examine their mental health perspectives. Thematic analysis of these interviews informed the development of a quantitative survey, which was administered to 169 Arab Muslim participants using Qualtrics. The quantitative data were analysed using SPSS 29. The integration of qualitative and quantitative findings revealed that
Arab Muslim participants exhibited moderate to high levels of religiosity, which, along with their cultural and traditional beliefs influenced their mental health perceptions and treatment approaches. Notably, a discrepancy between participants' self-identification as religious and their actual religious practices suggests a cultural value placed on modesty. The preference
for traditional healing practices and supernatural explanations for mental disorders indicates a strategic approach to navigating mental health stigma. Furthermore, education was identified as a crucial element in dispelling mental health misconceptions, with higher levels of education associated with a more accurate understanding of mental disorders and an increased likelihood of utilising formal mental health services. These insights highlight the challenges of integrating cultural, religious, and educational factors in shaping mental health perceptions and underscore the need for culturally and religiously sensitive mental health interventions and education. This study advocates for bridging the gap between traditional beliefs and formal mental health services to improve access and attitudes towards mental health care among Arab Muslims in Australia.
... Race classification has become significant in applications like surveillance [37] and market advertising [38]. Recently, various deep learning models have been used for race identification [37,39,40,41,42]. ...
Technologies for recognizing facial attributes like race, gender, age, and emotion have several applications, such as surveillance, advertising content, sentiment analysis, and the study of demographic trends and social behaviors. Analyzing demographic characteristics based on images and analyzing facial expressions have several challenges due to the complexity of humans' facial attributes. Traditional approaches have employed CNNs and various other deep learning techniques, trained on extensive collections of labeled images. While these methods demonstrated effective performance, there remains potential for further enhancements. In this paper, we propose to utilize vision language models (VLMs) such as generative pre-trained transformer (GPT), GEMINI, large language and vision assistant (LLAVA), PaliGemma, and Microsoft Florence2 to recognize facial attributes such as race, gender, age, and emotion from images with human faces. Various datasets like FairFace, AffectNet, and UTKFace have been utilized to evaluate the solutions. The results show that VLMs are competitive if not superior to traditional techniques. Additionally, we propose "FaceScanPaliGemma"--a fine-tuned PaliGemma model--for race, gender, age, and emotion recognition. The results show an accuracy of 81.1%, 95.8%, 80%, and 59.4% for race, gender, age group, and emotion classification, respectively, outperforming pre-trained version of PaliGemma, other VLMs, and SotA methods. Finally, we propose "FaceScanGPT", which is a GPT-4o model to recognize the above attributes when several individuals are present in the image using a prompt engineered for a person with specific facial and/or physical attributes. The results underscore the superior multitasking capability of FaceScanGPT to detect the individual's attributes like hair cut, clothing color, postures, etc., using only a prompt to drive the detection and recognition tasks.
... The objective is to test the capability of MaxViT to recognize cognate facial features that implicitly represent the discriminative appearance traits which distinguish one ethnic group from other using facial images. The proposed model is trained on a database created by merging three different ethnicity datasets namely FairFace [10], UTKFace [11], and Arab face dataset [12]. The main contribution is that the proposed model achieves better generalization capabilities compared to other models with an ...
... The experiments were conducted on a database created by merging three datasets FairFace [10], UTKFace [11], and Arab face dataset [12]. Dataset is described in Fig. 3 Fig. 4. Random samples from each dataset are shown in Fig. 5. ...
... The datasets available currently are often based on traditional ethnicity categories. However, the existence of many very specialized datasets (such as in And and Prince [5], Gao et al. [22], Heng et al. [32], Jilani et al. [37], Terada et al. [89]) suggests that technologies used for ethnicity classification are being shaped by their social context, as implied by the Social Shaping of Technology theory [94]. ...
... Since the reporting of ethnicity followed the historical notation, we observed that the majority of algorithms that classify humans according to ethnicity are shaped to classify the same classes. Many studies try to focus on their local ethnical groups (see [5,39,41] for some examples), which is an indicative that systems shaped for localized social contexts are in demand and may present a better fit than general-purpose models. ...
... But, it is hard to remember the PINs and passwords, and also, it is a possibility to guess and steal tokens, keys, and cards very easily [7]. In most cases, the magnetic cards of the individuals are missed, and then it can be misplaced, duplicated, purloined, and forgotten by intruders; hence, it becomes unreadable and corrupted [24]. Using biometric-based techniques, an individual's biological traits cannot be stolen, forgotten, forged, or misplaced [2]. ...
... The LBP [24] operator uses the thresholding approach to label the pixels of an image, where the thresholding operation is performed over the 3 3 nearest of all pixels in accordance with the center value, and the resultant patterns are obtained fro m the face images are binary numbers. Then, the binary numbers are described as a label pres ented in the histogram, which can be utilized as a texture descriptor. ...
... When a pre-trained CNN was trained on the whole Arab dataset and then evaluated on a different dataset, the best result was 56.97 percent, and 52.12 percent when the Arab dataset labels were fairly distributed. [11], In this study, we used facial images to train a machine learning classifier to predict Chinese, Japanese, and Korean. Of all the approaches we provided, CNN had the greatest accuracy rate (89.2%)In the three-class categorization of the three different languages. ...
A new intelligent monitoring model is developed by us for determining gender and nationality from frontal picture candidates utilizing the face area based on deep learning. Face recognition is influenced by a number of characteristics, including picture quality, illumination, rotation angle, blockage, and facial expression. As a result, we must first recognize an input image before converting it to a genuine input. Nationality is the most well distinguishing characteristic that is applied in every nation, and it is also important to secure authentication. Image detection is crucial in this case. Then, we can determine the facial shape, gender, and nationality of the candidate image. In the end, we return the result based on the distance comparison using the use of a library to measure the sample. There were significant discrepancies across photographs while measuring samples based on their gender and facial features. The photos used in the input must be the same as those used in the output. Picture of a frontal face with clean lighting and no blemishes at every angle of rotation. The model may be used by ordinary people, models, celebrities, actors, and others. In the end, computer can tell nationality by looking at a picture of a person’s face.
... Additionally, they provide an in-depth evaluation of many prominent deep network frameworks, including ResNet-50, VGG-Face, VGG-16, and MobileNet v2. Al-Humaidan et al. [16] give an Arab dataset labelled appropriately for Arab subethnic groupings, which they subsequently classify using DL algorithms. Three Arab image datasets were created: GCC, Egyptian, and Levant. ...
Recently, computer vision-based face image analysis has sparked considerable interest in a variety of applications such as surveillance, security, biometrics and so on. The goal of the facial analysis was to derive facial soft biometrics such as identification, gender, age, ethnicity, expression and so on. Among these, ethnicity recognition remains a hot study topic, a major aspect of society with profound linkages to a variety of environmental and social concerns. The introduction of machine learning (ML) and deep learning (ML) technologies has proven advantageous for effective ethnicity recognition and classification. In this regard, the IDL-ERCFI technique, which is based on intelligent DL, is designed in this paper. The IDL-ERCFI technique's purpose is to distinguish and classify ethnicity based on facial photos. The IDL-ERCFI technique uses face landmarks to align photos before sending them to the network. Furthermore, the proposed model employs an Exception network as a feature extractor. Because the retrieved features are high-dimensional, the feature reduction procedure employs the principal component analysis (PCA) technique, which is effective in overcoming the “curse of dimensionality.” Furthermore, the ethnicity classification procedure is carried out using an optimal kernel extreme learning machine (KELM), with parameter tuning of the KELM model carried out using the glow worm swarm optimization (GSO) technique. A complete experimental analysis is carried out to demonstrate the superiority of the IDL-ERCFI technique over the other techniques.
... The gender and face identification helps the recognition approach to focus further on the limited number of entries from a huge database, the efficiency of the retrieval systems, improving the search speed, and identifying the features related to it [13,14]. Furthermore, in many social applications, gender and ethnicity are useful for demographic statistics [45,46]. Unlike identity, the categories of ethnicity are loosely defined classes. ...
One of the biggest outlooks of face recognition is to create a useful application for human identification at immigration offices for security purposes. The human face is a complex visual pattern that consists of general categorical information, identity specification, primary information, and eccentricity. In this domain, the ethnic identification of humans finds its uses in various real-time applications. Yet, automatic identification does not produce prompt results due to complex characteristics and computational complexity. This paper intends to propose automated human ethnicity identification using facial features, comprising three major processes: pre-processing, feature extraction, and classification. At first, the input image is subjected to a pre-processing method, in which the face detection is carried out using the Viola-Jones face detection algorithm. Then, the pre-processed image is subjected to the feature extraction process, where the color feature, texture feature, forehead area extraction, and the improved active appearance model (AAM) based on unique features are extracted. These extracted features are then subjected to the optimized convolutional neural network (CNN) for ethnicity classification. As the major contribution, training of CNN is carried out by the proposed Moth Spiral adopted Grey Wolf Algorithm (MSGWA) model via tuning the optimal weights. Finally, the performance of the proposed work is compared against the adopted and existing approaches on the basis of certain metrics such as NPV, sensitivity, FDR, accuracy, specificity, FPR, precision, MCC, FNR, and F1-score, respectively.
