Spatial Relationships Of Objects Research Articles

Small Object Detection in Unmanned Aerial Vehicle Images Using Feature Fusion and Scaling-Based Single Shot Detector With Spatial Context Analysis

Objects in unmanned aerial vehicle (UAV) images are generally small due to the high-photography altitude. Although many efforts have been made in object detection, how to accurately and quickly detect small objects is still one of the remaining open challenges. In this paper, we propose a feature fusion and scaling-based single shot detector (FS-SSD) for small object detection in the UAV images. The FS-SSD is an enhancement based on FSSD, a variety of the original single shot multibox detector (SSD). We add an extra scaling branch of the deconvolution module with an average pooling operation to form a feature pyramid. The original feature fusion branch is adjusted to be better suited to the small object detection task. The two feature pyramids generated by the deconvolution module and feature fusion module are utilized to make predictions together. In addition to the deep features learned by the FS-SSD, to further improve the detection accuracy, spatial context analysis is proposed to incorporate the object spatial relationships into object redetection. The interclass and intraclass distances between different object instances are computed as a spatial context, which proves effective for multiclass small object detection. Six experiments are conducted on the PASCAL VOC dataset and the two UAV image datasets. The experimental results demonstrate that the proposed method can achieve a comparable detection speed but an accuracy superior to those of the six state-of-the-art methods.

Semantic and Spatial Co-Occurrence Analysis on Object Pairs for Urban Scene Classification

Urban scene classifications with very-high-resolution (VHR) satellite images play an important role in urban functional-zone investigation and landscape analysis. Most scene classification techniques utilize visual features which however are weak in identifying scenes, as scenes are usually composed of diverse objects with variant visual cues. To resolve this issue, we use spatial object relations to characterize urban scenes and recognize their categories. Accordingly, two questions are considered in this study: how to measure spatial object relations and how to use these relations to classify urban scenes? First, we propose a novel scene feature, semantic and spatial co-occurrence probability (SSCP), to measure spatial relations between objects with considering their directions, distances, and semantics, which can hopefully resolve the three key issues in measuring spatial object relations, i.e., anisotropy, scale- and semantic-dependencies. Then, a semilatent Dirichlet allocation is employed to classify scenes based on the proposed SSCP. In experiments, our method is first verified based on the UC Merced data set, and then used to generate an urban functional-zone map for Beijing, China. Both experiments indicate high accuracies and state-of-the-art performances of our method. In addition, we give a deep insight into SSCP's characteristics including convergence, high efficiency, and invariance to affine transformation, which can make our method more applicable.

Uso De Los Sensores Remotos En Mediciones Forestales

To understand the use of remote sensors in forestry measurements, some of the most relevant definitions have been taken from a wide variety that currently exist, therefore, we will say that these generally play a predominant role in the Dasometry and that all The forest resource requires qualitative and quantitative information regarding the state of the forest and its evolution over time, with sampling that follows one of the existing methods. Historically, modern discipline arises with the invention of flight and the first photographs are obtained from a balloon in the years 1858 and 1859, in 1909 and on board the airplane the first photographic observation is acquired. The first aerial camera was developed in 1915 by J.T.C. Moore, starting the interesting way forward in the techniques of aerial photography using modified combat aircraft giving way to what was called systematic aerial photography in the late 50s. The development on a global scale of the first artificial satellites, allowed remote sensing in the middle of the 20th century on board the first satellite placed in orbit is that of the TYROS series in 1960 by NASA, becoming the pioneer in missions of meteorological observation, which also led to the appearance of satellite image processing, using mathematical procedures such as the Fourier transform. During the 70s missions were promoted with the objective of exploring the solar system and the moon; giving rise to the first spatial photographs taken by Alan B. Shepard rudimentary, Subsequently, Apollo-9 was used for the first multi-spectral experiment in which 4 Hasselblad cameras with different filters were installed. On July 23, 1972, the Landsat project appeared with the launch of the first satellite of the ERTS series (Earth Resources Technology Satellite). Google Earth in the 21st century, introduces online web services making remote sensing accessible to all audiences, with many techniques and processes that allow an image of the earth's surface to be obtained remotely captured by remote sensors located on satellites or airplanes that gather the spectral and spatial relations of objects. Interferometric radar synthetic aperture They are used to producing accurate digital models of large areas of land. LiDAR(An acronym for the English Light Detection and Ranging) is a monochrome active sensor, its mode of operation consists of measuring the distance between the sensor and the target. It is less expensive compared to manual inventory is multi-purpose, allows a complete survey of the study area, is more efficient than photogrammetry. The multi-spectral acquisition is based on the collection and analysis of areas or objects that emit or reflect radiation at a higher level than nearby objects. The quality of the information collected remotely, once the correction of errors through georeferencing with the help of specialized programs, will depend on their resolutions: spatial, spectral, radiometric and temporal.

Channel bar feature extraction for a mining-contaminated river using high-spatial multispectral remote-sensing imagery

Mapping and monitoring changes of geomorphological features over time are important for understanding fluvial process and effects of its controlling factors. Using high spatial resolution multispectral images has become common practice in the mapping as these images become widely available. Traditional pixel-based classification relies on statistical characteristics of single pixels and performs poorly in detailed mapping using high resolution multispectral images. In this work, we developed a hybrid method that detects and maps channel bars, one of the most important geomorphological features, from high resolution multispectral aerial imagery. This study focuses on the Big River which drains the Ozarks Plateaus region in southeast Missouri and the Old Lead Belt Mining District which was one of the largest producers of lead worldwide in the early and middle 1900s. Mapping and monitoring channel bars in the Big River is essential for evaluating the fate of contaminated mining sediment released to the Big River. The dataset in this study is 1 m spatial resolution and is composed of four bands: Red (Band 3), Green (Band 2), Blue (Band 1) and Near-Infrared (Band 4). The proposed hybrid method takes into account both spectral and spatial characteristics of single pixels, those of their surrounding contextual pixels and spatial relationships of objects. We evaluated its performance by comparing it with two traditional pixel-based classifications including Maximum Likelihood (MLC) and Support Vector Machine (SVM). The findings indicate that derived characteristics from segmentation and human knowledge can highly improve the accuracy of extraction and our proposed method was successful in extracting channel bars from high spatial resolution images.

Using manifold structure for automatic image annotation by fusion of multiple feature spaces

Automatic image annotation has been an active research topic in recent years. Low level features like as color, texture, shape as well as object spatial relations are extracted to represent images in general. These syntaxes are further used to retrieve images from large image data sets. However, the similarity of images could not be found correctly by similarity measures such as Euclidean distance in many situations. On the other hand, graph models have been shown powerful in solving many machine learning problems in recent years. In this paper, we propose a graph-based learning approach, named Conceptual Manifold Structure (CMS), based on transition from conceptual to observation space. In the proposed method, a graph including both the trained and tested samples is constructed by fusion of multiple feature spaces. Conceptual transition in graph structure is found by altering the edge values in an innovative manner. This is caused to learn the manifold structure where the samples dissimilarity is closer to the conceptual distance. Furthermore, the continuity between the instances of a semantic in the conceptual space is kept in feature space. Keeping the continuity in manifold structure is the main idea to decrease the semantic gap in this study. The experiments on different image data sets indicated that the geometrical distances between the samples on the manifold space are closer to their conceptual distance. The proposed method has been compared to other well-known approaches. The results confirmed the effectiveness and validity of the proposed method.

Expertise and the representation of space.

Expertise and the representation of space.

Mental Spatial Transformations in 14‐ and 16‐Month‐Old Infants: Effects of Action and Observational Experience

Infants' ability to mentally track the orientation of an object during a hidden rotation was investigated (N=28 in each experiment). A toy on a turntable was fully covered and then rotated 90°. When revealed, the toy had turned with the turntable (probable event), remained at its starting orientation (improbable event in Experiment 1), or turned to the opposite side (improbable event in Experiment 2). Results demonstrated a developmental progression between 14 and 16months of age in infants' sensitivity to spatial object relations and their ability to track the orientation of an object during hidden rotation. Experiment 3 showed that 14-month-olds' performance improved with hands-on training, highlighting the role of action experience in cognitive development.

Methodical Spatial Database Design with Topological Polygon Structures

Spatial databases form the foundation for a Spatial Data Infrastructure (SDI). For this, a spatial database should be methodically developed to accommodate its role in SDI. It is desirable to have an approach to spatial database development that considers maintenance from the early stage of database design and in a flexible way. Moreover, there is a lack of a mechanism to capture topological relations of spatial objects during the design process. This paper presents an approach that integrates maintenance of topological integrity constraints into the whole spatial database development cycle. The approach is based on the concept of Abstract Data Types. A number of topological classes have been identified and modelling primitives developed for them. Topological integrity constraints are embedded into maintenance functions associated with the topological classes. A semi-automatic transformation process has been developed following the principles of Model Driven Architecture to simplify the design process.

Simultaneous Localization and Mapping with Identification of Landmarks Based on Monocular Vision

How to identify objects is a hot issue of robot simultaneous localization and mapping (SLAM) with monocular vision. In this paper, an algorithm of wheeled robot’s simultaneous localization and mapping with identification of landmarks based on monocular vision is proposed. In observation steps, identifying landmarks and locating position are performed by image processing and analyzing, which converts vision image projection of wheeled robots and geometrical relations of spatial objects into calculating robots’ relative landmarks distance and angle. The integral algorithm procedure follows the recursive order of prediction, observation, data association, update, mapping to have simultaneous localization and map building. Compared with Active Vision algorithm, Three dimensional vision and stereo vision algorithm, the proposed algorithm is able to identify environmental objects and conduct smooth movement as well.

Kea, Nestor notabilis, produce dynamic relationships between objects in a second-order tool use task

Studies on advanced forms of tool use in birds have mainly been concentrated on corvids. In this study, captive kea, a neophilic New Zealand parrot species, produced different orders of spatial object relationships in a tube-lifting/object-inserting paradigm. Hence, we found that kea, which are neither natural tool users nor nest builders, could readily solve a second-order tool use task. They also learned to produce highly complicated means–means–end sequences in a short period of time.

Inability to perceive the spatial relationship of objects revolving too quickly to attentively track

Inability to perceive the spatial relationship of objects revolving too quickly to attentively track

Non-Connective Linear Cartograms for Mapping Traffic Conditions

Cartograms have the advantage of bringing a greater visual impact to map readers. Geographic locations or spatial relationships of objects are intentionally modified to suit the attributes pertaining to objects. In area cartograms, it is the size of the object that is intentionally modified, while in linear cartograms it is the length or direction that is intentionally modified. Traffic conditions in urban transportation networks are very dynamic phenomenon as they change through time. During highly congested hours, travel speeds are low, and travel times are long, and vice versa. In previous studies, traffic conditions were visualized by color and width of road segments. In this paper, non-connective linear cartograms are introduced as a way to represent traffic conditions. Non-connective linear cartograms are linear cartograms that do not show the connectivity between line segments. Lengths of road segments are modified to represent a specific theme in traffic conditions. When the length of road segments represents the congestion level, longer segments indicate higher congestion levels, meaning near road maximum capacity. When the length of the segments represents the travel speed, longer segments indicate higher travel speed and, therefore, shorter travel time. When the length of the segments represents the travel time, longer segments indicate longer travel time, and therefore lower travel speed. In the non-connective linear cartograms, lengths of line segments are not limited to the physical length of represented road segments. The flexibility of adjusting it makes length of line segment a visual variable just like color and width of line segment. All three visual variables work together to create dramatic visual effects and attract greater attention from readers.

Multi‐Perspective Modelling, Rendering and Imaging

Abstract A perspective image represents the spatial relationships of objects in a scene as they appear from a single viewpoint. In contrast, a multi‐perspective image combines what is seen from several viewpoints into a single image. Despite their incongruity of view, effective multi‐perspective images are able to preserve spatial coherence and can depict, within a single context, details of a scene that are simultaneously inaccessible from a single view, yet easily interpretable by a viewer. In computer vision, multi‐perspective images have been used for analysing structure revealed via motion and generating panoramic images with a wide field‐of‐view using mirrors. In this STAR, we provide a practical guide on topics in multi‐perspective modelling and rendering methods and multi‐perspective imaging systems. We start with a brief review of multi‐perspective image techniques frequently employed by artists such as the visual paradoxes of Escher, the Cubism of Picasso and Braque and multi‐perspective panoramas in cel animations. We then characterize existing multi‐perspective camera models, with an emphasis on their underlying geometry and image properties. We demonstrate how to use these camera models for creating specific multi‐perspective rendering effects. Furthermore, we show that many of these cameras satisfy the multi‐perspective stereo constraints and we demonstrate several multi‐perspective imaging systems for extracting 3D geometry for computer vision. The participants learn about topics in multi‐perspective modelling and rendering for generating compelling pictures for computer graphics and in multi‐perspective imaging for extracting 3D geometry for computer vision. We hope to provide enough fundamentals to satisfy the technical specialist without intimidating curious digital artists interested in multi‐perspective images. The intended audience includes digital artists, photographers and computer graphics and computer vision researchers using or building multi‐perspective cameras. They will learn about multi‐perspective modelling and rendering, along with many real world multi‐perspective imaging systems.

Quantitative fuzzy topological relations of spatial objects by induced fuzzy topology

This paper presents a study on the modeling of fuzzy topological relations between uncertain objects in Geographic Information Systems (GIS). Based on the recently developed concept of computational fuzzy topological space, topological relations between simple fuzzy spatial objects are modeled. The fuzzy spatial objects here cover simple fuzzy region, simple fuzzy line segment and fuzzy point. To compute the topological relations between the simple spatial objects, intersection concepts and integration methods are applied and a computational 9-intersection model are proposed and developed. There are different types of intersection, and we have proposed different integration methods for computation in different cases. For example, surface integration method is applied to the case of the fuzzy region-to-fuzzy region relation, while the line integration method is used in the case of fuzzy line segment-to-fuzzy line segment relation. Moreover, this study has discovered that there are (a) sixteen topological relations between simple fuzzy region to line segment; (b) forty-six topological relations between simple fuzzy line segments; (c) three topological relations between simple fuzzy region to fuzzy point; and (d) three topological relations between simple fuzzy line segment to fuzzy point.

Spatial relationship representation for visual object searching

Image representation has been a key issue in vision research for many years. In order to represent various local image patterns or objects effectively, it is important to study the spatial relationship among these objects, especially for the purpose of searching the specific object among them. Psychological experiments have supported the hypothesis that humans cognize the world using visual context or object spatial relationship. How to efficiently learn and memorize such knowledge is a key issue that should be studied. This paper proposes a new type of neural network for learning and memorizing object spatial relationship by means of sparse coding. A group of comparison experiments for visual object searching between several sparse features are carried out to examine the proposed approach. The efficiency of sparse coding of the spatial relationship is analyzed and discussed. Theoretical and experimental results indicate that the newly developed neural network can well learn and memorize object spatial relationship and simultaneously the visual context learning and memorizing have certainly become a grand challenge in simulating the human vision system.

Directional Disorientation Following Left Retrosplenial Hemorrhage: a Case Report with FMRI Studies

We report a 55 year-old right-handed man who presented with topographical disorientation following left retrosplenial hemorrhage. His directional information about familiar places, encoded by previous navigation, was severely impaired, and he could not learn the direction to new places in large-scale spaces beyond the range of visual surveillance. By contrast, he had no difficulties with directional information encoded in a tabletop manner: he could locate major cities or countries on a map, and he also could memorize the spatial relationship of objects in a room. Six months after the ictus, when he had recovered from his directional disorientation, a functional magnetic resonance imaging (fMRI) study of mental navigation demonstrated prominent activation in the retrosplenial area along the right parieto-occipital sulcus and the circumference of the injured area on the left side. The present study, together with previous investigations including clinical case reports, functional neuroimaging, and anatomical and physiological studies on monkeys, suggests that the ‘sense of direction’ in a large-scale locomotor environment is subserved by the visual area along the parieto-occipital sulcus, and that bilateral deterioration of this function causes directional disorientation.

Computing the fuzzy topological relations of spatial objects based on induced fuzzy topology

For modeling the topological relations between spatial objects, the concepts of a bound on the intersection of the boundary and interior, and the boundary and exterior are defined in this paper based on the newly developed computational fuzzy topology. Furthermore, the qualitative measures for the intersections are specified based on the α‐cut induced fuzzy topology, which are (Aα∧∂A)(x)<1−α and ((Ac)α∧∂A)(x)<1−α. In other words, the intersection of the interior and boundary or boundary and exterior are always bounded by 1−α, where α is a value of a level cutting. Specifically, the following areas are covered: (a) the homeomorphic invariants of the fuzzy topology; (b) a definition of the connectivity of the newly developed fuzzy topology; (c) a model of the fuzzy topological relations between simple fuzzy regions in GIS; and (d) the quantitative values of topological relations can be calculated.

Combining estimators for Monte Carlo volume rendering with shading

Monte Carlo volume rendering (MCVR), which allows for generating X-ray like images, offers significant advantages over existing volume rendering techniques in terms of time and memory complexity. The classical shading for MCVR, however, results in a low quality image. In this paper, a so-called estimators combining technique is presented that estimates a realistic rendering using the Lambertian reflection illumination model. Such a technique is particularly adapted to changes in viewing and lighting, and provides high quality shading images exhibiting strong shape and spatial relationship of objects.

Brain SPECT Findings in a Patient With Alice in Wonderland Syndrome

Alice in Wonderland syndrome (AIWS) refers to distortions in body images and in the apparent size, shape, and spatial relations of objects seen. The authors present a 7-year-old girl diagnosed with AIWS. Fifteen days after she had had an upper respiratory tract infection and tonsillitis, she began to complain of several episodes in which the objects and the people around her seemed too small or too big. Brain SPECT with Tc-99m HMPAO performed during the acute stage of AIWS showed hypoperfusion in the right frontal and the right frontoparietal regions.

Knowledge-assisted semantic video object detection

An approach to knowledge-assisted semantic video object detection based on a multimedia ontology infrastructure is presented. Semantic concepts in the context of the examined domain are defined in an ontology, enriched with qualitative attributes (e.g., color homogeneity), low-level features (e.g., color model components distribution), object spatial relations, and multimedia processing methods (e.g., color clustering). Semantic Web technologies are used for knowledge representation in the RDF(S) metadata standard. Rules in F-logic are defined to describe how tools for multimedia analysis should be applied, depending on concept attributes and low-level features, for the detection of video objects corresponding to the semantic concepts defined in the ontology. This supports flexible and managed execution of various application and domain independent multimedia analysis tasks. Furthermore, this semantic analysis approach can be used in semantic annotation and transcoding systems, which take into consideration the users environment including preferences, devices used, available network bandwidth and content identity. The proposed approach was tested for the detection of semantic objects on video data of three different domains.