scholarly journals Single Target SAR 3D Reconstruction Based on Deep Learning

Sensors ◽  
2021 ◽  
Vol 21 (3) ◽  
pp. 964
Author(s):  
Shihong Wang ◽  
Jiayi Guo ◽  
Yueting Zhang ◽  
Yuxin Hu ◽  
Chibiao Ding ◽  
...  
Keyword(s):  
Deep Learning ◽  
3D Reconstruction ◽  
Time Window ◽  
Learning Algorithm ◽  
Signal To Noise Ratio ◽  
Super Resolution ◽  
Mapping Method ◽  
Deep Learning Algorithm ◽  
Deep Learning Network ◽  
Voxel Grid

Synthetic aperture radar tomography (TomoSAR) is an important 3D mapping method. Traditional TomoSAR requires a large number of observation orbits however, it is hard to meet the requirement of massive orbits. While on the one hand, this is due to funding constraints, on the other hand, because the target scene is changing over time and each observation orbit consumes lots of time, the number of orbits can be fewer as required within a narrow time window. When the number of observation orbits is insufficient, the signal-to-noise ratio (SNR), peak-to-sidelobe ratio (PSR), and resolution of 3D reconstruction results will decline severely, which seriously limits the practical application of TomoSAR. In order to solve this problem, we propose to use a deep learning network to improve the resolution and SNR of 3D reconstruction results under the condition of very few observation orbits by learning the prior distribution of targets. We use all available orbits to reconstruct a high resolution target, while only very few (around 3) orbits to reconstruct a low resolution input. The low-res and high-res 3D voxel-grid pairs are used to train a 3D super-resolution (SR) CNN (convolutional neural network) model, just like ordinary 2D image SR tasks. Experiments on the Civilian Vehicle Radar dataset show that the proposed deep learning algorithm can effectively improve the reconstruction both in quality and in quantity. In addition, the model also shows good generalization performance for targets not shown in the training set.

scholarly journals Combining deep learning with 3D stereophotogrammetry for craniosynostosis diagnosis

2020 ◽  
Vol 10 (1) ◽  
Author(s):  
Guido de Jong ◽  
Elmar Bijlsma ◽  
Jene Meulstee ◽  
Myrte Wennen ◽  
Erik van Lindert ◽  
...  
Keyword(s):  
Deep Learning ◽  
Learning Algorithm ◽  
Normal Brain ◽  
Healthy Controls ◽  
Learning Network ◽  
Deep Learning Algorithm ◽  
Anterior Plagiocephaly ◽  
Deep Learning Network ◽  
Shape Data ◽  
Cranial Shape

Abstract Craniosynostosis is a condition in which cranial sutures fuse prematurely, causing problems in normal brain and skull growth in infants. To limit the extent of cosmetic and functional problems, swift diagnosis is needed. The goal of this study is to investigate if a deep learning algorithm is capable of correctly classifying the head shape of infants as either healthy controls, or as one of the following three craniosynostosis subtypes; scaphocephaly, trigonocephaly or anterior plagiocephaly. In order to acquire cranial shape data, 3D stereophotographs were made during routine pre-operative appointments of scaphocephaly (n = 76), trigonocephaly (n = 40) and anterior plagiocephaly (n = 27) patients. 3D Stereophotographs of healthy infants (n = 53) were made between the age of 3–6 months. The cranial shape data was sampled and a deep learning network was used to classify the cranial shape data as either: healthy control, scaphocephaly patient, trigonocephaly patient or anterior plagiocephaly patient. For the training and testing of the deep learning network, a stratified tenfold cross validation was used. During testing 195 out of 196 3D stereophotographs (99.5%) were correctly classified. This study shows that trained deep learning algorithms, based on 3D stereophotographs, can discriminate between craniosynostosis subtypes and healthy controls with high accuracy.

scholarly journals Reconstruction of cardiovascular black-blood T2-weighted image by deep learning algorithm: A comparison with intensity filter

2021 ◽  
Vol 10 (9) ◽  
pp. 205846012110447
Author(s):  
Ryo Ogawa ◽  
Tomoyuki Kido ◽  
Masashi Nakamura ◽  
Atsushi Nozaki ◽  
R Marc Lebel ◽  
...  
Keyword(s):  
Deep Learning ◽  
Image Quality ◽  
Image Quality Assessment ◽  
Learning Algorithm ◽  
Signal To Noise Ratio ◽  
Contrast Ratio ◽  
Black Blood ◽  
Endocardial Border ◽  
Deep Learning Algorithm ◽  
Subjective Scale

Background Deep learning–based methods have been used to denoise magnetic resonance imaging. Purpose The purpose of this study was to evaluate a deep learning reconstruction (DL Recon) in cardiovascular black-blood T2-weighted images and compare with intensity filtered images. Material and Methods Forty-five DL Recon images were compared with intensity filtered and the original images. For quantitative image analysis, the signal to noise ratio (SNR) of the septum, contrast ratio (CR) of the septum to lumen, and sharpness of the endocardial border were calculated in each image. For qualitative image quality assessment, a 4-point subjective scale was assigned to each image (1 = poor, 2 = fair, 3 = good, 4 = excellent). Results The SNR and CR were significantly higher in the DL Recon images than in the intensity filtered and the original images ( p < .05 in each). Sharpness of the endocardial border was significantly higher in the DL Recon and intensity filtered images than in the original images ( p < .05 in each). The image quality of the DL Recon images was significantly better than that of intensity filtered and original images ( p < .001 in each). Conclusions DL Recon reduced image noise while improving image contrast and sharpness in the cardiovascular black-blood T2-weight sequence.

scholarly journals Image Super-Resolution Based on CNN Using Multilabel Gene Expression Programming

2020 ◽  
Vol 10 (3) ◽  
pp. 854 ◽  
Author(s):  
Jiali Tang ◽  
Chenrong Huang ◽  
Jian Liu ◽  
Hongjin Zhu
Keyword(s):  
Gene Expression ◽  
Deep Learning ◽  
Image Restoration ◽  
Learning Algorithm ◽  
Gene Expression Programming ◽  
Super Resolution ◽  
Image Block ◽  
Deep Learning Algorithm ◽  
Deep Convolution Neural Network ◽  
Image Super Resolution

Current mainstream super-resolution algorithms based on deep learning use a deep convolution neural network (CNN) framework to realize end-to-end learning from low-resolution (LR) image to high-resolution (HR) images, and have achieved good image restoration effects. However, as the number of layers in the network is increased, better results are not necessarily obtained, and there will be problems such as slow training convergence, mismatched sample blocks, and unstable image restoration results. We propose a preclassified deep-learning algorithm (MGEP-SRCNN) using Multilabel Gene Expression Programming (MGEP), which screens out a sample sub-bank with high relevance to the target image before image block extraction, preclassifies samples in a multilabel framework, and then performs nonlinear mapping and image reconstruction. The algorithm is verified through standard images, and better objective image quality is obtained. The restoration effect under different magnification conditions is also better.

scholarly journals Detecting objects behind scattering media using vortex beams and deep learning

2021 ◽  
Author(s):  
Ganesh M. Balasubramaniam ◽  
Netanel Biton ◽  
Shlomi Arnon
Keyword(s):  
Deep Learning ◽  
Learning Algorithm ◽  
Signal To Noise Ratio ◽  
Computer Assisted Surgery ◽  
Computer Assisted ◽  
Scattering Media ◽  
Deep Learning Algorithm ◽  
Main Challenge ◽  
Vortex Beams ◽  
Set Up

Abstract Reconstructing objects behind scattering media is a challenging issue with applications in biomedical imaging, non-distractive testing, computer-assisted surgery, and autonomous vehicular systems. Such systems’ main challenge is the multiple scattering of the photons in the angular and spatial domain, which results in a blurred image. Previous works try to improve the reconstructing ability using deep learning algorithms, with some success. We enhance these methods by illuminating the set-up using several modes of vortex beams obtaining a series of time-gated images corresponding to each mode. The images are accurately reconstructed using a deep learning algorithm by analyzing the pattern captured in the camera. This study shows that using vortex beams instead of Gaussian beams enhances the deep learning algorithm’s image reconstruction ability in terms of the peak signal-to-noise ratio (PSNR) by ~ 2.5 dB and ~1 dB when low and high scattering scatterers are used respectively.

scholarly journals Diffeomorphic unsupervised deep learning model for mono- and multi-modality registration

2020 ◽  
Vol 14 ◽  
pp. 174830262097352
Author(s):  
Anis Theljani ◽  
Ke Chen
Keyword(s):  
Deep Learning ◽  
Learning Algorithm ◽  
Training Data ◽  
Variational Model ◽  
Learning Network ◽  
Deep Learning Algorithm ◽  
Unsupervised Deep Learning ◽  
Deep Learning Network ◽  
Affine Transforms ◽  
Deep Learning Model

Different from image segmentation, developing a deep learning network for image registration is less straightforward because training data cannot be prepared or supervised by humans unless they are trivial (e.g. pre-designed affine transforms). One approach for an unsupervised deep leaning model is to self-train the deformation fields by a network based on a loss function with an image similarity metric and a regularisation term, just with traditional variational methods. Such a function consists in a smoothing constraint on the derivatives and a constraint on the determinant of the transformation in order to obtain a spatially smooth and plausible solution. Although any variational model may be used to work with a deep learning algorithm, the challenge lies in achieving robustness. The proposed algorithm is first trained based on a new and robust variational model and tested on synthetic and real mono-modal images. The results show how it deals with large deformation registration problems and leads to a real time solution with no folding. It is then generalised to multi-modal images. Experiments and comparisons with learning and non-learning models demonstrate that this approach can deliver good performances and simultaneously generate an accurate diffeomorphic transformation.

scholarly journals Classification of Retinal Image for Early Detection of Diabetic Retinopathy Using Deep Learning

2021 ◽  
Vol 9 (8) ◽  
pp. 1552-1555
Author(s):  
Piyush Kumar
Keyword(s):  
Diabetic Retinopathy ◽  
Deep Learning ◽  
Early Detection ◽  
Learning Algorithm ◽  
Retinal Images ◽  
Data Sets ◽  
Deep Learning Algorithm ◽  
The Difference ◽  
Learning Machine ◽  
Deep Learning Network

Abstract: Diabetic retinopathy is a disease which cause of blindness due to diabetes. For this reason, it is very important to detect diabetic retinopathy in early stages. A deep learning-based approach is used for the early detection of diabetic retinopathy from retinal images. The proposed approach consists of two steps. In the first stage, pre treatments were performed to remove retinal images from different data sets and standardize them to size. In the second stage, classification is done by the help of Convolutional Neural Network using deep learning algorithm and 98.5% success is achieved. The difference of this technique from similar studies is that instead of creating the feature set manually as in traditional methods, the deep learning network automatically constructs and trained by using CPU and GPU in a very short time. Keywords: CNN, Early detection, Artificial intelligence, Deep learning, Machine-learning, Fundus Image, Optical coherence tomography, Ophthalmology.

Automating the Detection of Dynamically Triggered Earthquakes via a Deep Metric Learning Algorithm

2020 ◽  
Vol 91 (2A) ◽  
pp. 901-912 ◽  
Author(s):  
Vivian Tang ◽  
Prem Seetharaman ◽  
Kevin Chao ◽  
Bryan A. Pardo ◽  
Suzan van der Lee
Keyword(s):  
Deep Learning ◽  
Learning Algorithm ◽  
Signal To Noise Ratio ◽  
Metric Learning ◽  
Signal To Noise ◽  
Large Magnitude ◽  
Triggered Earthquakes ◽  
Learning Network ◽  
Deep Metric Learning ◽  
Deep Learning Network

Abstract Detecting subtle signals from small earthquakes triggered by transient stresses from the surface waves of large magnitude earthquakes can contribute to a more general understanding of how earthquakes nucleate and interact with each other. However, searching for signals from such small earthquakes in thousands of seismograms is overwhelming, and discriminating them from a miscellany of noise is challenging. Here, we explore how we can automate the detection of such dynamically triggered earthquakes using a simple, diagnostic signal-to-noise ratio (SNR) threshold as well as a convolutional deep metric learning network. Our analysis shows that the deep learning network was more reliable at detecting small earthquakes than the SNR method.

Sign in / Sign up

Export Citation Format

Share Document