Bounding Box Method Research Articles

Localization and Clustering Based on Swarm Intelligence in UAV Networks for Emergency Communications

In recent years, unmanned aerial vehicle (UAV) networks have been a focus area of the academic and industrial research community. They have been used in many military and civilian applications. Emergency communication is one of the essential requirements for first responders and victims in the aftermath of natural disasters. In such scenarios, UAVs may configure ad hoc wireless networks to cover a large area. In UAV networks, however, localization and routing are challenging tasks owing to the high mobility, unstable links, dynamic topology, and limited energy of UAVs. Here, we propose swarm-intelligence-based localization (SIL) and clustering schemes in UAV networks for emergency communications. First, we propose a new 3-D SIL algorithm based on particle swarm optimization (PSO) that exploits the particle search space in a limited boundary by using the bounding box method. In the 3-D search space, anchor UAV nodes are randomly distributed and the SIL algorithm measures the distance to existing anchor nodes for estimating the location of the target UAV nodes. Convergence time and localization accuracy are improved with lower computational cost. Second, we propose an energy-efficient swarm-intelligence-based clustering (SIC) algorithm based on PSO, in which the particle fitness function is exploited for intercluster distance, intracluster distance, residual energy, and geographic location. For energy-efficient clustering, cluster heads are selected based on improved particle optimization. The proposed SIC outperforms five typical routing protocols regarding packet delivery ratio, average end-to-end delay, and routing overhead. Moreover, SIC consumes less energy and prolongs network lifetime.

Mathematic formulation for the generation of combined paths for mounting parts in assembly

Today, the assembly process is the least automated phase of production for mechanical products. The percentage of the automated assembly line operations of automobiles fluctuates between 10 and 20% according to manufacturers. The use of industrial robots in the assembly lines played a decisive role in the development of their activity despite the high cost. Indeed, this strategy is a key element that increases productivity. The use of robots in assembly tasks requires the trajectory determination of parts to be assembled from the design phase of the product. This problem is difficult if the part may interfere with other elements and requires combined trajectories to avoid the interference. Therefore, the simulation of part assembling trajectories from the design phase is an interesting and topical research issue. The main objective of this paper is to develop a new approach to generate multidirectional combined trajectories of parts. A mathematical formulation of the proposed approach is firstly proposed, being based on both the position matrix concept and the bounding box method. A didactic example is used to illustrate the different stages of the proposed approach. To highlight the computer implementation of the proposed method, an industrial case study is presented. This example was chosen to illustrate the effectiveness of the proposed approach such as the gain of assembly time, the reduction of costs by robotization of the work, and the programming assistance of these robots.

Simulating unmanned aerial vehicle flight control and collision detection

An unmanned aerial vehicle (UAV) is a small, fast aircraft with many useful features. It is widely used in military reconnaissance, aerial photography, searches, and other fields; it also has very good practical-application and development prospects. Since the UAV’s flight orientation is easily changeable, its orientation and flight path are difficult to control, leading to its high damage rate. Therefore, UAV flight-control technology has become the focus of attention. This study focuses on simulating a UAV’s flight and orientation control, and detecting collisions between a UAV and objects in a complex virtual environment. The proportional-integral-derivative control algorithm is used to control the orientation and position of the UAV in a virtual environment. A version of the bounding-box method that combines a grid with a k-dimensional tree is adopted in this paper, to improve the system performance and accelerate the collision-detection process. This provides a practical method for future studies on UAV flight position and orientation control, collision detection, etc.

License Number Plate Recognition using Template Matching and Bounding Box Method

Recognition of vehicle number plates is very important in the field of transportation security systems. Template matching is one of the most popular techniques used for number plate recognition. This algorithm works by evaluating the image pattern that will be recognized and then comparing it with the image pattern in the template. A number plate is an object in an image that has many characters which usually consists of numbers and letters. Character recognition cannot be done simultaneously in an image. A number plate object in the image needs to be done early processing, normalization, and segmentation to recognize the character. While to separate each character, a bounding box algorithm is used. This technique is relatively simple and has fast computing. So this study examines a combination of template matching and bounding box methods to detect Indonesian car license plates. Based on the tests that have been done, the accuracy reaches 80%.

Research on 3D Simulation of Fabricated Building Structure Based on BIM

In order to solve the problem of low accuracy and poor analysis of traditional fabricated building structure, the new BIM-based 3D simulation method for fabricated building structure is proposed. Drawings and related documents are obtained from the database. Through the REVIT software, the 3D simulation model is drawn by the BIM technical team and the fabricated building structure is simulated in 3D. The design of the REVIT platform is realized by three stages: structural space constraint relationship determination, geometric component generation and color processing. The spatial mesh structure is used to describe the geometric components, and the homomorphic filtering is used to process the colors. The bounding box and the hierarchical bounding box method are used to detect the collision of the steel structure. For static combinations, the bounding box method is used to implement collision detection, and the hierarchical bounding box method is used to detect dynamic combinations. The experimental results show that the proposed method can effectively realize the 3D simulation of the fabricated building structure, and the detection results are accurate, the overall performance is excellent, and the user satisfaction is high.

Minimum elastic bounding box algorithm for dimension detection of 3D objects: a case of airline baggage measurement

Motivated by the interference of appendages in airline baggage dimension detection using three-dimensional (3D) point cloud, a minimum elastic bounding box (MEBB) algorithm for dimension detection of 3D objects is developed. The baggage dimension measurements using traditional bounding box method or shape fitting method can cause large measurements due to the interference of appendages. Starting from the idea of ‘enclosing’, an elastic bounding box model with the deformable surface is established. On the basis of using principal component analysis to obtain the main direction of the bounding box, the elastic rules for deformable surfaces are developed so as to produce a large elastic force when it comes into contact with the main body part and to produce a small elastic force when it comes into contact with the appendages part. The airline baggage measurement shows how to use MEBB for dimension detection, especially for the processing of isotropic density distribution, the elasticity computing and the adaptive adjustment of elasticity. Results on typical baggage samples, comparisons to other methods, and error distribution experiments with different algorithm parameters show that the authors’ method can reliably obtain the size of the main body part of the object under the interference of appendages.

Practically Feasible Optimal Assembly Sequence Planning with Tool Accessibility

Practically Feasible Optimal Assembly Sequence Plan (PFOASP) achievement is most important for an industrial engineer to meet the demand of producing the final product in less time and cost. A PFOASP directs the industrial engineer to generate a complete product with less number of manipulations from individual parts. An assembly sequence is said to be optimal from its nature of performing parallel operations to form stable subassemblies which directly reduces the overall assembly time for handling large scale products. Most the researchers followed to ignore tool accessibility in their approaches to avoid the computational complexities. However, the generated solution lacks completeness in practice because tools play the significant role while assembling operation. PFOASP determination is most challenging due to validation of multiple feasibilities qualifying criterion. In this paper, an attempt is made to explain the complexity and the importance of tool accessibility consideration in PFOASP with suitable illustrations. A tool based bounding box method is proposed to solve geometrical feasibility in PFOASP.

3-D Data Processing to Extract Vehicle Trajectories from Roadside LiDAR Data

High-resolution vehicle data including location, speed, and direction is significant for new transportation systems, such as connected-vehicle applications, micro-level traffic performance evaluation, and adaptive traffic control. This research developed a data processing procedure for detection and tracking of multi-lane multi-vehicle trajectories with a roadside light detection and ranging (LiDAR) sensor. Different from existing methods for vehicle onboard sensing systems, this procedure was developed specifically to extract high-resolution vehicle trajectories from roadside LiDAR sensors. This procedure includes preprocessing of the raw data, statistical outlier removal, a Least Median of Squares based ground estimation method to accurately remove the ground points, vehicle data clouds clustering, a principle component-based oriented bounding box method to estimate the location of the vehicle, and a geometrically-based tracking algorithm. The developed procedure has been applied to a two-way-stop-sign intersection and an arterial road in Reno, Nevada. The data extraction procedure has been validated by comparing tracking results and speeds logged from a testing vehicle through the on-board diagnostics interface. This data processing procedure could be applied to extract high-resolution trajectories of connected and unconnected vehicles for connected-vehicle applications, and the data will be valuable to practices in traffic safety, traffic mobility, and fuel efficiency estimation.

Collision detection during planning for sheet metal bending by bounding volume hierarchy approaches

ABSTRACTCollision detection is a computationally intensive task within process planning for sheet metal bending. An efficient collision detection algorithm can greatly improve the speed of the process planning. In this work, relevant features are extracted first from STL format of the part and tool models. Next, the collision detection strategies for the sheet metal bending problem are investigated considering bounding volume hierarchies involving oriented-bounding box (OBB) and axis-aligned bounding box (AABB) methods. The approaches are explained using two example parts. By analysing the data for 10 different sheet metal parts, it is demonstrated that although OBB hierarchy is more efficient in terms of minimising the number of collision tests between part and tool models, AABB hierarchy is superior in terms of computation time. The collision detection method based on AABB can be integrated with the sheet metal bend planning to realise CAD–CAM integration.

Recognition of Vehicle Number Plate Using Image Processing Technique

The video surveillance system is preferred for security monitoring, however, the detection of moving object is a challenging part. In areas where parking space is allotted to a particular vehicle, a wrongly parked vehicle can be recognized. It is noted that the number plates of the vehicle are found in different shape, size and also they are different in color. This makes the detection of the vehicle by their number plates, the most interesting and challenging research topic. Number plate detection is helpful in finding stolen cars, car parking management system and identification of the vehicle in traffic. This paper presents an approach for Number Plate Recognition (NPR) using morphological operation and Sobel edge detection method. In this approach, all the letters and numbers used in the number plate are segmented by using bounding box method. After segmentation, template matching approach is used to recognize numbers and characters. The number plate decoded can be used further for identification, matching and documentation purpose of vehicle details.

Research on Collision Detection in Virtual Measurement Scene

Facing the need of virtual laboratory of engineering drawing, virtual measurement module is developed. According to the collision features, the collision detection algorithm based on OBB hierarchical bounding box method is applied to the virtual measurement scene. A scanning sampling method is used to simulate the dynamic measurement process. As to the sampling point, the means of dividing the value by two points and limiting the step distance is obtained. The method of determining the traversal order by the distance of the center of the bounding box is proposed in the process of traversing the tree to reduce the average lookup depth of the bounding box tree intersection test. Accurate collision detection between STL model [1] and virtual measurement tool using VS2013 and OpenGL. Through experimental verification, the whole process satisfies the accuracy and real-time requirement of the measurement, and proves the validity and fastness of the collision detection algorithm and its optimization method.

Algorithm of combined reduction based on scattered point cloud

In recent years, the 3D laser scanning technology has been becoming a development hot spot in the survey. Due to the lack of monotonic point cloud data compression method, the paper proposes to use the combination algorithm to streamline point cloud data. Curvature is one of the important geometric features of 3D surface. The point cloud data can be reduced through the curvature of the surface, which means that the point cloud data feature can be better reserved. If the amount of point cloud data is very large, curving surface fitting is slower and it consumes more time. Space distance compression algorithm is simple, and the compression is faster, but it is not very good to retain the features. This paper combined the two algorithms together. Firstly, it uses the minimum distance method to reduce the amount of the point cloud data. Then, based on the space bounding box method, the k-nearest neighbors point cloud data are established. After that, the point cloud is fitted to the quadratic surface in the neighborhood, and the average curvature is calculated. After setting the angle threshold, the point cloud data are reduced based on adjacent normal angle values and threshold. Next, through comparing reconstructed triangulation network and modeling with the original model, it demonstrates that the new method of the compression effect is ideal.

Accuracy of tree diameter estimation from terrestrial laser scanning by circle-fitting methods

This study compares the accuracies of diameter at breast height (DBH) estimations by three initial (minimum bounding box, centroid, and maximum distance) and two refining (Monte Carlo and optimal circle) circle-fitting methods The circle-fitting algorithms were evaluated in multi-scan mode and a simulated single-scan mode on 157 European beech trees (Fagus sylvatica L.). DBH measured by a calliper was used as reference data. Most of the studied circle-fitting algorithms significantly underestimated the mean DBH in both scanning modes. Only the Monte Carlo method in the single-scan mode significantly overestimated the mean DBH. The centroid method proved to be the least suitable and showed significantly different results from the other circle-fitting methods in both scanning modes. In multi-scan mode, the accuracy of the minimum bounding box method was not significantly different from the accuracies of the refining methods The accuracy of the maximum distance method was significantly different from the accuracies of the refining methods in both scanning modes. The accuracy of the Monte Carlo method was significantly different from the accuracy of the optimal circle method in only single-scan mode. The optimal circle method proved to be the most accurate circle-fitting method for DBH estimation from point clouds in both scanning modes.

Collision detection for virtual environment using particle swarm optimization with adaptive cauchy mutation

Rapid and accurate detection of collision between virtual objects is crucial for many virtual reality based applications. In order to ensure a high-level of accuracy and to meet the real-time requirement, a fast collision detection algorithm between soft bodies is proposed. The developed algorithm is a combination of stochastic methods and particle swarm optimization with adaptive Cauchy mutation. The hierarchical bounding box method is used for a rough detection in order to filter out obvious disjoint space, and the problem is converted to a nonlinear optimization problem based on the distance of points. Then particle swarm optimization with adaptive Cauchy mutation is used to find the optimal solution. When it is updated iteratively, keeping some particles experience value and variation of other particles avoids particles trapped in local optimum, which further accelerates the speed of collision detection. The algorithm is implemented and evaluated through experiments and the results confirm the advantages of the developed algorithm.

Automated Method for Optimizing Feasible Locations of Mobile Cranes Based on 3D Visualization

The precise operation of mobile cranes contributes to the safety and efficiency of lift works, among which accurate location is one of the most critical aspects. Considering the complexities and dynamics of construction sites, this paper presents an approach for crane location determination based on the simulation of mobile crane operations. First a data processing method is developed to enable the automated conversion from BIM (Building Information Modeling) and schedule plan to the simulation system. Then the reachability and work radius of mobile cranes are examined for selecting the initial feasible location points. The bounding box method is used for the collision detection to ensure the safety of crane operations. Also, the minimum number of relocation and the lift time are taken into account to determine the final feasible location points of mobile cranes.

Effect of variations in atelectasis on tumor displacement during radiation therapy for locally advanced lung cancer

PurposeAtelectasis (AT), or collapsed lung, is frequently associated with central lung tumors. We investigated the variation of atelectasis volumes during radiation therapy and analyzed the effect of AT volume changes on the reproducibility of the primary tumor (PT) position. Methods and materialsTwelve patients with lung cancer who had AT and 10 patients without AT underwent repeated 4-dimensional fan beam computed tomography (CT) scans during radiation therapy per protocols that were approved by the institutional review board. Interfraction volume changes of AT and PT were correlated with PT displacements relative to bony anatomy using both a bounding box (BB) method and change in center of mass (COM). Linear regression modeling was used to determine whether PT and AT volume changes were independently associated with PT displacement. PT displacement was compared between patients with and without AT. ResultsThe mean initial AT volume on the planning CT was 189 cm3 (37-513 cm3), and the mean PT volume was 93 cm3 (12-176 cm3). During radiation therapy, AT and PT volumes decreased on average 136.7 cm3 (20-369 cm3) for AT and 40 cm3 (−7 to 131 cm3) for PT. Eighty-three percent of patients with AT had at least one unidirectional PT shift that was greater than 0.5 cm outside of the initial BB during treatment. In patients with AT, the maximum PT COM shift was ≥0.5 cm in all patients and >1 cm in 58% of patients (0.5-2.4 cm). Changes in PT and AT volumes were independently associated with PT displacement (P < .01), and the correlation was smaller with COM (R2 = 0.58) compared with the BB method (R2 = 0.80). The median root mean squared PT displacement with the BB method was significantly less for patients without AT (0.45 cm) compared with those with AT (0.8cm, P = .002). ConclusionsChanges in AT and PT volumes during radiation treatment were significantly associated with PT displacements that often exceeded standard setup margins. Repeated 3-dimensional imaging is recommended in patients with AT to evaluate for PT displacements during treatment.

A Novel Denoising and Segmentation of Brain Tumors in MRI Images

Objectives: The objective of this work is to study denoising and segmentation methods to extract brain tumor area from the MRI image, implemented using MATLAB2013b and to examine its performance metrics. Methods/Statistical Analysis: As preprocessing stage is essential for better segmentation as it removes noise that makes images having similar qualities so that tumor area can be shown and extracted with great accuracy. An anisotropic diffusion filter with 8-connected neighborhood is employed for noise removal and Fast Bounding Box (FBB) for exactly showing tumor area on MRI images. Finally Support vector machine classifies the boundary and extracts the tumor from the MRI image. Findings: Brain tumor is the major cause of cancer deaths in human which is due to uncontrollable cells growth in brain portion. Prior detection, diagnosis and accurate healing of brain tumor are the primary work to prevent human death. Image segmentation can also be done in several approaches like thresholding, region growing, watersheds and contours. Specialists with their basic knowledge do manual segmentation, which is time consuming process, where this limitation can be overcome by our fully automatic proposed method. Employing of an isotropic diffusion filter with 8-connected neighborhood compare to 4-connected neighborhood results in considerable improvements in terms of lower identical error rates. Our proposed Fast Bounding Box (FBB) method is applied that exactly shows tumor area on MRI images and its central region is selected judiciously to have sample points required for functionality of one class SVM classifier training. To achieve optimal classification level there is necessity of SVM with optimum efficiency, so that we adapted Support vector machine that immediately stops its operation once all the points are separated. Application/Improvements: Segmented tumor obtained with precision are very useful for radiologists and specialists to had good idea of estimating tumor position and size with great dealt with ease and without any prior information.

Common phase error estimation in coherent optical OFDM systems using best-fit bounding box.

In this paper, we investigate and characterize a new approach of adopting best-fit bounding box method for common phase error estimation in coherent optical OFDM systems. The method is based on the calculation of the 2-D convex hull of the received signal constellation, which is generally adopted in image processing area to correct the skew of images. We further perform detailed characterizations including root mean square error analysis, laser linewidth tolerance, noise tolerance, and computation complexity analysis, via numerical simulations and experiments. The results show the proposed method achieves much improved spectral efficiency and comparable system performance than the pilot-aided method, while it exhibits good estimation accuracy and reduced complexity than the blind phase searching method.

Perancangan dan Pengujian Piranti Pemantauan Visual untuk Menentukan Volume Lalu Lintas

Intelligent Transportation System (ITS) applies information and communication technologies for transportation management. The problem in the study is the importance of computer vision tools to determine the volume of traffic on the highway. The design of computer vision device is performed using a static camera and the bounding box method for determining the traffic volume based on vehicle type, namely light vehicles, heavy vehicles (trucks) and motorcycles that are applied on the Gaussian Mixture Models (GMM). The classification is based on the pixel area of vehicle objects. The tests were carried out at Jalan T. Nyak Arief, Jambotape, Banda Aceh. The test results of the system achieved an accuracy of 79.32%.

Node localization algorithm for wireless sensor networks using compressive sensing theory

In recent years, localization has been recognized as an important supporting technology for wireless sensor networks (WSNs). Along with the increase in WSN indoor applications, indoor localization has become a hot research topic and many localization algorithms have been studied. Among these algorithms, the localization method based on compressive sensing theory emerges as a popular approach to indoor localization. In this approach, the nodes are sparse when compared to the number of grids utilized to represent the locations of the nodes, so the locations are considered as sparse signal and can be reconstructed using the compressive sensing techniques. The localization problem is formulated as the sparse reconstruction of sparsifying matrix which is comprised of measurement of received signal at grids. In order to improve the localization accuracy and meet the real-time requirement of localization applications in large indoor area, an indoor localization algorithm based on dynamic measurement compressive sensing for wireless sensor networks is proposed. Using the bounding-box method, we firstly identify a potential area that possesses the independent features. Instead of using the entire node deployment region as the measurement area, our method can decrease the number of meshing and also the dimension of measurement matrix. Meanwhile, we assume that only the anchor nodes which have communication relationship with the unknown nodes can be used as the measuring nodes; the measurement matrix of unknown nodes which need to be localized can be dynamically constructed according to the potential area and the received anchor node information, and the maximum number of measurement is decided by the number of grids of potential area. The proposed algorithm can mitigate the measurement redundancy and improve the real-time feature. Simulation results indicate that the proposed algorithm can reduce the time complexity and also maintain good localization accuracy and localization efficiency.