• Corpus ID: 18930788

Super-resolution using Sparse Representations over Learned Dictionaries: Reconstruction of Brain Structure using Electron Microscopy

  title={Super-resolution using Sparse Representations over Learned Dictionaries: Reconstruction of Brain Structure using Electron Microscopy},
  author={Tao Hu and Juan Nunez-Iglesias and Shiv Naga Prasad Vitaladevuni and Louis K. Scheffer and Shan Xu and Mehdi Bolorizadeh and Harald F. Hess and Richard D. Fetter and Dmitri B. Chklovskii},
A central problem in neuroscience is reconstructing neuronal circuits on the synapse level. Due to a wide range of scales in brain architecture such reconstruction requires imaging that is both high-resolution and high-throughput. Existing electron microscopy (EM) techniques possess required resolution in the lateral plane and either high-throughput or high depth resolution but not both. Here, we exploit recent advances in unsupervised learning and signal processing to obtain high depth… 

Deep Learning for Isotropic Super-Resolution from Non-isotropic 3D Electron Microscopy

This work has adapted one of the most successful architectures (FSRCNN) for 3D super-resolution, and compared its performance to a 3D U-Net architecture that has not been used previously to generate super- resolution.

Sparse imaging for fast electron microscopy

This work proposes and demonstrates on an operational SEM a fast method to sparsely sample and reconstruct smooth images, and reports image fidelity as a function of acquisition speed by comparing traditional raster to sparse imaging modes.

SuperSlicing Frame Restoration for Anisotropic ssTEM and Video Data

The details and additional evaluation of an approach originally introduced in Laptev et al. (2014) called SUPERSLICING to decompose the observed frame into a sequence of plausible hidden sub-frames are presented and a novel automated method to perform neuronal structure segmentation is proposed.

Neural Connectivity Reconstruction from Calcium Imaging Signal using Random Forest with Topological Features

A method of incorporating topological knowledge inside data representation for Random Forest classifier in order to reconstruct the neural connections from patterns of their activities leads to the model competitive with state-of-the art methods like Deep Convolutional Neural Networks and Graph Decomposition techniques.

Image Super-Resolution with Fast Approximate Convolutional Sparse Coding

This work presents a computationally efficient architecture for image super-resolution that achieves state-of-the-art results on images with large spatial extend and empirically shows that upsampling methods work much better on latent representations than in the original spatial domain.

Superslicing frame restoration for anisotropic sstem

This work presents an approach called SuperSlicing to decompose the observed frame into a sequence of plausible hidden sub-frames and proposes a novel automated method to perform neuronal structure segmentation.

Interpolation and denoising of high-dimensional seismic data by learning a tight frame

ABSTRACTSparse transforms play an important role in seismic signal processing steps, such as prestack noise attenuation and data reconstruction. Analytic sparse transforms (so-called implicit



Increasing depth resolution of electron microscopy of neural circuits using sparse tomographic reconstruction

First, it is shown that the brain tissue can be represented as sparse linear combination of local basis functions that are thin membrane-like structures oriented in various directions that enables tracing of neuronal connections across layers and, hence, high throughput reconstruction of neural circuits to the level of individual synapses.

Supervised Learning of Image Restoration with Convolutional Networks

This work shows that convolutional networks can be used as a general method for low-level image processing and suggests that high model complexity is the single most important factor for good performance.

Image super-resolution as sparse representation of raw image patches

It is shown that a small set of randomly chosen raw patches from training images of similar statistical nature to the input image generally serve as a good dictionary, in the sense that the computed representation is sparse and the recovered high-resolution image is competitive or even superior in quality to images produced by other SR methods.

Natural Image Denoising with Convolutional Networks

An approach to low-level vision is presented that combines the use of convolutional networks as an image processing architecture and an unsupervised learning procedure that synthesizes training samples from specific noise models to avoid computational difficulties in MRF approaches that arise from probabilistic learning and inference.

Serial Block-Face Scanning Electron Microscopy to Reconstruct Three-Dimensional Tissue Nanostructure

It is demonstrated that datasets meeting these requirements can be obtained by automated block-face imaging combined with serial sectioning inside the chamber of a scanning electron microscope, opening the possibility of automatically obtaining the electron-microscope-level 3D datasets needed to completely reconstruct the connectivity of neuronal circuits.

Image Super-Resolution Via Sparse Representation

This paper presents a new approach to single-image superresolution, based upon sparse signal representation, which generates high-resolution images that are competitive or even superior in quality to images produced by other similar SR methods.

Image Deblurring and Super-Resolution by Adaptive Sparse Domain Selection and Adaptive Regularization

Extensive experiments on image deblurring and super-resolution validate that by using adaptive sparse domain selection and adaptive regularization, the proposed method achieves much better results than many state-of-the-art algorithms in terms of both PSNR and visual perception.

Image Denoising Via Sparse and Redundant Representations Over Learned Dictionaries

This work addresses the image denoising problem, where zero-mean white and homogeneous Gaussian additive noise is to be removed from a given image, and uses the K-SVD algorithm to obtain a dictionary that describes the image content effectively.

Natural image statistics and efficient coding.

It is suggested that a good objective for an efficient coding of natural Scenes is to maximize the sparseness of the representation, and it is shown that a network that learns sparse codes of natural scenes succeeds in developing localized, oriented, bandpass receptive fields similar to those in the mammalian striate cortex.

The Emergence of Electron Tomography as an Important Tool for Investigating Cellular Ultrastructure

  • B. McEwenM. Marko
  • Biology
    The journal of histochemistry and cytochemistry : official journal of the Histochemistry Society
  • 2001
The technology has matured to the point at which application of electron tomography to specimens in plastic sections is routine, and new developments to overcome limitations due to beam exposure and specimen geometry promise to further improve its capabilities.