Search results for "convolution"
showing 10 items of 334 documents
Depth-of-Field Enhancement in Integral Imaging by Selective Depth-Deconvolution
2014
One of the major drawbacks of the integral imaging technique is its limited depth of field. Such limitation is imposed by the numerical aperture of the microlenses. In this paper, we propose a method to extend the depth of field of integral imaging systems in the reconstruction stage. The method is based on the combination of deconvolution tools and depth filtering of each elemental image using disparity map information. We demonstrate our proposal presenting digital reconstructions of a 3-D scene focused at different depths with extended depth of field.
Introduction to Digital Signal Processing
2018
Signal processing deals with the representation, transformation, and manipulation of signals and the information they contain. Typical examples include extracting the pure signals from a mixture observation (a field commonly known as deconvolution) or particular signal (frequency) components from noisy observations (generally known as filtering). This chapter outlines the basics of signal processing and then introduces the more advanced concepts of time‐frequency and time‐scale representations, as well as emerging fields of compressed sensing and multidimensional signal processing. When moving to multidimensional signal processing, a modern approach is taken from the point of view of statis…
Deep learning algorithms for gravitational waves core-collapse supernova detection
2021
The detection of gravitational waves from core-collapse supernova (CCSN) explosions is a challenging task, yet to be achieved, in which it is key the connection between multiple messengers, including neutrinos and electromagnetic signals. In this work, we present a method for detecting these kind of signals based on machine learning techniques. We tested its robustness by injecting signals in the real noise data taken by the Advanced LIGO-Virgo network during the second observation run, O2. We trained three newly developed convolutional neural networks using time-frequency images corresponding to injections of simulated phenomenological signals, which mimic the waveforms obtained in 3D nume…
Energy balance in single exposure multispectral sensors
2013
International audience; Recent simulations of multispectral sensors are based on a simple Gaussian model, which includes filters transmittance and substrate absorption. In this paper we want to make the distinction between these two layers. We discuss the balance of energy by channel in multispectral solid state sensors and propose an updated simple Gaussian model to simulate multispectral sensors. Results are based on simulation of typical sensor configurations.
An attention-based weakly supervised framework for spitzoid melanocytic lesion diagnosis in whole slide images
2021
[EN] Melanoma is an aggressive neoplasm responsible for the majority of deaths from skin cancer. Specifically, spitzoid melanocytic tumors are one of the most challenging melanocytic lesions due to their ambiguous morphological features. The gold standard for its diagnosis and prognosis is the analysis of skin biopsies. In this process, dermatopathologists visualize skin histology slides under a microscope, in a highly time-consuming and subjective task. In the last years, computer-aided diagnosis (CAD) systems have emerged as a promising tool that could support pathologists in daily clinical practice. Nevertheless, no automatic CAD systems have yet been proposed for the analysis of spitzoi…
Automatic Sleep Stage Identification with Time Distributed Convolutional Neural Network
2021
Polysomnography (PSG), the gold standard for sleep stage classification, requires a sleep expert for scoring and is both resource-intensive and expensive. Many researchers currently focus on the real-time classification of the sleep stages based on biomedical signals, such as Electroencephalograph (EEG) and electrooculography (EOG). However, most of the research work is based on machine learning models with multiple signal inputs or hand-engineered features requiring prior knowledge of the sleep domain. We propose a novel encoded Time-Distributed Convolutional Neural Network (TDConvNet) to automatically classify sleep stages based on a single raw PSG signal. The TDConvNet can infer sleep st…
Deconvolution of XPS spectra
1991
The resolution of XPS spectra is limited mainly by instrumental parameters like the spectral line width of the exciting X-ray source and the finite energy resolution of the electron analyzer. If the line broadening functions resulting from the instrumental setup can be estimated and expressed by a spectrometer function, a mathematical recalculation of the intrinsic signal is possible by deconvolution. With the method presented in this paper, a resolution enhancement by a factor of 3 can be obtained. Measured spectra of physically correlated spin orbit doublets have been deconvoluted, and it is shown, that the intensity ratios and the positions are comparable with results obtained by highly …
TheINTEGRALspectrometer SPI: performance of point-source data analysis
2005
The performance of the SPI point-source data analysis system is assessed using a combination of simulations and of observations gathered during the first year of INTEGRAL operations. External error estimates are derived by comparing source positions and fluxes obtained from independent analyses. When the source detection significance provided by the SPIROS imaging reconstruction program increases from ∼10 to ∼100, the errors decrease as the inverse of the detection significance, with values from ∼10 to ∼1 arcmin in positions, and from ∼10 to ∼1 per cent in relative flux. These errors are dominated by Poisson counting noise. Our error estimates are consistent with those provided by the SPIRO…
Block Based Deconvolution Algorithm Using Spline Wavelet Packets
2010
This paper presents robust algorithms to deconvolve discrete noised signals and images. The idea behind the algorithms is to solve the convolution equation separately in different frequency bands. This is achieved by using spline wavelet packets. The solutions are derived as linear combinations of the wavelet packets that minimize some parameterized quadratic functionals. Parameters choice, which is performed automatically, determines the trade-off between the solution regularity and the initial data approximation. This technique, which id called Spline Harmonic Analysis, provides a unified computational scheme for the design of orthonormal spline wavelet packets, fast implementation of the…
A convolutional neural network for virtual screening of molecular fingerprints
2019
In the last few years, Deep Learning (DL) gained more and more impact on drug design because it allows a huge increase of the prediction accuracy in many stages of such a complex process. In this paper a Virtual Screening (VS) procedure based on Convolutional Neural Networks (CNN) is presented, that is aimed at classifying a set of candidate compounds as regards their biological activity on a particular target protein. The model has been trained on a dataset of active/inactive compounds with respect to the Cyclin-Dependent Kinase 1 (CDK1) a very important protein family, which is heavily involved in regulating the cell cycle. One qualifying point of the proposed approach is the use of molec…