Remote photoplethysmography (rPPG) has been at the forefront recently, thanks to its capacity in estimating non-contact physiological parameters such as heart rate and heart rate variability (Wang et al. in FBB 6:33, ...
详细信息
Remote photoplethysmography (rPPG) has been at the forefront recently, thanks to its capacity in estimating non-contact physiological parameters such as heart rate and heart rate variability (Wang et al. in FBB 6:33, 2018). rPPG signals are typically extracted from facial videos by performing spatial averaging to obtain temporal RGB traces. Although this spatial averaging simplifies computation, it is accompanied by loss of essential spatial information which might reveal interesting relationships between signals from different spatial regions. In this article, we present a novel algorithm adapted from generalized eigenvalue decomposition (GEVD) to estimate this spatial rPPG distribution. GEVD is an extremely versatile algorithm that finds uses in signal and imageprocessing and analytical problems such as principal component analysis and Fisher discriminant analysis (Ghojogh et al. in Tutorial 2: 1-8, 2019)(Han and Clemmensen in PR 49:43-54, 2016). It is performed using the QZ algorithm (Moler and Stewart in JNA 10(2):241-256, 2010), which in turn uses Householder transformations (Householder in JACM 5(4):339-342, 1958) to extract generalized eigenvectors of a pair of matrices. We adapt the QZ algorithm for the domain of spatio-temporal biomedical signals such as remote photoplethysmography (rPPG), electrocardiography and electroencephalography signals. We call this algorithm Temporal-QZ, which employs vectorization techniques to extract generalized eigenvectors over spatial data points simultaneously. We validate this extension in the domain of remote photoplethysmography (rPPG) measurement, for the estimation of spatial rPPG distribution of skin.
Background The use of different imaging modalities to assist in skin cancer diagnosis is a common practice in clinical scenarios. Different features representative of the lesion under evaluation can be retrieved from ...
详细信息
Background The use of different imaging modalities to assist in skin cancer diagnosis is a common practice in clinical scenarios. Different features representative of the lesion under evaluation can be retrieved from image analysis and processing. However, the integration and understanding of these additional parameters can be a challenging task for physicians, so artificial intelligence (AI) methods can be implemented to assist in this process. This bibliographic research was performed with the goal of assessing the current applications of AI algorithms as an assistive tool in skin cancer diagnosis, based on information retrieved from different imaging modalities. Materials and methods The bibliography databases ISI Web of Science, PubMed and Scopus were used for the literature search, with the combination of keywords: skin cancer, skin neoplasm, imaging and classification methods. Results The search resulted in 526 publications, which underwent a screening process, considering the established eligibility criteria. After screening, only 65 were qualified for revision. Conclusion Different imaging modalities have already been coupled with AI methods, particularly dermoscopy for melanoma recognition. Learners based on support vector machines seem to be the preferred option. Future work should focus on image analysis, processing stages and image fusion assuring the best possible classification outcome.
Precise localization of Wireless Capsule Endoscopy (WCE) inside the curly, long, and compact small intestine remains a challenging problem facing researchers for more than a decade. Conventional Radio Frequency (RF) l...
详细信息
Precise localization of Wireless Capsule Endoscopy (WCE) inside the curly, long, and compact small intestine remains a challenging problem facing researchers for more than a decade. Conventional Radio Frequency (RF) localization techniques, commonly used in outdoor and indoor area, have demonstrated a few centimeters accuracy when applying to the inside of human body. In this paper, using 3D Posterior Cramer-Rao Lower Bound (PCRLB) as a framework for performance evaluation, we demonstrated that millimetric accuracy can be achieved using hybrid RF and imageprocessing localization technique. This level of accuracy enables precise simultaneous localization and mapping of the WCE movement path inside the small intestine. Using the PCRLB framework, we provided in-depth analysis on hybrid localization performance regarding the effects of WCE movement estimation, the effects of system bandwidth as well as the effects of on-body sensor numbers and placements.
The task of document binarization of degraded complex documents is tremendously challenging due to the various forms of noise often present in these documents. While the current state-of-the-art deep learning approach...
详细信息
The task of document binarization of degraded complex documents is tremendously challenging due to the various forms of noise often present in these documents. While the current state-of-the-art deep learning approaches are capable for the removal of various noise types in documents with high accuracy, they employ a supervised learning scheme which requires matching clean and noisy document image pairs which are difficult and costly to obtain for complex documents such as engineering drawings. In this paper, we propose our method for document binarization of engineering drawings using 'Multi Noise CycleGAN'. The method utilizing unsupervised learning using adversarial and cycle-consistency loss is trained on unpaired noisy document images of various noise and image conditions. Experimental results for the removal of various noise types demonstrated that the method is able to reliably produce a clean image for any given noisy image and in certain noisy images achieve significant improvements over existing methods.
The temperature monitoring system (TMS) aims to reduce the infection spread and outbreak of COVID-19 through early detection. Conventional and currently deployed TMS have high implementation cost and require a substan...
详细信息
The temperature monitoring system (TMS) aims to reduce the infection spread and outbreak of COVID-19 through early detection. Conventional and currently deployed TMS have high implementation cost and require a substantial amount of space. Also, the performance often depends on the accuracy of the thermal camera. To address this, we propose Edge TMS wherein a multitask cascaded convolutional neural networks (MTCNNs)based TMS is deployed on an edge AI device. To overcome the resource constraints of edge AI devices, an optimization method is applied to compress MTCNN up to 100x. The compressed MTCNN is deployed on the local PC, Jetson Xavier, Jetson TX2, and Jetson Nano which yields pruning-per-reduction ratio (PPRR) values of 1.21, 1.63, 1.99, and 2.10, respectively. We proposed the PPRR metric to measure the performance of the compressed model. Low PPRR values indicate an improvement in the hardware performance and computational efficiency of the optimized model. The optimized model deployed in all the Jetson series achieved an average percent power reduced (%R) of 53.18% with a percent difference of 35.9% from the results of the local PC.
Feature-based method for detecting landmarks from facial images was designed. The method was based on extracting oriented edges and constructing edge maps at two resolution levels. Edge regions with characteristic edg...
详细信息
Feature-based method for detecting landmarks from facial images was designed. The method was based on extracting oriented edges and constructing edge maps at two resolution levels. Edge regions with characteristic edge pattern formed landmark candidates. The method ensured invariance to expressions while detecting eyes. Nose and mouth detection was deteriorated by happiness and disgust.
Multimodal interfaces offer ever-changing tasks and challenges for designers to accommodate newer technologies, and as these technologies become more accessible, newer application scenarios emerge. Prototype developme...
详细信息
Multimodal interfaces offer ever-changing tasks and challenges for designers to accommodate newer technologies, and as these technologies become more accessible, newer application scenarios emerge. Prototype development and user evaluation are important steps in the creation of solutions to these challenges. Furthermore, playful interactions and games are shown to be important settings to study social signals of interaction people. Research in multimodal analysis brings together people with diverse skills and specializations on the integration of tools in different modalities, to collect and annotate data, and to exchange ideas and skills, and this special issue is a reflection of that collective effort.
The Visual computer receives more than 1000 submissions a year.[...]the Visual computer hosts a few special issues, among them the computer Graphics Conference (CGI) Special Issue.In order to accelerate the publicatio...
详细信息
The Visual computer receives more than 1000 submissions a year.
[...]the Visual computer hosts a few special issues, among them the computer Graphics Conference (CGI) Special Issue.
In order to accelerate the publication of the regular accepted papers published online, this issue contains 26 regular papers: 1 Juan Bajo et al Physically inspired technique for modeling wet absorbent materials 2 Xin Xu et al Salient object detection from low contrast images based on local contrast enhancing and non-local feature learning 3 Prasen Sharma et al Deep learning-based image de-raining using discrete Fourier transformation 4 Nikhil Mhala et al A secure visual secret sharing (VSS) scheme with CNN-based image enhancement for underwater images 5 Xin Ouyang et al A presentation and retrieval hash scheme of images based on principal component analysis 6 Sultan Khan et al Scale and density invariant head detection deep model for crowd counting in pedestrian crowds 7 Changbo Wang et al VEFP: visual evaluation of flight procedure in airport terminal 8 Rahul Singh et al computer-aided diagnostic network for brain tumor classification employing modulated Gabor filter banks 9 Venkatesh Munagala et al Enhanced holoentropy-based encoding via whale optimization for highly efficient video coding 10 Mengxiao Yin et al Three-view generation based on a single front view image for car 11 Aditya Sole et al Measurement and rendering of complex non-diffuse and goniochromatic packaging materials 12 Nikolas Ladas et al Background segmentation in multicolored illumination environments 13 Ji Liu et al Skeleton extraction from point clouds of trees with complex branches via graph contraction 14 Pengyi Hao et al Self-supervised deep subspace clustering network for faces in videos 15 Boguslaw Obara et al The relationship between curvilinear structure enhancement and ridge detection methods 16 Liguo Zhang et al S2RGAN: sonar-image super-resolution based on generative adversarial network 17 Lai-Yu Cheng
暂无评论