16 research outputs found
On automatic age estimation from facial profile view
No full textYesIn recent years, automatic facial age estimation has gained popularity due to its numerous applications. Much work has been done on frontal images and lately, minimal estimation errors have been achieved on most of the benchmark databases. However, in reality, images obtained in unconstrained environments are not always frontal. For instance, when conducting a demographic study or crowd analysis, one may get profile images of the face. To the best of our knowledge, no attempt has been made to estimate ages from the side-view of face images. Here we exploit this by using a pre-trained deep residual neural network (ResNet) to extract features. We then utilize a sparse partial least squares regression approach to estimate ages. Despite having less information as compared to frontal images, our results show that the extracted deep features achieve a promising performance
A nonlinear appearance model for age progression
No full textNoRecently, automatic age progression has gained popularity due to its nu-merous applications. Among these is the search for missing people, in the UK alone up to 300,000 people are reported missing every year. Although many algorithms have been proposed, most of the methods are affected by image noise, illumination variations, and most importantly facial expres-sions. To this end we propose to build an age progression framework that utilizes image de-noising and expression normalizing capabilities of kernel principal component analysis (Kernel PCA). Here, Kernel PCA a nonlinear form of PCA that explores higher order correlations between input varia-bles, is used to build a model that captures the shape and texture variations of the human face. The extracted facial features are then used to perform age progression via a regression procedure. To evaluate the performance of the framework, rigorous tests are conducted on the FGNET ageing data-base. Furthermore, the proposed algorithm is used to progress images of Mary Boyle; a six-year-old that went missing over 39 years ago, she is considered Ireland's youngest missing person. The algorithm presented in this paper could potentially aid, among other applications, the search for missing people worldwide
On facial age progression based on modified active appearance models with face texture
No full textNoAge progression that involves the reconstruction of facial appearance with a natural ageing effect has several applications. These include the search for missing people and identification of fugitives. The majority of age progression methods reported in the literature are data driven. Hence, such methods learn from training data and utilise statistical models such as 3D morphable models and active appearance models (AAM). Principal component analysis (PCA) which is a vital part of these models has an unfortunate drawback of averaging out texture details. Therefore, they work as a low pass filter and as such many of the face skin deformations and minor details become faded. Interestingly, recent work in 2D and 3D animation has shown that patches of the human face are somewhat similar when compared in isolation. Thus, researchers have proposed generating novel faces by compositing small face patches, usually from large image databases. Following these ideas, we propose a novel age progression model which synthesises aged faces using a hybrid of these two techniques. First, an invertible model of age synthesis is developed using AAM and sparse partial least squares regression (sPLS). Then the texture details of the face are enhanced using the patch-based synthesis approach. Our results show that the hybrid algorithm produces both unique and realistic images. Furthermore, our method demonstrates that the identity and ageing effects of subjects can be more emphasised
Discrimination of Healthy Skin, Superficial Epidermal Burns, and Full-Thickness Burns from 2D-Colored Images Using Machine Learning
No full text
Transfer learning based histopathologic image classification for burns recognition
No full textBurn is one of the most leading devastating
injuries affecting people worldwide with high impact rate in
low-and middle-income countries subjecting hundreds of
thousands to loss of lives and physical deformities. Both
affected individuals and health institutions are faced with
challenges such as inadequate experience/well trained
workforce and high diagnostics cost. The demand of having
efficient, cost-effective and user-friendly technique to aid in addressing the problem is on the rise. Deep neural networks have recently attracted the attention of many researchers and achieved impressive results in many applications. Therefore, this paper proposed the use of off-the-shelf Convolutional Neural Network features from two ImageNet pre-trained models (GoogleNet and ResNet152), VGG-Face. The features are used to train Support Vector Machine (SVM) and Decision Tree (DT). 100% identification accuracy was recorded using ImageNet model and SVM
On the ethnic classification of Pakistani face using deep learning
No full textDemographic-based identification plays an active role in the field of face identification. Over the past decade, machine learning algorithms have been used to investigate challenges surrouding ethnic classification for specific populations, such as African, Asian and Caucasian people. Ethnic classification for individuals of South Asian, Pakistani heritage, however, remains to be addressed. The present paper addresses a two-category (Pakistani Vs Non-Pakistani) classification task from a novel, purpose-built dataset. To the best of our knowledge, this work is the first to report a machine learning ethnic classification task with South Asian (Pakistani) faces. We conduted a series of experiments using deep learning algorithms (ResNet-50, ResNet-101 and ResNet-152) for feature extraction and a linear support vector machine (SVM) for classification. The experimental results demonstrate ResNet-101 achieves the highest performance accuracy of 99.2% for full-face ethnicity classification, followed closely by 91.7% and 95.7% for the nose and mouth respectively
An approach to failure prediction in a cloud based environment
No full textYesFailure in a cloud system is defined as an even that occurs when the delivered service deviates from the correct intended behavior. As the cloud computing systems continue to grow in scale and complexity, there is an urgent need for cloud service providers (CSP) to guarantee a reliable on-demand resource to their customers in the presence of faults thereby fulfilling their service level agreement (SLA). Component failures in cloud systems are very familiar phenomena. However, large cloud service providers' data centers should be designed to provide a certain level of availability to the business system. Infrastructure-as-a-service (Iaas) cloud delivery model presents computational resources (CPU and memory), storage resources and networking capacity that ensures high availability in the presence of such failures. The data in-production-faults recorded within a 2 years period has been studied and analyzed from the National Energy Research Scientific computing center (NERSC). Using the real-time data collected from the Computer Failure Data Repository (CFDR), this paper presents the performance of two machine learning (ML) algorithms, Linear Regression (LR) Model and Support Vector Machine (SVM) with a Linear Gaussian kernel for predicting hardware failures in a real-time cloud environment to improve system availability. The performance of the two algorithms have been rigorously evaluated using K-folds cross-validation technique. Furthermore, steps and procedure for future studies has been presented. This research will aid computer hardware companies and cloud service providers (CSP) in designing a reliable fault-tolerant system by providing a better device selection, thereby improving system availability and minimizing unscheduled system downtime
African Neuroscience on the Global Stage: Nigeria as a Model
No full textSeveral challenges contribute to Africa’s trailing position in the global production of knowledge. Decades of focused work through international and local programmes have thus far been unable to lift the continent onto its scientific feet. To learn more about the strengths and weaknesses of neuroscience research carried out on the continent today, that would enable the development of robust programmes focusing on specific needs, a strategy is required to extract information about specific contributions of African laboratories. Nigeria, Africa’s most populous nation, is among the top beneficiaries of international programmes promoting neuroscience research in Africa. Therefore, to establish and test a framework for evaluating neuroscience output from the continent, we here focussed on Nigeria’s neuroscience publications over the past two decades. Using PubMed key-word search and defined exclusion criteria, we extracted 572 neuroscience articles from Nigeria-based laboratories published between 1996 and 2017. Articles were automatically categorised into clinical and epidemiological studies (55.5%) or basic neuroscience (44.5%) using a support vector machine and decision tree algorithm. From here, we extracted each publication’s use of model species, methods, citations received and the publishing journal’s metrics.
We find that over the 21 year period surveyed, only one Nigerian-led neuroscience paper was published in a “top-tier” international journal with an impact factor of >8. However, about half (55%) of PubMed indexed articles were published in reputable journals with an impact factor between 1-4. These publications primarily comprised basic (61%), rather than clinical and epidemiological studies (39%) which were instead mostly published in lower-ranking journals. Next, we find a worrying account of model species and research tools employed in Nigerian-based neuroscience. For example, no studies used genetically amenable model systems such as zebrafish, Drosophila, C.elegans, or transgenic mouse strains. Instead, popular model species were human (54%), rat (30%) and wild-type mice (11%). In line, research techniques employed were dominated by “basic” techniques such as Hematoxylin and Eosin stainings or classical behavioural analysis, with only 8% of studies using more modern techniques like PCR, Western blotting or forms of fluorescence microscopy. Perhaps as one consequence, even though medicinal plants have been used to treat diseases for decades by locals, and 41% of basic neuroscience studies investigated their potential utility in treating disease, none made it into local clinical research.
Together, these findings highlight two clear access points for the support of Nigerian neuroscience in the future: Investment in the training and infrastructure in the use of more modern research techniques, and the widespread promotion of genetically amenable model species. Moreover, any such effort might consider specifically targeting existing basic over clinical or epidemiological research efforts. In time, it will be important to also assess the neuroscience output across the entire continent
