Objects In Complex Environments Research Articles

The spatiotemporal neural dynamics of object location representations in the human brain

To interact with objects in complex environments, we must know what they are and where they are in spite of challenging viewing conditions. Here, we investigated where, how and when representations of object location and category emerge in the human brain when objects appear on cluttered natural scene images using a combination of functional magnetic resonance imaging, electroencephalography and computational models. We found location representations to emerge along the ventral visual stream towards lateral occipital complex, mirrored by gradual emergence in deep neural networks. Time-resolved analysis suggested that computing object location representations involves recurrent processing in high-level visual cortex. Object category representations also emerged gradually along the ventral visual stream, with evidence for recurrent computations. These results resolve the spatiotemporal dynamics of the ventral visual stream that give rise to representations of where and what objects are present in a scene under challenging viewing conditions.

Досвід створення та впровадження інтелектуалізованих систем керування динамічними об’єктами

Introduction. An important area of application of advanced information technologies is intelligent control of dynamic objects in complex environment. The purpose of the article is to analyze and generalize theoretical as well as practical results related to elaboration of systems for intelligent control of dynamic objects. Future research prospects in this area are also outlined. Methods. Methodological tools include theories of intelligent control, decision making, artificial intelligence, computer vision, system analysis, acoustics, problem solving, conflict resolution. Results. The article outlines main results related to intelligent control of dynamic objects in complex and changing environment. Issues of system integrity, distinctions between traditional and intelligent control, usage of artificial intelligence in autonomous systems, data structuring, image comprehension of current situation have been discussed. Information about the implementation of high-precision control systems for various kinds of dynamic objects (sea vessels, aircrafts, UAVs) in the conditions of conflicts as well as essential informational restrictions has been presented. Though the article covers a wide spectrum of research activities with a considerable period, the authors tried to stress both interconnections and succession of scientific as well as technological achievements in the examined area. Conclusions. Main directions of future research include development and implementation of new approaches to the increase of intellectualization level for information technologies and systems.

Joint Learning of Salient Object Detection, Depth Estimation and Contour Extraction

Benefiting from color independence, illumination invariance and location discrimination attributed by the depth map, it can provide important supplemental information for extracting salient objects in complex environments. However, high-quality depth sensors are expensive and can not be widely applied. While general depth sensors produce the noisy and sparse depth information, which brings the depth-based networks with irreversible interference. In this paper, we propose a novel multi-task and multi-modal filtered transformer (MMFT) network for RGB-D salient object detection (SOD). Specifically, we unify three complementary tasks: depth estimation, salient object detection and contour estimation. The multi-task mechanism promotes the model to learn the task-aware features from the auxiliary tasks. In this way, the depth information can be completed and purified. Moreover, we introduce a multi-modal filtered transformer (MFT) module, which equips with three modality-specific filters to generate the transformer-enhanced feature for each modality. The proposed model works in a depth-free style during the testing phase. Experiments show that it not only significantly surpasses the depth-based RGB-D SOD methods on multiple datasets, but also precisely predicts a high-quality depth map and salient contour at the same time. And, the resulted depth map can help existing RGB-D SOD methods obtain significant performance gain.

Salient-VPR: Salient Weighted Global Descriptor for Visual Place Recognition

Visual place recognition (VPR) is a widely investigated but challenging problem, which must deal with the twin problems of changed appearance caused by varying weather conditions, illumination, and seasons, as well as dynamic objects in complex environments. In this paper, we proposed Salient-VPR by combining image retrieval, semantic information, and saliency cues to achieve accurate estimations. We provides a novel formulation for combining the local semantic features into global descriptors. Unlike traditional global descriptors that typically summarize the whole visual content in the images or image regions, we aggregate dense local semantic descriptors based on pixel-level semantic scores to form global semantic descriptors. Then we calculate predicted saliency descriptors to learn the representation of static objects from images. These salient descriptions are learned from a dataset tailored for VPR missions. Lastly, we introduce a late-fusion module, which increases the stability of the descriptor and avoids performance degradation caused by the limitations of semantic and saliency prediction. Our method outperforms traditional global descriptors in experiments, attaining state-of-the-art visual location recognition scores on a variety of challenging datasets such as Pitts30k, MSLS, and Tokyo24/7.

Low-altitude small-sized object detection using lightweight feature-enhanced convolutional neural network

Unauthorized operations referred to as "black flights" of unmanned aerial vehicles (UAVs) pose a significant danger to public safety, and existing low-attitude object detection algorithms encounter difficulties in balancing detection precision and speed. Additionally, their accuracy is insufficient, particularly for small objects in complex environments. To solve these problems, we propose a lightweight feature-enhanced convolutional neural network able to perform detection with high precision detection for low-attitude flying objects in real time to provide guidance information to suppress black-flying UAVs. The proposed network consists of three modules. A lightweight and stable feature extraction module is used to reduce the computational load and stably extract more low-level feature, an enhanced feature processing module significantly improves the feature extraction ability of the model, and an accurate detection module integrates low-level and advanced features to improve the multiscale detection accuracy in complex environments, particularly for small objects. The proposed method achieves a detection speed of 147 frames per second (FPS) and a mean average precision (mAP) of 90.97% for a dataset composed of flying objects, indicating its potential for low-altitude object detection. Furthermore, evaluation results based on microsoft common objects in context (MS COCO) indicate that the proposed method is also applicable to object detection in general.

Geometrical Assessment of Sunlit and Shaded Area of Urban Trees Based on Aligned Orthographic Views

To quantify the ecosystem services of trees in urban environments, it is necessary to assess received direct solar radiation of each tree. While the Sky View Factor (SVF) is suitable for assessing the total incoming short- and longwave radiation fluxes, its information is limited to specific points in space. For a spatial analysis, it is necessary to sample the area for SVF. A new geometrical method, Area View Factor (AVF), for the calculation of sunlit areas is proposed. AVF is the ratio of the unhidden, projected surface of an object to the whole projected surface of an object in a complex environment. Hereby, a virtual, orthographic camera is oriented in accordance to the sun’s position in the 3D model domain. The method is implemented in the microscale model SkyHelios, utilizing efficient rendering techniques to assess AVF of all urban trees in parallel. The method was applied to Rieselfeld in Freiburg, Germany. The assessed sunlit area is compared to the SVF at the top of each tree and solar altitude angle, revealing a strong relationship between sunlit areas to solar altitude angles. This study shows that AVF is an efficient methodology to assess received direct radiation of urban trees. Based on AVF, it is possible to identify urban areas with shaded and sunlit trees, but it can also be applied to other objects in complex environments. Therefore, AVF is applicable for urban architecture or energetic research questions.

Reconfigurable magnetic soft robots with multimodal locomotion

Magnetic soft robots have recently attracted increasing attention owing to their remarkable advantages and potential applications, but the reconfigurable magnetization inside soft-composite materials remains challenging, otherwise they can only achieve fixed magnetic response under external magnetic fields with limited morphological features and motion modes. To tackle this issue, a direct magnetization method based on pulsed high magnetic field focusing is developed, where the local magnetization distribution can be programmed flexibly and reconfigurably after fabrication within milliseconds and millimeter-scale resolution. Based on this method, reconfigurable magnetic soft robots are developed with superior merits, including a fishing soft robot to resist the dynamic drag, an inchworm-inspired crawling robot with a speed more than 1 body/s, and a six-arm rolling soft robot to transport objects in complex environments. The underlying locomotion mechanisms of these soft robots are also well discussed. The developed magnetic soft robots, together with the proposed magnetization method, are expected to pave avenues for the future development of mass production and controllable magnetization of soft robots for practical applications.

Improved simultaneous localization and mapping algorithm combined with semantic segmentation model

In the past decades, emerging technologies such as unmanned driving and indoor navigation have developed rapidly, and simultaneous localization and mapping has played unparalleled roles as core technologies. However, dynamic objects in complex environments will affect the positioning accuracy. In order to reduce the influence of dynamic objects, this article proposes an improved simultaneous localization and mapping algorithm combined with semantic segmentation model. First, in the pre-processing stage, in order to reduce the influence of dynamic features, fully convolutional network model is used to find the dynamic object, and then the output image is masked and fused to obtain the final image without dynamic object features. Second, in the feature-processing stage, three parts are improved to reduce the computing complexity, which are extracting, matching, and eliminating mismatching feature points. Experiments show that the absolute trajectory accuracy in high dynamic scene is improved by 48.58% on average. Meanwhile, the average processing time is also reduced by 21.84%.

High-Content Analysis Monitoring Intracellular Trafficking and Replication of Mycobacterium tuberculosis Inside Host Cells.

Mycobacterium tuberculosis is able to colonize, persist, and massively replicate in host cells, such as phagocytes and epithelial cells. The intracellular stage of the bacteria is critical to the development of tuberculosis pathogenesis. The detailed mechanisms of intracellular trafficking of the bacillus are not fully understood and require further investigations. Therefore, increasing the knowledge of this process will help to develop therapeutic tools that will lower the burden of tuberculosis. M. tuberculosis is genetically tractable and tolerates the expression of heterologous fluorescent proteins. Thus, the intracellular distribution of the bacteria expressing fluorescent tracers can be easily defined using confocal microscopy. Advances in imaging techniques and images-based analysis allow the rapid quantification of biological objects in complex environments. In this chapter, we detailed high-content / high-throughput imaging methods to track the bacillus within host cell settings.

Recent Advances in Design and Actuation of Continuum Robots for Medical Applications

Traditional rigid robot application in the medical field is limited due to the limited degrees of freedom caused by their material and structure. Inspired by trunk, tentacles, and snakes, continuum robot (CR) could traverse confined space, manipulate objects in complex environment, and conform to curvilinear paths in space. The continuum robot has broad prospect in surgery due to its high dexterity, which can reach circuitous areas of the body and perform precision surgery. Recently, many efforts have been done by researchers to improve the design and actuation methods of continuum robots. Several continuum robots have been applied in clinic surgical interventions and demonstrated superiorities to conventional rigid-link robots. In this paper, we provide an overview of the current development of continuum robots, including the design principles, actuation methods, application prospect, limitations, and challenge. And we also provide perspective for the future development. We hope that with the development of material science, Engineering ethics, and manufacture technology, new methods can be applied to manufacture continuum robots for specific surgical procedures.

Diffusion of Anisotropic Colloids in Periodic Arrays of Obstacles

Colloidal suspensions in confined geometries exhibit rich diffusion dynamics governed by particle shapes and particle-confinement interactions. Here, we propose a colloidal system, consisting of ellipsoids in periodic array of obstacles, to investigate the confined diffusion of anisotropic colloids. From the obstacle density-dependent diffusion, we discover a decoupling of translational and rotational diffusion in which only rotational motion is localized while translational motion remains diffusive. Moreover, by evaluating the probability distributions of displacements, we found Brownian but non-Gaussian diffusion behaviors with increasing the obstacle densities, which originates from the shape anisotropy of the colloid and the multiplicity of the local configurations of the ellipsoids with respect to the obstacle. Our results suggest that the shape anisotropy and spatial confinements play a vital role in the diffusion dynamics. It is important for understanding the transportations of anisotropic objects in complex environments.

Fuzzy Detection Aided Real-Time and Robust Visual Tracking Under Complex Environments

Today, a new generation of artificial intelligence has brought several new research domains such as computer vision (CV). Thus, target tracking, the base of CV, has been a hotspot research domain. Correlation filter (CF)-based algorithm has been the basis of real-time tracking algorithms because of the high tracking efficiency. However, CF-based algorithms usually failed to track objects in complex environments. Therefore, this article proposes a fuzzy detection strategy to prejudge the tracking result. If the prejudge process determines that the tracking result is not good enough in the current frame, the stored target template is used for following tracking to avoid the template pollution. During testing on the OTB100 dataset, the experimental results show that the proposed auxiliary detection strategy improves the tracking robustness under complex environment by ensuring the tracking speed.

Experimental study of diffusion behaviors of an ellipsoidal colloid in spherical colloid systems

The diffusive transport in complex confined media is ubiquitous such as diffusions of micro- or nano-particles in glassy liquids and polymer solutions, protein diffusions under crowded conditions, and deliveries of drugs in the biological media. Therefore, the understanding of the diffusive transport arouses the great interest of researchers in the physics, materials science, and biology circles. Despite the fact that the shape of the colloidal particles acts as one of the important physical factors influencing their dynamic behaviors, the study of the anisotropic particles diffusing in confined media is still lacking. In this work, we propose a simple experimental model to investigate the confined diffusion of shape-anisotropic particles. The diffusion of an ellipsoid at different area fractions (<i>ϕ</i>) of colloidal spheres is investigated through video microscopy. At low <i>ϕ</i>, ellipsoid exhibits a random trajectory and free diffusion in translational and rotational degree of freedom; while at high <i>ϕ</i>, the trajectory is in a small spatial range with a nearly constant orientation of the particle, indicating that the arrested diffusion takes place in translational and rotational degree of freedom. The translational and rotational mean square displacement decrease with the increase of <i>ϕ</i>. By power-law fitting (~<i>t<sup>β</sup></i>), it is found that <i>β</i> decreases from 1 to a small value at high <i>ϕ</i>, demonstrating that the ellipsoid experiences a transition from normal diffusion to sub-diffusion. Moreover, <i>β</i> for rotational motion decreases faster than that for translational motion at high <i>ϕ</i>, which signifies that the the rotational motion decouples from the translational motion with increasing <i>ϕ</i>. The results from the van Hove correlation function show that the translational displacement along the major axis of the ellipsoid is always larger than that along the minor axis, manifesting the ellipsoid prefers to diffuse along its major axis independent of <i>ϕ</i>. Significant non-Gaussian tail is observed in the distribution of the translational displacement along the major axis with increasing <i>ϕ</i>. However, the distribution of the translational displacement along the minor axis presents a nearly Gaussian behavior independent of <i>ϕ</i>. This indicates that the translational motion along the major axis decouples from the translational motion along the minor with increasing <i>ϕ</i>. For the rotational displacement, the non-Gaussian tail is only observed at the intermediate <i>ϕ</i>. These non-Gaussian behaviors are confirmed by calculating the non-Gaussian parameter (<i>α</i><sub>2</sub>). Our experiments demonstrate that the confinements give rise to the anomalous diffusion behaviors of the anisotropic colloids, which is conducive to the understanding of transportations of anisotropic objects in complex environments.

A Robust Visual Tracking Method for Unmanned Mobile Systems

This paper introduces a robust visual tracking of objects in complex environments with blocking obstacles and light reflection noises. This visual tracking method utilizes a transfer matrix to project image pixels back to real-world coordinates. During the image process, a color and shape test is used to recognize the object and a vector is used to represent the object, which contains the information of orientation and body length of the object. If the object is partially blocked by the obstacles or the reflection from the water surface, the vector predicts the position of the object. During the real-time tracking, a Kalman filter is used to optimize the result. To validate the method, the visual tracking algorithm was tested by tracking a submarine and a fish on the water surface of a water tank, above which three pieces of blur glass were blocking obstacles between the camera and the object. By using this method, the interference from the reflection of the side glass and the fluctuation of the water surface can be also avoided.

Hierarchical visual attention model for saliency detection inspired by avian visual pathways

Visual attention is a mechanism that enables the visual system to detect potentially important objects in complex environment. Most computational visual attention models are designed with inspirations from mammalian visual systems. However, electrophysiological and behavioral evidences indicate that avian species are animals with high visual capability that can process complex information accurately in real time. Therefore, the visual system of the avian species, especially the nuclei related to the visual attention mechanism, are investigated in this paper. Afterwards, a hierarchical visual attention model is proposed for saliency detection. The optic tectum neuron responses are computed and the self-information is used to compute primary saliency maps in the first hierarchy. The &#x201C winner-take-all &#x201D network in the tecto-isthmal projection is simulated and final saliency maps are estimated with the regularized random walks ranking in the second hierarchy. Comparison results verify that the proposed model, which can define the focus of attention accurately, outperforms several state-of-the-art models. This study provides insights into the relationship between the visual attention mechanism and the avian visual pathways. The computational visual attention model may reveal the underlying neural mechanism of the nuclei for biological visual attention.

Salient Object Detection: Integrate Salient Features in the Deep Learning Framework

Salient object detection in complex environments brings the challenge from the collections of large number of training images for deep learning algorithm. It is difficult to collect the enough number of training data for varied salient objects in different scenes, and furthermore the salient objects are usually compared with the background. This paper proposes a novel method to integrate the salient features into the deep learning framework, and design a parallel multi-scale structure of the neural network to enhance the ability to detect salient objects. In addition, a multiple stage method is proposed to optimize the training process and make the datasets more prominent in the commonality of the salient features, which effectively reduces the difficulty of correctly distinguishing salient objects in complex environments. Experiments show that the proposed approach enhances the ability of neural networks to learn specified features and improves the detection effect of salient objects in complex scenes.

Some examples of the application of high-performance computing for time-domain full-wave simulations in underwater acoustics using a spectral-element method

Being able to perform accurate time-domain full-wave simulations in underwater acoustics is important within the context of complex environments. Among the numerical methods that can be used to perform such types of simulations, the spectral-element method is a method of choice because it can take advantage of the possibilities offered by high-performance computing. This capability is very important because accurate numerical simulations require a high computational power. In this presentation we will show some examples of the use a spectral-element method for underwater acoustics applications. 2D simulations as well as 3D simulations will be considered for various configurations, illustrating the interest of performing such simulations. Shallow water propagation, deep water propagation as well as diffraction by objects in complex environments will be analyzed.

Visual tracking via salient feature extraction and sparse collaborative model

Object tracking is always a very attractive research topic in computer vision and image processing. In this paper, an innovative method called salient-sparse-collaborative tracker (SSCT) is put forward, which exploits both object saliency and sparse representation. Within the proposed collaborative appearance model, the object salient feature map is built to create a salient-sparse discriminative model (SSDM) and a salient-sparse generative model (SSGM). In the SSDM module, the presented sparse model effectively distinguishes the target region from its background by using the salient feature map that further helps locate the object in complex environment. In the SSGM module, a sparse representation method with salient feature map is designed to improve the effectiveness of the templates and deal with occlusions. The update scheme takes advantage of salient correction, thus the SSCT algorithm can both handle the appearance variation as well as reduce tracking drifts effectively. Plenty of experiments with quantitative and qualitative comparisons on benchmark reveal the SSCT tracker is more competitive than several popular approaches.

Automatic tracking and positioning algorithm of motion object in complex environment

The problem of large error in automatic tracking and positioning of moving targets in complex environment is discussed. A method of target tracking and positioning based on probability analysis for moving images in complex environment is proposed. The likelihood and prior probability characteristics of moving target under complex background is calculated, and the optimal label of foreground jitter image contour is calculated using probability theory to identify the target region of moving image of the current frame in the complex state, and effectively complete the moving image tracking and marking in complex state. Experimental results show that the proposed method can accurately predict foreground objects in complex images, and improve the stability of target tracking.

Topographical Manipulation of Microparticles and Cells with Acoustic Microstreaming.

Precise and reproducible manipulation of synthetic and biological microscale objects in complex environments is essential for many practical biochip and microfluidic applications. Here, we present an attractive acoustic topographical manipulation (ATM) method to achieve efficient and reproducible manipulation of diverse microscale objects. This new guidance method relies on the acoustically induced localized microstreaming forces generated around microstructures, which are capable of trapping nearby microobjects and manipulating them along a determined trajectory based on local topographic features. This unique phenomenon is investigated by numerical simulations examining the local microstreaming in the presence of microscale boundaries under the standing acoustic wave. This method can be used to manipulate a single microobject around a complex structure as well as collectively manipulate multiple objects moving synchronously along complicated shapes. Furthermore, the ATM can serve for automated maze solving by autonomously manipulating microparticles with diverse geometries and densities, including live cells, through complex maze-like topographical features without external feedback, particle modification, or adjustment of operational parameters.