Objects In Different Environments Research Articles

Research on Vision of Intelligent Car Based on Broad Learning System.

The broad learning system (BLS) of intelligent vehicle in different target environments is studied in this article. First, this article provides with the target recognition image data to be trained and detected through the automated guided vehicle (AGV) mobile platform, which can grab the recognition image of different angles and backgrounds. In order to avoid the data generalization phenomenon, the dataset can be expanded by the data normalization and data enhancement. Second, the data are input into the shared convolution layer to extract the feature image and maintain the image. The parameters of image height, width, and channel number are invariable, and the new feature image is obtained by further extraction. Furthermore, the region proposal network (RPN) prefiltering algorithm based on hierarchical clustering is used to filter the objects in the candidate box to determine the region image corresponding to the feature image. Then, the feature images of different sizes input into region of interest (ROI) pooling are used to keep the size of the image in the ROI consistent. Finally, the normalized image is input into the classifier module to obtain the category of the target recognition image to be detected. Through the simulation experiments of different groups, it can be seen that the target recognition system proposed in this design can not only accurately detect the objects but also stably recognize the objects in different environments. The target recognition accuracy for the optimized system is about 95%.

DGFlow-SLAM: A Novel Dynamic Environment RGB-D SLAM without Prior Semantic Knowledge Based on Grid Segmentation of Scene Flow.

Currently, using semantic segmentation networks to distinguish dynamic and static key points has become a mainstream designing method for semantic SLAM systems. However, the semantic SLAM systems must have prior semantic knowledge of relevant dynamic objects, and their processing speed is inversely proportional to the recognition accuracy. To simultaneously enhance the speed and accuracy for recognizing dynamic objects in different environments, a novel SLAM system without prior semantics called DGFlow-SLAM is proposed in this paper. A novel grid segmentation method is used in the system to segment the scene flow, and then an adaptive threshold method is used to roughly detect the dynamic objects. Based on this, a deep mean clustering segmentation method is applied to find potential dynamic targets. Finally, the results of grid segmentation and depth mean clustering segmentation are jointly used to find moving objects accurately, and all the feature points of the moving objects are removed on the premise of retaining the static part of the moving object. The experimental results show that on the dynamic sequence dataset of TUM RGB-D, compared with the DynaSLAM system with the highest accuracy for detecting moderate and violent motion and the DS-SLAM with the highest accuracy for detecting slight motion, DGflow-SLAM obtains similar accuracy results and improves the accuracy by 7.5%. In addition, DGflow-SLAM is 10 times and 1.27 times faster than DynaSLAM and DS-SLAM, respectively.

Unambiguous Identification of Objects in Different Environments and Conditions Based on Holographic Machine Learning Algorithms

Unambiguous provision of results in different environments and conditions by machine learning algorithms is an unresolved problem until now. Solving the problem of machine learning with unambiguous provision of results in different environments and conditions can be approached by focusing on the psychophysical holographic process of human learning. A person, with a mental concentration of attention, experimentally teaches vision, hearing, psyche and mind in a holographic way and in a resonant way to perceive, recognize and recognize phenomena, processes, objects, subjects, meanings, music and other entities in various environments and conditions. A person experimentally teaches the psyche and feelings to rationally navigate in various environments and conditions. Holographic algorithms of experienced machine learning will help neural network ensembles to unambiguously recognize objects, subjects, music, texts in various environments and conditions using a model of recognizing their own or someone else's. Machine learning simulates holographic processes of human communication memorization of entities. Searching for objects in different environments in different conditions based on experienced machine learning simulates resonant associative processes of human entity detection. By simulating holographic processes of the human psyche based on artificial intelligence of machine learning with Fourier transformation, using full parametric sequences of necessary and sufficient data of holograms of target objects, it is possible to solve the problem of their unambiguous detection in different environments and in different conditions.

Unambiguous Identification of Objects in Different Environments and Conditions Based on Holographic Machine Learning Algorithms

Unambiguous Identification of Objects in Different Environments and Conditions Based on Holographic Machine Learning Algorithms

Methods of automated detection of anomalies and nonlinear transitions by autonomous unmanned aerial vehicles

The paper considers the possibility of automating the recognition of anomalies and nonlinear properties of objects in different environments by assembling the UAV payload in the form of a nonlinear locator with a minimum participation of the operator in the processes of probing influence. The process of selecting the main flight platform has a number of requirements that meet the required mission, and by its tactics and technical characteristics it is able to fully implement monitoring algorithms. The theoretical basis of the model construction is given and the main parameters for the standard operation of the model in terms of height, speed and required payload set are defined. The regression model of the interval of scanning of terrain areas depending on the speed of the device and the load of the main processor is compiled. The results of the calculations are summarized and reduced to the method of automated detection of anomalies and nonlinear transitions of objects detected as a result of terrain scanning missions. The main operating properties and optimal conditions of external technological impacts of the developed mathematical model are described.

LCP-Net: A local context-perception deep neural network for medical image segmentation

Automatic image segmentation is an indispensable step in medical image analysis, and it plays an important role in computer-assisted radiotherapy, disease diagnosis and treatment effect evaluation. The difficulty of medical image segmentation is greatly enhanced by the blurry nature of medical image, the complex shape of objects and the existence of noise. In recent years, segmentation methods based on deep learning, especially convolutional neural network, have made great progress in improving the accuracy of medical image segmentation. However, these methods also have poor ability to distinguish similar objects in different environments, because of insufficient use of the local context information of images during the process of feature extraction. To address this problem, this paper proposes a deep neural network (LCP-Net) that can perceive multi-scale context information of images. LCP-Net improves the utilization of context information of feature encoders by using Parallel Dilated Convolution (PDC) and Local Context Embedding (LCE), which are beneficial to get feature map rich in environmental information. In addition, to improve the segmentation accuracy of the model for small objects and alleviate the swing issue during training, we propose a novel improved cross-entropy loss (DDCLoss), which can adaptively adjust the weight of loss according to the certainty and deviation distance of the predicted pixel value and enable the model to focus on optimizing the sample points with low certainty and tend to be mislabeled. Experimental results on three different medical datasets demonstrate that compared with the state-of-the-art medical image segmentation models, our proposed LCP-Net can achieve better segmentation performance.

Analysis on the Algorithm and Practical Application of Computer Intelligent Image Processing Technology

With the continuous improvement of computer science and technology, the role of computer in the development of the whole society is more and more prominent. Based on this development background, people begin to pay attention to the research of computer intelligent processing technology, which can accurately identify and process objects in different environments. At present, the technology in this field has been greatly improved in China, and the application scope of computer intelligent image processing technology has been gradually expanded. The purpose of this paper is to realize the further breakthrough of the computer intelligent processing technology through the exploration of the algorithm and the analysis of the practical application. This article first to the intelligent image processing technology has carried on the summary of relevant, then combined with the existing research materials and the characteristics of the intelligent image processing technology, mathematical morphology of image processing algorithm is proposed, and the effectiveness of the proposed algorithm has carried on the inspection, finally the technology application in the field of industry, agriculture and transportation related instructions. The experimental results show that the image processing algorithm of mathematical form proposed in this paper can realize the accurate processing and recognition of the image and promote the improvement of the technology.

Hybrid End-to-End VPN Security Approach for Smart IoT Objects

The End-to-End VPN security has an essential role especially in connecting smart objects in the Internet of Things (IoT) environments. It noted that security is a crucial issue in the End-to-End VPN approach. IPSec/IPv6 and OpenSSL are the most common VPN approaches used separately to secure smart IoT objects in different environments. These approaches have strengths and weaknesses. This paper addresses the development of a Hybrid End-to-End VPN security approach achieved by combining the IPSec/IPv6 and OpenSSL security approaches that can secure smart IoT objects in different settings. This proposed solution achieves the best combinations of complementary advantages of IPSec/IPv6 VPN approach and OpenSSL VPN approach and eliminates their shortcomings. Performance evaluation of the proposed Hybrid approach and the analysis of the comparison outputs of the performance of both IPSec/IPv6 VPN and OpenSSL VPN approaches assessed in terms of a group of standard measurements, for example, throughput, round-trip delay, and bandwidth consumption. The results of the simulation for the proposed solution demonstrate the correlation with acceptance levels compared to IPSec/IPv6 VPN, and OpenSSL VPN approaches.

Parameter Identification in a Magnetic Suspension Force Transduction System

Abstract In force transduction systems, magnetic suspension can be used to facilitate the measurement of a force acting between two objects in different environments, namely, between the force transducer in one environment and the test object in another environment. In the past, it has been employed in densimetry and equations-of-state measurements of fluids and gases. An instrument being constructed at the National Institute of Standards and Technology (NIST), referred to as the magnetic suspension mass comparator (MSMC), also employs force transduction between two environments to provide a direct means of vacuum-to-air mass dissemination. Within this work, we discuss the identification, through a combination of finite element analysis and experimental measurements, of the system parameters used for feedback control of the suspension. A Hall effect magnetometer is used to determine the position of the suspended object. Emphasis is placed on experimentally and numerically characterizing the magnetic field and its appropriate gradients to determine the set point of operation. The information presented here is useful for a variety of applications that utilize magnetic suspension for force transduction.

Dichotomous classification of black-colored metal using spectral analysis

The task of detecting metal objects in different environments has always been important. To solve it metal detectors are used. They are designed to detect and identify objects that in their electric or magnetic properties different from the environment in which they are located. The most common among them are the metal detectors of the «detection of very low frequency» type (Very Low Frequency (VLF) detectors). They use eddy current testing for detecting metal targets, which solves the problem of dichotomous distinction, that is a problem of splitting (or set) into two parts (subsets): black or colored target.The target distinction is performed by a threshold level of the received signal. However, this approach does not allow to identify the type of target, if two samples of different metals are nearby.To overcome the above described limitations we propose another way of distinction based on the use of spectral analysis, which occurs in the metal detector antenna by Foucault current. We show that the problem of dichotomous distinction can be solved in just a measurement of width and area by the envelope of amplitude spectrum (hereinafter spectrum) of the received signal.In this regard the laboratory model using eddy current metal detector will combat withdrawal from two samples – steel and copper, located along and calculate its range.The task of distinguishing between metal targets reduced to determining the hit spectra of reference samples obtained spectrum. The ratio between the areas is measured and reference spectra indicates the percentage of specific metals (e.g. two identical samples of different metals lying side by side).Signal processing is performed by specially designed program that compares two spectra along posted samples of black and colored metals with base.

The TM-RTree: an index on generic moving objects for range queries

Existing works on moving objects mainly focus on a single environment such as free space and road network, and do not investigate the complete trip for humans who can pass several environments, e.g., road network, pavement areas, indoor. In this paper, we consider multiple environments and study moving objects with different transportation modes, also called generic moving objects. We aim to answer a new class of queries supporting three kinds of conditions: temporal, spatial, and transportation modes. To efficiently provide the result, we propose an index structure called TM-RTree, which takes into account the feature of moving objects in different environments and has the capability of managing objects on not only temporal and spatial data but also transportation modes. This property is not maintained by existing indices for moving objects. Different cases on transportation modes are supported. Correspondingly, several algorithms are developed. The TM-RTree and related algorithms are developed in a real DBMS to have a practical and solid result for applications. In the experiment, we conduct the performance evaluation using extensive datasets and compare the proposed technique with the other two competitors, demonstrating the efficiency and significant superiority of our solution in various settings.

GMOBench: Benchmarking generic moving objects

In real world scenarios, people's movement include several environments rather than one, for example, road network, pavement areas and indoor. This imposes a new challenge for moving objects database that the complete trip needs to be managed by a database system. In the meantime, novel queries regarding different transportation modes should also be supported. Since existing methods are limited to trips in a single environment and do not support queries on moving objects with different transportation modes, new technologies are essentially needed in a database system. In this paper, we introduce a benchmark called GMOBench that aims to evaluate the performance of a database system managing moving objects in different environments. GMOBench is settled in a realistic scenario and is comprised of three components: (1) a data generator with the capability of creating a scalable set of trips representing the complete movement of humans (both indoor and outdoor); (2) a set of carefully designed and benchmark queries; (3) Mode-RTree, an index structure for managing generic moving objects. The generator defines some parameters so that users can control the characteristics of results. We create the benchmark data in such a way that the dataset can mirror important characteristics and real world distributions of human mobility. Efficient access methods and optimization techniques are developed for query processing. In particular, we propose an index structure called Mode-RTree to manage moving objects in different environments. By employing the proposed index, the cost of benchmark queries is greatly reduced. GMOBench is implemented in a real database system to have a practical result. We perform an extensive experimental study on comprehensive datasets to evaluate the performance. The results show that by using the Mode-RTree we achieve significant performance improvement over the baseline method, demonstrating the effectiveness and efficiency of our approaches.

Creation and application methodology of the intelligent information technology of complexity objects space and ground based monitoring

Methodological and techniques foundations of complexity objects monitoring and structural dynamics control theory are proposed in this paper. These foundations include multiple-models descriptions, integrated methods, algorithms and new intelligent information technology which is used for automation designing space and ground based monitoring and control for complex technical-organizational objects in different environments in practice. Some applied examples of proposed intelligent information technology are suggested in the paper.

A generic data model for moving objects

Moving objects databases should be able to manage trips that pass through several real world environments, e.g., road network, indoor. However, the current data models only deal with the movement in one situation and cannot represent comprehensive trips for humans who can move inside a building, walk on the pavement, drive on the road, take the public vehicles (bus or train), etc. As a result, existing queries are solely limited to one environment. In this paper, we design a data model that is able to represent moving objects in multiple environments in order to support novel queries on trips in different surroundings and various transportation modes (e.g., Car, Walk, Bus). A generic and precise location representation is proposed that can apply in all environments. The idea is to let the space for moving objects be covered by a set of so-called infrastructures each of which corresponds to an environment and defines the available places for moving objects. Then, the location is represented by referencing to the infrastructure. We formulate the concept of space and infrastructure and propose the methodology to represent moving objects in different environments with the integration of precise transportation modes. Due to different infrastructure characteristics, a set of novel data types is defined to represent infrastructure components. To efficiently support new queries, we design a group of operators to access the data. We present how such a data model is implemented in a database system and report the experimental results. The new model is designed with attention to the data models of previous work for free space and road networks to have a consistent type system and framework of operators. In this way, a powerful set of generic query operations is available for querying, together with those dealing with infrastructures and transportation modes. We demonstrate these capabilities by formulating a set of sophisticated queries across all infrastructures.

Optimization of Short Probe Linear Frequency Modulated Signal Parameters

In many physical experiments, linear frequency modulated (LFM) signals are widely used to probe objects in different environments, from outer-space to underwater. These signals allow a significant improvement in measurement resolution, even when the observation distance is great. For example, using LFM probe signals in underwater investigations enables discovery of even small objects covered by bottom sediments. Recognition of LFM (chirp) signals depends on their compression based on matched filtering. This work presents two simple solutions to improve the resolution of the short chirp signals recognition. These methods are effective only if synchronization between the signal and matched filter (MF) is obtained. This work describes both the aforementioned methods and a method of minimizing the effects of the lack of synchronization. The proposed matched filtering method, with the use of $n$ parallel MFs and other techniques, allows only one sample to be obtained in the main lobe and to accurately locate its position in the appropriate sampling period $T_s$ with accuracy $T_{s/n}$. These approaches are appropriate for use in probe signal processing.

Aspect-Trees: Generation and Interpretation

This contribution presents a new approach to the recognition of a priori known 3-D objects in single 2-D images. The underlying model is embedded in the domain of CAD-based vision using a viewer-centered approach to generate a set of normalized views. They serve as a basis for an optimal selection of properties of features. The aspect idea is used for grouping the values of the properties into aspect-trees. The aim of this approach is to identify the correct view of an object seen in the image and thereby to distinguish between different objects. This is achieved with an appropriate traversion of the aspect-trees which proves the ability of the image interpretation system Aspik to recognize complex objects in different environments robustly and efficiently. The time complexity for recognizing nonoccluded objects is O (n2 · m), where n is the number of the considered aspect-trees of the object and m the number of image features.

Ultraviolet interstellar absorption lines in the LMC: Searching for hidden SNRs

Strong x-ray emission detected in Large Magellanic Cloud (LMC) superbubbles has been explained as the result of interior supernova remnants (SNRs) hitting the dense superbubble shell. Such SNRs cannot be found using conventional criteria. We thus investigate the possibility of using the interstellar absorption properties in the ultraviolet (UV) as a diagnostic of hidden SNR shocks. The International Ultraviolet Explorer (IUE) archives provide the database for this pilot study. They contain high-dispersion spectra of several stars in x-ray bright superbubbles. To distinguish the effects of SNR shocks from those of local stellar winds and a global hot halo around the LMC, we included control objects in different environments. We find that almost all interstellar absorption properties can be explained by the interstellar environment associated with the objects. Summarizing the two most important results of this study: (1) a large velocity shift between the high-ionization species (C IV and Si IV) and the low-ionization species (S II, Si II, and C II*) is a diagnostic of hidden SNR shocks; however, the absence of a velocity shift does not preclude the existence of SNR shocks; (2) there is no evidence that the LMC is uniformly surrounded by hot gas; hot gas is preferentially found associated with large interstellar structures like superbubbles and supergiant shells, which may extend to large distances from the plane.