A multimodal deep learning framework using local feature representations for face recognition

Al-Waisy, Alaa S.; Qahwaji, Rami; Ipson, Stanley S.; Al-Fahdawi, Shumoos

doi:10.1007/s00138-017-0870-2

Cited by 38 publications

(18 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…By taking advantage of the flexibility and trainability of deep neural networks [26][27][28], a similarity maximization framework is designed to produce a full-resolution hyperspectral cube from a low-resolution hyperspectral mosaic using a CNN.…”

Section: Introductionmentioning

confidence: 99%

Hyperspectral demosaicking and crosstalk correction using deep learning

Dijkstra

Loosdrecht

Schomaker

et al. 2018

Machine Vision and Applications

View full text Add to dashboard Cite

Precision agriculture using unmanned aerial vehicles (UAVs) is gaining popularity. These UAVs provide a unique aerial perspective suitable for inspecting agricultural fields. With the use of hyperspectral cameras, complex inspection tasks are being automated. Payload constraints of UAVs require low weight and small hyperspectral cameras; however, such cameras with a multispectral color filter array suffer from crosstalk and a low spatial resolution. The research described in this paper aims to reduce crosstalk and to increase spatial resolution using convolutional neural networks. We propose a similarity maximization framework which is trained to perform end-to-end demosaicking and crosstalk-correction of a 4 × 4 raw mosaic image. The proposed method produces a hyperspectral image cube with 16 times the spatial resolution of the original cube while retaining a median structural similarity (SSIM) index of 0.85 (compared to an SSIM of 0.55 when using bilinear interpolation). Furthermore, this paper provides insight into the beneficial effects of crosstalk for hyperspectral demosaicking and gives best practices for several architectural and hyperparameter variations as well as a theoretical reasoning behind certain observations.

show abstract

Section: Introductionmentioning

confidence: 99%

Hyperspectral demosaicking and crosstalk correction using deep learning

Dijkstra

Loosdrecht

Schomaker

et al. 2018

Machine Vision and Applications

View full text Add to dashboard Cite

show abstract

“…The training set images have been rescaled, zoomed, sheared, and horizontally flipped and the test set images have been rescaled through this generator. The images have been augmented in this way so that the model would never see twice the exact same image and it helps the model to prevent overfitting and helps the model generalize better [29,30]. The flow_from_directory() method has been used to generate the batches of image data and prepare the data generator from the training set and test set images.…”

Section: Data Augmentation Is One Of the Methods Of Artificiallymentioning

confidence: 99%

Verification of Biometric Traits using Deep Learning

2019

IJITEE

View full text Add to dashboard Cite

Multimodal biometric systems have been widely applied in many real-world applications due to its ability to deal with a number of significant limitations of unimodal biometric systems including non-universality, noise, population coverage, vulnerability and intra-class variability for verification, authentication and identification of an individual. In this paper, the impact of deep learning in the field of biometrics is investigated where supervised learning is primarily involved in identifying biometric traits using Graphical User Interface. The trained deep learning system proposed is called MultiTraitConvNet whose architecture is based on a combination of Convolutional Neural Network (CNN) and Softmax classifier to extract discriminative features from the input image without prior domain knowledge and classify the class of the biometric trait. A discriminative CNN training scheme is based on a combination of back-propagation algorithm and mini-batch Adam optimization method that is used for weights updating and learning rate adaption, respectively. To evaluate various CNN architecture data augmentation and dropout method training techniques have been used. A centralized database of different biometric trait images has been created using data sets of CASIA1 for iris trait, google 11K hand for palmprint trait, for face UTKFace is used. The accuracy of the proposed system is found to be 100%. Python library Keras is used to develop CNN model and TKinter is used to create GUI of the proposed system.

show abstract

“…Deep learning networks (DNNs) have been efficiently employed in the medical field with remarkable results and significant performance compared with the human-level performance in various challenging image analysis and classification tasks (Al-Waisy et al 2017a , b , 2018 ). Several medical imaging systems based on deep learning approaches have also been employed to support the clinicians in the early detection of COVID-19 infection, treatment, and follow-up investigation (De Fauw et al 2018 ).…”

Section: Related Workmentioning

confidence: 99%

COVID-CheXNet: hybrid deep learning framework for identifying COVID-19 virus in chest X-rays images

Al-Waisy¹,

et al. 2020

Self Cite

View full text Add to dashboard Cite

The outbreaks of Coronavirus (COVID-19) epidemic have increased the pressure on healthcare and medical systems worldwide. The timely diagnosis of infected patients is a critical step to limit the spread of the COVID-19 epidemic. The chest radiography imaging has shown to be an effective screening technique in diagnosing the COVID-19 epidemic. To reduce the pressure on radiologists and control of the epidemic, fast and accurate a hybrid deep learning framework for diagnosing COVID-19 virus in chest X-ray images is developed and termed as the COVID-CheXNet system. First, the contrast of the X-ray image was enhanced and the noise level was reduced using the contrast-limited adaptive histogram equalization and Butterworth bandpass filter, respectively. This was followed by fusing the results obtained from two different pre-trained deep learning models based on the incorporation of a ResNet34 and high-resolution network model trained using a large-scale dataset. Herein, the parallel architecture was considered, which provides radiologists with a high degree of confidence to discriminate between the healthy and COVID-19 infected people. The proposed COVID-CheXNet system has managed to correctly and accurately diagnose the COVID-19 patients with a detection accuracy rate of 99.99%, sensitivity of 99.98%, specificity of 100%, precision of 100%, F1-score of 99.99%, MSE of 0.011%, and RMSE of 0.012% using the weighted sum rule at the score-level. The efficiency and usefulness of the proposed COVID-CheXNet system are established along with the possibility of using it in real clinical centers for fast diagnosis and treatment supplement, with less than 2 s per image to get the prediction result. Keywords Coronavirus COVID-19 epidemic Á Deep learning Á Transfer learning Á ResNet34 model Á Chest radiography imaging Á Chest X-ray images Communicated by Valentina E. Balas.

show abstract

A multimodal deep learning framework using local feature representations for face recognition

Cited by 38 publications

References 59 publications

Hyperspectral demosaicking and crosstalk correction using deep learning

Hyperspectral demosaicking and crosstalk correction using deep learning

Verification of Biometric Traits using Deep Learning

COVID-CheXNet: hybrid deep learning framework for identifying COVID-19 virus in chest X-rays images

Contact Info

Product

Resources

About