Automated analysis of platelet microstructures using a feature length orientation space

The ability to measure elongated structures such as platelets and colonies, is an important step in the microstructural analysis of many materials. Widely used techniques and standards require extensive manual interaction making them slow, laborious, difficult to repeat and prone to human error. Automated approaches have been proposed but often fail when analysing complex microstructures. This paper addresses these challenges by proposing a new, automated image analysis technique, to reliably assess platelet microstructure. Tools from Mathematical Morphology are designed to probe the image and map the response onto a new feature-length orientation space (FLOS). This enables automated measurement of key microstructural features such as platelet width, orientation, globular volume fraction, and colony size. The method has a wide field of view, low dependency on input parameters, and does not require prior thresholding, common in other automated analysis techniques. Multiple datasets of complex Titanium alloys were used to evaluate the new techniques which are shown to match measurements from expert materials scientists using recognized standards, while drastically reducing measurement time and ensuring repeatability. The per-pixel measurement style of the technique also allows for the generation of useful colourmaps, that aid further analysis and provide evidence to increase user confidence in the quantitative measurements.

3D fluorescence microscopy data synthesis for segmentation and benchmarking

Automated image processing approaches are indispensable for many biomedical experiments and help to cope with the increasing amount of microscopy image data in a fast and reproducible way. Especially state-of-the-art deep learning-based approaches most often require large amounts of annotated training data to produce accurate and generalist outputs, but they are often compromised by the general lack of those annotated data sets. In this work, we propose how conditional generative adversarial networks can be utilized to generate realistic image data for 3D fluorescence microscopy from annotation masks of 3D cellular structures. In combination with mask simulation approaches, we demonstrate the generation of fully-annotated 3D microscopy data sets that we make publicly available for training or benchmarking. An additional positional conditioning of the cellular structures enables the reconstruction of position-dependent intensity characteristics and allows to generate image data of different quality levels. A patch-wise working principle and a subsequent full-size reassemble strategy is used to generate image data of arbitrary size and different organisms. We present this as a proof-of-concept for the automated generation of fully-annotated training data sets requiring only a minimum of manual interaction to alleviate the need of manual annotations.

Research on Automatic Construction Method of Three-Dimensional Complex Fault Model

Three-dimensional complex fault modeling is an important research topic in three-dimensional geological structure modeling. The automatic construction of complex fault models has research significance and application value for basic geological theories, as well as engineering fields such as geological engineering, resource exploration, and digital mines. Complex fault structures, especially complex fault networks with multilevel branches, still require a large amount of manual participation in the characterization of fault transfer relationships. This paper proposes an automatic construction method for a three-dimensional complex fault model, including the generation and optimization of fault surfaces, automatic determination of the contact relationship between fault surfaces, and recording of the model. This method realizes the automatic construction of a three-dimensional complex fault model, reduces the manual interaction in model construction, improves the automation of fault model construction, and saves manual modeling time.

Fully-automated root image analysis (faRIA)

High-throughput root phenotyping in the soil became an indispensable quantitative tool for the assessment of effects of climatic factors and molecular perturbation on plant root morphology, development and function. To efficiently analyse a large amount of structurally complex soil-root images advanced methods for automated image segmentation are required. Due to often unavoidable overlap between the intensity of fore- and background regions simple thresholding methods are, generally, not suitable for the segmentation of root regions. Higher-level cognitive models such as convolutional neural networks (CNN) provide capabilities for segmenting roots from heterogeneous and noisy background structures, however, they require a representative set of manually segmented (ground truth) images. Here, we present a GUI-based tool for fully automated quantitative analysis of root images using a pre-trained CNN model, which relies on an extension of the U-Net architecture. The developed CNN framework was designed to efficiently segment root structures of different size, shape and optical contrast using low budget hardware systems. The CNN model was trained on a set of 6465 masks derived from 182 manually segmented near-infrared (NIR) maize root images. Our experimental results show that the proposed approach achieves a Dice coefficient of 0.87 and outperforms existing tools (e.g., SegRoot) with Dice coefficient of 0.67 by application not only to NIR but also to other imaging modalities and plant species such as barley and arabidopsis soil-root images from LED-rhizotron and UV imaging systems, respectively. In summary, the developed software framework enables users to efficiently analyse soil-root images in an automated manner (i.e. without manual interaction with data and/or parameter tuning) providing quantitative plant scientists with a powerful analytical tool.

Design and optimization of CNC based polishing process

In recent years many aerospace manufacturing corporations have taken an initiative to eliminate the manual interaction of workers and moving work pieces. At the same time there is an ever-increasing demand for efficiency in the manufacturing sector. The research performed investigates the design and development of passive compliance polishing tools for both external and internal polishing of parts on a CNC turning center. The goal of this thesis is to develop tools that meet new safety demands while at the same time optimizing the tools through the Taguchi method and process planning in order to meet quality standards cost effectively. Experiments on both test pieces and production parts were conducted not only at Ryerson University but at Messier-Dowty as well.

Design and optimization of CNC based polishing process

In recent years many aerospace manufacturing corporations have taken an initiative to eliminate the manual interaction of workers and moving work pieces. At the same time there is an ever-increasing demand for efficiency in the manufacturing sector. The research performed investigates the design and development of passive compliance polishing tools for both external and internal polishing of parts on a CNC turning center. The goal of this thesis is to develop tools that meet new safety demands while at the same time optimizing the tools through the Taguchi method and process planning in order to meet quality standards cost effectively. Experiments on both test pieces and production parts were conducted not only at Ryerson University but at Messier-Dowty as well.

Time Measurement and Control Blocks for Bare-Metal C++ Applications

Precisely timed execution of resource constrained bare-metal applications is difficult, because the embedded software developer usually has to implement and check the timeliness of the executed application through manual interaction with timers or counters. In the scope of this work, we propose a combined timing specification and concept for time annotation and control blocks in C++. Our proposed blocks can be used to measure and profile software block execution time. Furthermore, it can be used to control and enforce the software time behavior at runtime. After the application of these time blocks, a trace-based verification against the block-based timing specification can be performed to obtain evidence on the correct implementation and usage of the time blocks on the target platform. We have implemented our time block concept in a C++ library and tested it on an ARM Cortex A9 bare-metal platform. The combined usage of timing specification and our time block library has been successfully evaluated on a critical flight-control software for a multi-rotor system.

Pix2Prof: fast extraction of sequential information from galaxy imagery via a deep natural language “captioning” model

We present “Pix2Prof”, a deep learning model that can eliminate any manual steps taken when measuring galaxy profiles. We argue that a galaxy profile of any sort is conceptually similar to a natural language image caption. This idea allows us to leverage image captioning methods from the field of natural language processing, and so we design Pix2Prof as a float sequence “captioning” model suitable for galaxy profile inference. We demonstrate the technique by approximating a galaxy surface brightness (SB) profile fitting method that contains several manual steps. Pix2Prof processes ∼1 image per second on an Intel Xeon E5-2650 v3 CPU, improving on the speed of the manual interactive method by more than two orders of magnitude. Crucially, Pix2Prof requires no manual interaction, and since galaxy profile estimation is an embarrassingly parallel problem, we can further increase the throughput by running many Pix2Prof instances simultaneously. In perspective, Pix2Prof would take under an hour to infer profiles for 105 galaxies on a single NVIDIA DGX-2 system. A single human expert would take approximately two years to complete the same task. Automated methodology such as this will accelerate the analysis of the next generation of large area sky surveys expected to yield hundreds of millions of targets. In such instances, all manual approaches – even those involving a large number of experts – will be impractical.

A novel method for extracting skeleton of fruit tree from 3D point clouds

Tree skeleton could be useful to agronomy researchers because the skeleton describes the shape and topological structure of a tree. The phenomenon of organs’ mutual occlusion in fruit tree canopy is usually very serious, this should result in a large amount of data missing in directed laser scanning 3D point clouds from a fruit tree. However, traditional approaches can be ineffective and problematic in extracting the tree skeleton correctly when the tree point clouds contain occlusions and missing points. To overcome this limitation, we present a method for accurate and fast extracting the skeleton of fruit tree from laser scanner measured 3D point clouds. The proposed method selects the start point and endpoint of a branch from the point clouds by user’s manual interaction, then a backward searching is used to find a path from the 3D point cloud with a radius parameter as a restriction. The experimental results in several kinds of fruit trees demonstrate that our method can extract the skeleton of a leafy fruit tree with highly accuracy.