A Simulated Environment for Robot Vision Experiments

Training on simulation data has proven invaluable in applying machine learning in robotics. However, when looking at robot vision in particular, simulated images cannot be directly used no matter how realistic the image rendering is, as many physical parameters (temperature, humidity, wear-and-tear in time) vary and affect texture and lighting in ways that cannot be encoded in the simulation. In this article we propose a different approach for extracting value from simulated environments: although neither of the trained models can be used nor are any evaluation scores expected to be the same on simulated and physical data, the conclusions drawn from simulated experiments might be valid. If this is the case, then simulated environments can be used in early-stage experimentation with different network architectures and features. This will expedite the early development phase before moving to (harder to conduct) physical experiments in order to evaluate the most promising approaches. In order to test this idea we created two simulated environments for the Unity engine, acquired simulated visual datasets, and used them to reproduce experiments originally carried out in a physical environment. The comparison of the conclusions drawn in the physical and the simulated experiments is promising regarding the validity of our approach.

Simulated sensor characterization for virtual clinical trials in mammography and digital breast tomosynthesis

Virtual clinical trials in X-ray breast imaging permit to compare different technical solutions and imaging modalities at reduced costs, involved personnel, reduced times and reduced radiation risks to patients. In this context, the detector characteristics (spatial resolution, noise level and efficiency) play a key role for an appropriate generation of simulated images. The project AGATA proposes to compute images as dose deposit maps in a detector layer of defined materials. Simulated images are then post-processed on the basis of suitable comparison between intrinsic characteristics of real and simulated detectors. With this scope, as first step for the post-processing manipulations, we evaluated the presampled modulation transfer function (MTF), the detector-response function and the noise power spectrum (NPS) of the simulated detectors. Two detectors were simulated: (1) 0.20 mm-thick a-Se direct flat panel with 70 µm pixel pitch and (2) CsI(Tl) indirect flat panel with 100 µm pixel pitch and scintillator layer 0.25 mm thick. In addition, the impact of simulating the de-excitation processes (Auger emission and fluorescence) was explored. Simulated detector characteristics were evaluated for W/Rh spectra between 25 kV and 31 kV. The in-silico platform used a Monte Carlo software based on Geant4 toolkit (vers. 6). First, the simulation and tracking of electrons generated from photoelectric or Compton interactions was shown to have neglectable influence on the pixel values for the explored spectra, with the produced electrons presenting short ranges with respect to the pixel dimension. In the case of the CsI detector, which has fluorescence energies higher than those of the simulated X-ray photons, the deexcitation processes have not noticeable influence on the calculated pixel values. On the other hand, the MTF of the a-Se detector resulted slightly lower when the fluorescence is simulated in the detector materials, due to the dose spread derived from the fluorescence photons, which can travel far from the initial ionization interaction. Regarding the a-Se detector, the noise power spectrum resulted lower with simulated deexcitation.

Landscape Luminaire Design for Part of the Conservation of Chudhadhuj Royal Residence, Sichang Island

Chudhadhuj Royal Residence is a historically significant place, located on Si Chang island where a vast landscape area covers both coastal and mountainous areas. Its context connects natural landscape and culture which can be termed a cultural landscape in a holistic. The researcher would like to describe the link between the design criteria and the concepts of the cultural heritage conservation and some identities of the place to meet the function of each area in designing luminaires. The objective of this article is to demonstrate the design process of landscape luminaires for walkway. The design of prototype luminaires consists of a lamppost, a bollard, and a bollard with perforated patterns. Visual design elements and principles with the design concepts were applied to these luminaires. The image of luminaires and the effect of light of each prototype were designed, developed, and tested on-site both day and night-time. The questionnaires displaying computer-simulated images in the actual context were used for an assessment by experts from various design and architectural conservation fields. The results show that these luminaires respond to architectural conservation and identities of the place both day and night-time. The prototype luminaires were improved according to the comments of the experts. Consequently, these luminaires were granted three design patents. Ultimately, the designed luminaires and the lighting master plan were applied to the landscape improvement project of the Chudhadhuj Royal Residence, operated by the Fine Arts Department. This project can be an example of architectural conservation for a cultural heritage site in the future.

Synchronizability of Multi-Layer-Coupled Star-Composed Networks

In this paper, several multi-layer-coupled star-composed networks with similar symmetrical structures are defined by using the theory of graph operation. The supra-Laplacian matrix of the corresponding multi-layer networks is obtained according to the master stability equation (MSF). Two important indexes that reflect the synchronizability of these kinds of networks are derived in the case of bounded and unbounded synchronized regions. The relationships among the synchronizability, the number of layers, the length of the paths, the branchings, and the interlayer and intralayer coupling strengths in the two cases are studied. At the same time, the simulation experiments are carried out with the MATLAB software, and the simulated images of the two symmetrical structure networks’ synchronizability are compared. Finally, the factors affecting the synchronizability of multi-layer-coupled star-composed networks are found. On this basis, optimization schemes are given to improve the synchronizability of multi-layer-coupled star-composed networks and the influences of the number of central nodes on the networks’ synchronizability are further studied.

Feasibility of Four-Dimensional Similarity Filter For Radiation Dose Reduction in Dynamic Myocardial Computed Tomography Perfusion Imaging

Purpose: We aimed to evaluate the impact of four-dimensional noise reduction filtering using a similarity algorithm (4D-SF) on image noise during dynamic myocardial computed tomography perfusion (CTP) to simulate the reduction of radiation dose.Methods: A total of 43 patients who underwent dynamic myocardial CTP using 320-row CT were included in the study. The original images were reconstructed using iterative reconstruction (IR); three different CTP datasets with simulated noise, which corresponded to 25%, 50%, and 75% reduction of the original dose (= 300mA), were reconstructed using a combination of IR and 4D-SF. The signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were assessed, and CT-derived myocardial blood flow (CT-MBF) was quantified. The results were compared between the original and simulated images with radiation dose reduction.Results: The original, 25%-, 50%-, and 75%-dose reduced images with 4D-SF showed an SNR of 8.3 (6.5–10.2), 16.5 (11.9–21.7), 15.6 (11.0–20.1), and 12.8 (8.8–18.1) and a CNR of 4.4 (3.2–5.8), 6.7 (4.6–10.3), 6.6 (4.3–10.1), and 5.5 (3.5–9.1), respectively. Compared to the original images, the 25%-, 50%-, and 75%-dose reduced-simulated images showed significant improvement in both SNR and CNR with 4D-SF. There was no significant difference in CT-MBF between the original and 25%- or 50%-dose reduced-simulated images with 4D-SF, however, there was a significant difference in CT-MBF between the original and 75%-dose reduced-simulated images.Conclusion: 4D-SF has the potential to reduce the radiation dose associated with dynamic myocardial CTP imaging by half, without impairing the robustness of MBF quantification.

An efficient hexagonal image framework using pseudo hexagonal pixel for computer vision applications

Nature-inspired computing has been a real source of motivation for the development of many meta-heuristic algorithms. The biological optic system can be patterned as a cascade of sub-filters from the photoreceptors over the ganglion cells in the fovea to some simple cells in the visual cortex. This spark has inspired many researchers to examine the biological retina in order to learn more about information processing capabilities. The photoreceptor cones and rods in the human fovea resemble hexagon more than a rectangular structure. However, the hexagonal meshes provide higher packing density, consistent neighborhood connectivity, and better angular correction compared to the rectilinear square mesh. In this paper, a novel 2-D interpolation hexagonal lattice conversion algorithm has been proposed to develop an efficient hexagonal mesh framework for computer vision applications. The proposed algorithm comprises effective pseudo-hexagonal structures which guarantee to keep align with our human visual system. It provides the hexagonal simulated images to visually verify without using any hexagonal capture or display device. The simulation results manifest that the proposed algorithm achieves a higher Peak Signal-to-Noise Ratio of 98.45 and offers a high-resolution image with a lesser mean square error of 0.59.

Evaluation and Intercomparison of Topographic Correction Methods Based on Landsat Images and Simulated Data

Topographic effects in medium and high spatial resolution remote sensing images greatly limit the application of quantitative parameter retrieval and analysis in mountainous areas. Many topographic correction methods have been proposed to reduce such effects. Comparative analyses on topographic correction algorithms have been carried out, some of which drew different or even contradictory conclusions. Performances of these algorithms over different terrain and surface cover conditions remain largely unknown. In this paper, we intercompared ten widely used topographic correction algorithms by adopting multi-criteria evaluation methods using Landsat images under various terrain and surface cover conditions as well as images simulated by a 3D radiative transfer model. Based on comprehensive analysis, we found that the Teillet regression-based models had the overall best performance in terms of topographic effects’ reduction and overcorrection; however, correction bias may be introduced by Teillet regression models when surface reflectance in the uncorrected images do not follow a normal distribution. We recommend including more simulated images for a more in-depth evaluation. We also recommend that the pros and cons of topographic correction methods reported in this paper should be carefully considered for surface parameters retrieval and applications in mountain regions.

Using 4DCBCT simulation and guidance to evaluate inter-fractional tumor variance during SABR for lung tumor within the lower lobe

Inter-fractional tumor variance would lead to insufficient dosage or overdose in tumor region during lung cancer radiotherapy. However, previous works have not considered influence of inter-fractional tumor amplitude variance at treatment position due to lack of effective evaluation method during radiotherapy, especially for lung tumor within the lower lobe. Our objective was to investigate inter-fractional tumor baseline shift and amplitude variance due to respiratory motion with 4DCBCT simulation and guidance during stereotactic ablative body radiotherapy (SABR) for lung tumor. Subject included 19 patients with lung tumor within the lower lobe. 4DCBCT-simulated images at treatment position were acquired sequentially to determine internal tumor volume (ITV) and reference tumor motion at simulation process. Compared with reference tumor motion, 95 4DCBCT-guided images were acquired during each treatment to evaluate inter-fractional tumor baseline shift and amplitude variance, which were − 0.0 ± 1.3 mm and − 0.2 ± 1.4 mm in left–right(LR) direction, 0.9 ± 2.3 mm and 0.4 ± 2.9 mm in superior-inferior (SI) direction, 0.1 ± 1.5 mm and − 0.4 ± 2.0 mm in anterior–posterior (AP) direction. ITV margin were 3.5 mm, 7.5 mm and 5.3 mm in LR, SI and AP directions with van Herk’s (Int J Radiat Oncol Biol Phys 52(5):1407–1422, 2002) formula. 4DCBCT simulation and guidance is a reliable method to evaluate inter-fractional tumor variance during SABR for lung tumor within the lower lobe. ITV margin of 3.5 mm, 7.5 mm and 5.3 mm in LR, SI and AP directions would ensure greater tumor coverage during SABR for lung tumor within the lower lobe.