Entire Object Research Articles

РАЗРАБОТКА МЕТОДИКИ СОЗДАНИЯ ТРЕХМЕРНЫХ ВИДЕОСЦЕН УЧАСТКОВ ТЕРРИТОРИИ ПРИ СТРОИТЕЛЬСТВЕ ОБЪЕКТОВ ХИМИЧЕСКИХ ПАРКОВ В РЕСПУБЛИКЕ КАЗАХСТАН

The article is devoted to some issues concerning the improvement of techniques for stages control, while building the objects of chemical enterprises on territories in Kazakhstan. Nowadays, one of the types of control is the one through single photographs obtained by means of digital cam-eras. But, even visual inspection of the entire object is not possible with the application of single im-ages. The authors suggest to conduct aerial photographic survey of an object, photogrammetric im-ages processing, the object or the local areas 3D-model design while performing any work stage or its completion. The Agisoft PhotoScan Programme for experimental studies was chosen. The survey was carried out with Huawei Nova 2i RNE-L21 Graphite Black smartphone. As a result, due to to-photogrammetric images processing, the point digital model, providing visual object correctness was obtained.

Learning Linear Regression via Single-Convolutional Layer for Visual Object Tracking

Learning a large-scale regression model has proven to be one of the most successful approaches for visual tracking as in recent correlation filter (CF)- based trackers. Different from the conventional CF-based algorithms in which the regression model is solved based on circulant training samples, we propose learning linear regression models via a single-convolutional layer with the gradient descent (GD) technique. In our convolution-based approach, the samples are cropped from an image in a sliding-window manner rather than being circularly shifted from one base sample. As a result, the abundant background context in the images can be fully exploited to learn a robust tracker. The proposed tracker is based on two independent regression models: a holistic regression model and a texture regression model. The holistic regression model is trained based on the entire object patch to predict the object location, whereas the texture regression model is trained based on the local object textures. The foreground map outputted by the texture regression model is not only helpful to boost the location prediction in the case of large variations, but is also an important clue for estimating the object size. With the foreground map outputted by the texture regression model, we are able to estimate the object size by optimizing a novel objective function based on object-background contrast. Our extensive experiments on four popular visual tracking datasets OTB-50, OTB-100, VOT-2016, and TempleColor have proved that the proposed algorithm achieves outstanding performance and outperforms most CF-based trackers.

Higher order, multifeatural object encoding by the oculomotor system.

Previous behavioral and physiological research has demonstrated that as the behavioral relevance of potential saccade goals increases, they elicit more competition during target selection processing as evidenced by increased saccade curvature and neural activity. However, these effects have only been demonstrated for lower order feature singletons, and it remains unclear whether more complicated featural differences between higher order objects also elicit vector modulation. Therefore, we measured human saccades curvature elicited by distractors bilaterally flanking a target during a visual search saccade task and systematically varied subsets of features shared between the two distractors and the target, referred to as objective similarity (OS). Our results demonstrate that saccades deviated away from the distractor highest in OS to the target and that there was a linear relationship between the magnitude of saccade deviation and the number of feature differences between the most similar distractor and the target. Furthermore, an analysis of curvature over the time course of the saccade demonstrated that curvature only occurred in the first 20-30 ms of the movement. Given the multifeatural complexity of the novel stimuli, these results suggest that saccadic target selection processing involves dynamically reweighting vector representations for movement planning to several possible targets based on their behavioral relevance. NEW & NOTEWORTHY We demonstrate that small featural differences between unfamiliar, higher order object representations modulate vector weights during saccadic target selection processing. Such effects have previously only been demonstrated for familiar, simple feature singletons (e.g., color) in which features characterize entire objects. The complexity and novelty of our stimuli suggest that the oculomotor system dynamically receives visual/cognitive information processed in the higher order representational networks of the cortical visual processing hierarchy and integrates this information for saccadic movement planning.

Calipso: physics-based image and video editing through CAD model proxies

We present Calipso, an interactive method for editing images and videos in a physically-coherent manner. Our main idea is to realize physics-based manipulations by running a full physics simulation on proxy geometries given by non-rigidly aligned CAD models. Running these simulations allows us to apply new, unseen forces to move or deform selected objects, change physical parameters such as mass or elasticity, or even add entire new objects that interact with the rest of the underlying scene. In our method the user makes edits directly in 3D; these edits are processed by the simulation and then transfered to the target 2D content using shape-to-image correspondences in a photo-realistic rendering process. To align the CAD models, we introduce an efficient CAD-to-image alignment procedure that jointly minimizes for rigid and non-rigid alignment while preserving the high-level structure of the input shape. Moreover, the user can choose to exploit image flow to estimate scene motion, producing coherent physical behavior with ambient dynamics. We demonstrate physics-based editing on a wide range of examples producing myriad physical behavior while preserving geometric and visual consistency.

Manga face detection based on deep neural networks fusing global and local information

As more and more digitized manga (Japanese comics) books are available, efficient and effective access to manga is urgently needed. Among various elements of manga, character’s face plays one of the most important roles in access and retrieval. We propose a deep neural network method to do manga face detection, which is a challenging but relatively unexplored topic. Given a manga page, we first find candidate regions based on the selective search scheme. Three convolutional neural networks are then proposed to detect manga faces of various appearance. We extract information from the entire object region and several local regions, and integrate multi-scale information in an early fusion manner or a late fusion manner. The proposed methods are evaluated based on a large-scale benchmark. Convincing performance compared to the state-of-the-art face detection modules designed for human faces is demonstrated.

Strategies for data stream mining method applied in anomaly detection

Anomaly detection, which is a method of intrusion detection, detects anomaly behaviors and protects network security. Data mining technology has been integrated to improve the performance of anomaly detection and some algorithms have been improved for anomaly detection field. We think that most data mining algorithms are analyzed on static data sets and ignore the influence of dynamic data streams. Data stream is the potentially unbounded, ordered sequence of data objects which arrive over time. The entire data objects cannot be stored and they need to be handled in one-time scanning. The data distribution of data stream may change over time and this phenomenon is called concept drift. The properties of data stream make analysis method different from the method based on data set and the analysis model is required to be updated immediately when concept drift occurs. In this paper, we summarize the characteristics of data stream, compare the difference between data stream and data set, discuss the problems of data stream mining and propose some corresponding strategies.

Local Binary Pattern Circuit Generator With Adjustable Parameters for Feature Extraction

In the field of computer vision, local binary pattern (LBP) is one of the most popular feature extraction method and has been used in many object detection frameworks. To efficiently extract LBP features in high-resolution images, hardware architecture is needed to disperse CPU burden and to improve the entire object detection performance. In this paper, a hardware implementation of an approximated LBP method with adjustable parameters is introduced. For simulation, Taiwan Semiconductor Manufacturing Company <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$0.18~\mu \text{m}$ </tex-math></inline-formula> technology is used to implement the LBP hardware, and the hardware can achieve 500 MHz with lower gate count than previous study. The proposed LBP circuit is applied to the pedestrian classification application and the evaluation results show that the approximated LBP values generated by our circuit can achieve comparable classification accuracy with the primitive LBP method. Additionally, the proposed LBP hardware provides adjustable parameters to fit different applications while requires fewer hardware costs as compared with the existing work.

3D modelling of AGB stars with CO5BOLD

AbstractLocal three-dimensional radiation-hydrodynamics simulations of patches of the surfaces of solar-type stars, that are governed by small-scale granular convection, have helped analyzing and interpreting observations for decades. These models contributed considerably to the understanding of the atmospheres and indirectly also of the interiors and the active layers above the surface of these stars. Of great help was of course the availability of a close-by prototype of these stars – the sun.In the case of an asymptotic-giant-branch (AGB) star, the convective cells have sizes comparable to the radius of the giant. Therefore, the extensions of the solar-type-star simulations to AGB stars have to be global and cover the entire object, including a large part of the convection zone, the molecule-formation layers in the inner atmosphere, and the dust-formation region in the outer atmosphere. Three-dimensional radiation-hydrodynamics simulations with CO5BOLD show how the interplay of large and small convection cells, waves, pulsations, and shocks, but also molecular and dust opacities of AGB stars create conditions very different from those in the solar atmosphere.Recent CO5BOLD models account for frequency-dependent radiation transport and the formation of two independent dust species for an oxygen-rich composition. The drop of the comparably smooth temperature distribution below a threshold determines to onset of dust formation, further in, at higher temperatures, for aluminium oxides (Al2O3) than for silicates (Mg2SiO4). An uneven dust distribution is mostly caused by inhomogeneities in the density of the shocked gas.

Image Co-Saliency Detection and Co-Segmentation via Progressive Joint Optimization.

We present a novel computational model for simultaneous image co-saliency detection and co-segmentation that concurrently explores the concepts of saliency and objectness in multiple images. It has been shown that the co-saliency detection via aggregating multiple saliency proposals by diverse visual cues can better highlight the salient objects; however, the optimal proposals are typically region-dependent and the fusion process often leads to blurred results. Co-segmentation can help preserve object boundaries, but it may suffer from complex scenes. To address these issues, we develop a unified method that addresses co-saliency detection and co-segmentation jointly via solving an energy minimization problem over a graph. Our method iteratively carries out the region-wise adaptive saliency map fusion and object segmentation to transfer useful information between the two complementary tasks. Through the optimization iterations, sharp saliency maps are gradually obtained to recover entire salient objects by referring to object segmentation, while these segmentations are progressively improved owing to the better saliency prior. We evaluate our method on four public benchmark data sets while comparing it to the state-of-the-art methods. Extensive experiments demonstrate that our method can provide consistently higher-quality results on both co-saliency detection and co-segmentation.

Quantitative region-of-interest tomography using variable field of view.

In X-ray computed tomography, the task of imaging only a local region of interest (ROI) inside a larger sample is very important. However, without a priori information, this ROI cannot be exactly reconstructed using only the image data limited to the ROI. We propose here an approach of region-of-interest tomography, which reconstructs a ROI within an object from projections of different fields of view acquired on a specific angular sampling scheme in the same tomographic experiment. We present a stable procedure that not only yields high-quality images of the ROI but keeps as well the quantitative contrast on the reconstructed images. In addition, we analyze the minimum number of projections required for ROI tomography from the point of view of the band region of the Radon transform, which confirms this number must be estimated based on the size of the entire object and not only on the size of the ROI.

Joint Reconstruction of Tracer Distribution and Background in Magnetic Particle Imaging.

Magnetic particle imaging (MPI) is a novel tomographic imaging technique, which visualizes the distribution of a magnetic nanoparticle-based tracer material. However, reconstructed MPI images often suffer from an insufficiently compensated image background caused by rapid non-deterministic changes in the background signal of the imaging device. In particular, the signal-to-background ratio (SBR) of the images is reduced for lower tracer concentrations or longer acquisitions. The state-of-the-art procedure in MPI is to frequently measure the background signal during the sample measurement. Unfortunately, this requires a removal of the entire object from the scanner's field of view (FOV), which introduces dead time and repositioning artifacts. To overcome these considerable restrictions, we propose a novel method that uses two consecutive image acquisitions as input parameters for a simultaneous reconstruction of the tracer distribution, as well as the background signal. The two acquisitions differ by just a small spatial shift, while keeping the object always within the focus of a slightly reduced FOV. A linearly interpolated background between the initial and final background measurement is used to seed the iterative reconstruction. The method has been tested with simulations and phantom measurements. Overall, a substantial reduction of the image background was observed, and the image SBR is increased by a factor of 2(7) for the measurement (simulation) data.

Non-destructive evaluation of additively manufactured polymer objects using X-ray interferometry

X-ray interferometry provides a dark-field image, essentially a small-angle X-ray scattering image, of the voids and print defects in an additively manufactured polymer object. The interferometers used were tuned to scattering length 2–5 μm and configured to measure scattering along both vertical and horizontal directions. The samples studied included Stanford Bunnies, fabricated from acrylonitrile butadiene styrene (ABS) and polylactic acid (PLA), and a quadratic test object fabricated from PLA. The dark-field projection images show orientation-dependent X-ray scattering which is due to anisotropic voids and gaps at the filament-to-filament interface in these fused deposition modeling additive manufacturing objects. SEM corroborates the existence of gaps between filaments.The absorption and dark-field volumes are used to correlate printhead trajectory with print defect density. The absorption volume is used to generate perimeter points slice-by-slice, and from these points, the 2D curvature is calculated. There is a slight increase in X-ray scattering, hence print defect density, at regions with high curvature.Two X-ray interferometry techniques were used: stepped-grating and single-shot. As currently developed, stepped-grating has the larger field-of-view—examination of an entire test object—whilst single-shot has the potential for real-time, in situ measurement of the printing process within 1 mm of the printhead.

Assessment of Accuracy in the Identification of Rock Formations from Aerial and Terrestrial Laser-Scanning Data

Rock formations are among the most spectacular landscape features both for experts and the public. However, information about these objects is often stored inaccurately in existing spatial databases, their corresponding elevations are missing, or the entire rock object is completely absent. Cartographic depiction is also reduced to a point of areal symbology of a largely generalized character. This paper discusses options in identifying and analyzing rock formations from two digital terrain models (DTMs), DMR 5G and DMR 5G+, and irregularly spaced points of airborne laser-scanning (ALS) data with different point densities. A semi-automatic method allowing rock formations to be identified from DTMs is introduced at the beginning of the paper. A method to evaluate elevation models (volume differences) is subsequently applied and a 3D model of a selected rock object is created from terrestrial laser-scanning data. Finally, positional and volumetric comparisons of that 3D object are performed in 2D, 2.5D, and 3D. The results of the pilot study confirmed that the digital terrain models studied are a reliable source in identifying and updating rock formations using the semi-automatic method introduced. The results show that DMR 5G model quality decreases with increasing fragmentation and relative rock formation height, while the proportion of gross errors increases. The complementary DMR 5G+ is better in terms of location and altitude.

Video Foreground Cosegmentation Based on Common Fate

Extracting an object of interest from a single video still faces significant difficulties when the object has variegated appearance, manifests articulated motion, or experiences occlusions by other objects. In this paper, we present a video cosegmentation method to address the aforementioned challenges. Departing from the objectness attributes and motion coherence used by traditional foreground–background separation and video segmentation methods, we place central importance in the role of “common fate.” Specifically, the different parts of the object should persist together in all the videos despite the possible presence of incoherent (e.g., articulated) motions. To accomplish this idea, we first extract seed superpixels by a motion-based foreground segmentation method. We next formulate a set of initial to-link constraints between these superpixels based on whether they exhibit the characteristics of common fate. An iterative manifold ranking algorithm is then proposed to trim away the incorrect and accidental linkage relationships. Having discovered the parts that should cohere together, we next perform clustering to extract the entire object and to handle the case in which there might be multiple objects present. This clustering is performed at two levels: the superpixel level and the object level. This two-level clustering algorithm also performs automatic model selection to estimate the number of object classes extracted. Finally, a multiclass labeling Markov random field is used to obtain a refined segmentation result. To evaluate the performance of our framework, we introduce a new data set in which the videos have complex form and motion that are liable to ambiguity in interpretation. Our experimental results on this data set show that our method successfully addresses the challenges in the extraction of complex objects and outperforms the state-of-the-art video segmentation and cosegmentation methods in our data set.

A framework for fast automatic image cropping based on deep saliency map detection and gaussian filter

ABSTRACTExtracting robust visual saliency map and image cropping are fundamental problems in computer vision, graphics, and so on. It is not easy task to accurately detect and crop the entire salient object from images with complex background. In this paper, a deep learning strategy is adopted to train a large data-set of images, to get saliency map from the input image using graph-based segmentation and gray level adjustment to enhance and extract more accurate and clear saliency map. Furthermore, the Gaussian filter and image scaling used along with cropping method to keep better presentation of the visual object. The important task of overall framework is to take care about relevant image contents as well as to identify more region of interest and get optimum rectangle from the saliency map with minimum and maximum rectangular windows. Quality and low computational complexity have been focused while performing the cropping operation because automatic and efficient cropping technique should not only rely on geometric constraints, but it should be fast enough to consider the important image contents. The applied method use different data-set of images, to ensure the efficiency of this technique and the experimental results show that the framework is not only fast as well as much better for image cropping. We used Matlab and Caffe framework for efficient experimental results.

Percutaneous Coronary Intervention versus Coronary Artery Bypass Graft in Acute Coronary Syndrome patients with Renal Dysfunction

ACS patients with renal dysfunction tend to have a poorer prognosis than those with normal renal function. This retrospective cohort study was performed using The Second Drug-Eluting Stent Impact on Revascularization Registry, a retrospective registry, to evaluate the time-dependent relative risk of revascularization strategies in ACS patients with renal dysfunction. The study demonstrated that the short-term MACCE rate was lower after PCI than CABG. However, there was no significant difference in long-term MACCE rate. Subgroup analyses based on the degree of renal dysfunction resulted in similar findings. The revascularization strategy was identified as a time-dependent covariate by the time-dependent Cox model, and the regression coefficient was ‘−1.124 + 0.344 × ln (time + 1)’. For the entire object group and the separate subgroups, PCI was initially associated with a lower hazard for MACCE than CABG after revascularization, then the hazard ratio increases with time. In conclusion, the hazard ratio for MACCE in PCI relative to CABG is time-dependent. PCI tends to have a lower risk for MACCE than CABG in the short-term, then the hazard ratio increases with time.

X-ray diffraction tomography with limited projection information

X-ray diffraction tomography (XDT) records the spatially-resolved X-ray diffraction profile of an extended object. Compared to conventional transmission-based tomography, XDT displays high intrinsic contrast among materials of similar electron density and improves the accuracy in material identification thanks to the molecular structural information carried by diffracted photons. However, due to the weak diffraction signal, a tomographic scan covering the entire object typically requires a synchrotron facility to make the acquisition time more manageable. Imaging applications in medical and industrial settings usually do not require the examination of the entire object. Therefore, a diffraction tomography modality covering only the region of interest (ROI) and subsequent image reconstruction techniques with truncated projections are highly desirable. Here we propose a table-top diffraction tomography system that can resolve the spatially-variant diffraction form factor from internal regions within extended samples. We demonstrate that the interior reconstruction maintains the material contrast while reducing the imaging time by 6 folds. The presented method could accelerate the acquisition of XDT and be applied in portable imaging applications with a reduced radiation dose.

Minijatura iz Smilčića

U članku se raspravlja o dva keramička ulomka koja pripadaju keramičkom predmetu pronađenom prilikom istraživanja neolitičkog nalazišta u Smilčiću. Zahvaljujući njihovu međusobnom fizičkom povezivanju bilo je moguće predložiti rekonstrukciju koja prikazuje pročelje objekta koji posve odgovara predodžbi jedne neolitičke nastambe sa stiliziranim prikazom rogova postavljenim iznad ulaza. Minimalno sačuvana bočna strana na desnom ulomku pročelja vrlo uvjerljivo sugerira zaključak kako nije riječ samo o prikazu pročelja, nego o čitavom modelu objekta koji nalikuje neolitičkoj nastambi. Ipak, autor je mišljenja kako se ne radi o predočavanju neolitičke nastambe, nego neolitičkog svetišta u funkciji žrtvenika.

Collaboration of Multiple Autonomous Industrial Robots through Optimal Base Placements

Multiple autonomous industrial robots can be of great use in manufacturing applications, particularly if the environment is unstructured and custom manufacturing is required. Autonomous robots that are equipped with manipulators can collaborate to carry out manufacturing tasks such as surface preparation by means of grit-blasting, surface coating or spray painting, all of which require complete surface coverage. However, as part of the collaboration process, appropriate base placements relative to the environment and the target object need to be determined by the robots. The problem of finding appropriate base placements is further complicated when the object under consideration is large and has a complex geometric shape, and thus the robots need to operate from a number of base placements in order to obtain complete coverage of the entire object. To address this problem, an approach for Optimization of Multiple Base Placements (OMBP) for each robot is proposed in this paper. The approach aims to optimize base placements for multi-robot collaboration by taking into account task-specific objectives such as makespan, fair workload division amongst the robots, and coverage percentage; and manipulator-related objectives such as torque and manipulability measure. In addition, the constraint of robots maintaining an appropriate distance between each other and relative to the environment is taken into account. Simulated and real-world experiments are carried out to demonstrate the effectiveness of the approach and to verify that the simulated results are accurate and reliable.

The case of the missing visual details: Occlusion and long-term visual memory.

To investigate the critical information in long-term visual memory representations of objects, we used occlusion to emphasize 1 type of information or another. By occluding 1 solid side of the object (e.g., top 50%) or by occluding 50% of the object with stripes (like a picket fence), we emphasized visible information about the object, processing the visible details in the former and the object's overall form in the latter. On a token discrimination test, surprisingly, memory for solid or stripe occluded objects at either encoding (Experiment 1) or test (Experiment 2) was the same. In contrast, when occluded objects matched at encoding and test (Experiment 3) or when the occlusion shifted, revealing the entire object piecemeal (Experiment 4), memory was better for solid compared with stripe occluded objects, indicating that objects are represented differently in long-term visual memory. Critically, we also found that when the task emphasized remembering exactly what was shown, memory performance in the more detailed solid occlusion condition exceeded that in the stripe condition (Experiment 5). However, when the task emphasized the whole object form, memory was better in the stripe condition (Experiment 6) than in the solid condition. We argue that long-term visual memory can represent objects flexibly, and task demands can interact with visual information, allowing the viewer to cope with changing real-world visual environments. (PsycINFO Database Record