C++

WND-CHARM is a multi-purpose image classifier that can be applied to a wide variety of image classification tasks without modifications or fine-tuning, and yet provides classification accuracy comparable to state-of-the-art task-specific image classifiers. WND-CHARM can extract up to ~3,000 generic image descriptors (features) including polynomial decompositions, high contrast features, pixel statistics, and textures. These features are derived from the raw image, transforms of the image, and compound transforms of the image (transforms of transforms). The features are filtered and weighted depending on their effectiveness in discriminating between a set of predefined image classes (the training set). These features are then used to classify test images based on their similarity to the training classes. This classifier was tested on a wide variety of imaging problems including biological and medical image classification using several imaging modalities, face recognition, and other pattern recognition tasks. WND-CHARM is an acronym that stands for "Weighted Neighbor Distance using Compound Hierarchy of Algorithms Representing Morphology."

"we present a new fully automated 3D reconstruction algorithm, called TReMAP, short for Tracing, Reverse Mapping and Assembling of 2D Projections. Instead of tracing a 3D image directly in the 3D space as seen in majority of the tracing methods, we first trace the 2D projection trees in 2Dplanes, followed by reverse-mapping the resulting 2D tracing results back into the 3D space as 3D curves; then we use a minimal spanning tree (MST) method to assemble all the 3D curves to generate the final 3D reconstruction. Because we simplify a 3D reconstruction problem into 2D, the computational costs are reduced dramatically." 

Suitable for high throughput neuron image analysis (image sizes >10GB). This plugin can be used with default parameters or user-defined parameters.

All-path-pruning 2.0 (APP2) is a component of Vaa3D. APP2 prunes an initial reconstruction tree of a neuron’s morphology using a long-segment-first hierarchical procedure instead of the original termini-first-search process in APP. APP2 computes the distance transform of all image voxels directly for a gray-scale image, without the need to binarize the image before invoking the conventional distance transform. APP2 uses a fast-marching algorithm-based method to compute the initial reconstruction trees without pre-computing a large graph. This method allows to trace large images. This method can be used with default parameters or user-defined parameters.

Vaa3D is a handy, fast, and versatile 3D/4D/5D Image Visualization and Analysis System for Bioimages and Surface Objects. It also provides many unique functions that you may not find in other software. It is Open Source, and supports a very simple and powerful plugin interface and thus can be extended and enhanced easily.

Vaa3D is cross-platform (Mac, Linux, and Windows). This software suite is powerful for visualizing large- or massive-scale (giga-voxels and even tera-voxels) 3D image stacks and various surface data. Vaa3D is also a container of powerful modules for 3D image analysis (cell segmentation, neuron tracing, brain registration, annotation, quantitative measurement and statistics, etc) and data management. This makes Vaa3D suitable for various bioimage informatics applications, and a nice platform to develop new 3D image analysis algorithms for high-throughput processing. In short, Vaa3D streamlines the workflow of visualization-assisted analysis.

Vaa3D can render 5D (spatial-temporal) data directly in 3D volume-rendering mode; it supports convenient and interactive local and global 3D views at different scales... it comes with a number of plugins and toolboxes. Importantly, you can now write your own plugins to take advantage of the Vaa3D platform, possibly within minutes!

"We have developed an automatic graph algorithm, called the all-path pruning (APP), to trace the 3D structure of a neuron. To avoid potential mis-tracing of some parts of a neuron, an APP first produces an initial over-reconstruction, by tracing the optimal geodesic shortest path from the seed location to every possible destination voxel/pixel location in the image. Since the initial reconstruction contains all the possible paths and thus could contain redundant structural components (SC), we simplify the entire reconstruction without compromising its connectedness by pruning the redundant structural elements, using a new maximal- covering minimal-redundant (MCMR) subgraph algorithm. We show that MCMR has a linear computational complexity and will converge. We examined the performance of our method using challenging 3D neuronal image datasets of model organisms (e.g. fruit fly)"

This plugin can be used with default parameters or user-defined parameters.