Abstract.Face recognition is an important biometric method because of its potential applications in many fields, such as access control and surveillance. In surveillance applications, the distance between the subject and the camera is changing. Thus, in this paper, the effect of the distance between the subject and the camera, distance class, the effect of the number of images per class, and also the effect of database used for training have been investigated. The images in the database were equally divided into three classes: CLOSE, MEDIUM, and FAR, according to the distance of the subject from the camera.

In the past decade, face recognition has gained an important role among the most frequently used image processing applications and the availability of viable technologies in this field has also contributed significantly to this. Face recognition has become an enabler in healthcare, surveillance, photo cataloging, attendance, and much more, which will be discussed in this review paper. Despite rapid progress in face-recognition technology, various challenges such as variations, occlusion, facial expressions, aging and many more that affect the performance of the system still need to be addressed. This paper presents a review on the state-of-the-art, enablers, challenges and solutions for face recognition. Face recognition can be categorized into three groups; namely global approach, local feature approach, and hybrid approach. The global approach uses the whole face as input for the face recognition system. The local approach uses measurements between important landmarks of a face and certain face region selection for training. The hybrid approach blends global and local approaches in which the hybrid approach uses the best global and local approach methods. The challenges of face recognition are; (i) automated face detection where difficulties lies on detecting a person's face, (ii) pose variations cause by rotation of people's head, (iii) face occlusion, (iv) facial expression changes, (v) ageing of the face, (vi) varying illumination conditions, (vii) low image resolution, (viii) identity look-alike or twin, and (ix) other technical difficulties. Finally, the solutions to each of the highlighted challenges were described. The survey found that all the images considered for training and testing were made up of RGB images. With the rapid growth of computer technology in terms of computing speed and the increasingly sophisticated functions of smartphones, multispectral or even hyperspectral imagery could be considered for face-recognition research.

No abstract

In gastronomic tourism, food is viewed as the central tourist attraction. Specifically, indigenous food is known to represent the expression of local culture and identity. To promote gastronomic tourism, it is critical to have a model for the food business analytics system. This research undertakes an empirical evaluation of recent transfer learning models for deep learning feature extraction for a food recognition model. The VIREO-Food172 Dataset and a newly established Sabah Food Dataset are used to evaluate the food recognition model. Afterwards, the model is implemented into a web application system as an attempt to automate food recognition. In this model, a fully connected layer with 11 and 10 Softmax neurons is used as the classifier for food categories in both datasets. Six pre-trained Convolutional Neural Network (CNN) models are evaluated as the feature extractors to extract essential features from food images. From the evaluation, the research found that the EfficientNet feature extractor-based and CNN classifier achieved the highest classification accuracy of 94.01% on the Sabah Food Dataset and 86.57% on VIREO-Food172 Dataset. EFFNet as a feature representation outperformed Xception in terms of overall performance. However, Xception can be considered despite some accuracy performance drawback if computational speed and memory space usage are more important than performance.

No abstract

Face recognition is an important biometric method because of its potential applications in many fields, such as access control and surveillance. In this paper, the performance of the individual channels from the YCBCR color space on face recognition for surveillance applications is investigated and compared with the performance of the gray scale. In addition, the performance of fusing two or more color channels is also compared with that of the gray scale. Three cases with different number of training images per persons were used as a test bed. It was found out that, the gray scale always outperforms the individual channel. However, the fusion of CBxCR with any other channel outperforms the gray scale when three images of the same class from the same database are used for training. Regardless of the cases used, the CBxCR channel always gave the best performance for the individual color channels. It was found that, in general, increasing the number of fused channels increases the performance of the system. It was also found that the gray scale channel is the better choice for face recognition since it contain better quality edges and visual features which are essential for face recognition.

The COVID-19 pandemic has resulted in a global health crisis. The rapid spread of the virus has led to the infection of a significant population and millions of deaths worldwide. Therefore, the world is in urgent need of a fast and accurate COVID-19 screening. Numerous researchers have performed exceptionally well to design pioneering deep learning (DL) models for the automatic screening of COVID-19 based on computerised tomography (CT) scans; however, there is still a concern regarding the performance stability affected by tiny perturbations and structural changes in CT images. This paper proposes a fusion of a moment invariant (MI) method and a DL algorithm for feature extraction to address the instabilities in the existing COVID-19 classification models. The proposed method incorporates the MI-based features into the DL models using the cascade fusion method. It was found that the fusion of MI features with DL features has the potential to improve the sensitivity and accuracy of the COVID-19 classification. Based on the evaluation using the SARS-CoV-2 dataset, the fusion of VGG16 and Hu moments shows the best result with 90% sensitivity and 93% accuracy.

No abstract