Flexible feature detection by structural information

The distinction between operations in visual processing that are parallel and preattentive and those that are serial and attentional receives both theoretical and empirical support. According to Treisman's feature-integration theory, independent features are available preattentively, but attention is required to veridically combine features into objects. Certain evidence supporting this theory is consistent with a different interpretation, which was tested in four experiments. The first experiment compared the detection of features and feature combinations while eliminating a factor that confounded earlier comparisons. The resulting priority of access to combinatorial information suggests that features and nonlocal combinations of features are not connected solely by a bottom-up hierarchical convergence. Causes of the disparity between the results of Experiment 1 and the results of previous research were investigated in three subsequent experiments. The results showed that of the two confounded factors, it was the difference in the mapping of alternatives onto responses, not the differing attentional demands of features and objects, that underlaid the results of the previous research. The present results are thus counterexamples to the feature-integration theory. Aspects of this theory are shown to be subsumed by more general principles, which are discussed in terms of attentional processes in the detection of features, objects, and stimulus alternatives.

Open AccessCCS ChemistryMINI REVIEW1 Mar 2022Multidimensional Mass Spectrometry Assisted Metallo-Supramolecular Chemistry Heng Wang, Chenxing Guo and Xiaopeng Li Heng Wang College of Chemistry and Environmental Engineering, Shenzhen University, Shenzhen 518073 Shenzhen University General Hospital, Shenzhen University Clinical Medical Academy, Shenzhen 518055 , Chenxing Guo College of Chemistry and Environmental Engineering, Shenzhen University, Shenzhen 518073 and Xiaopeng Li *Corresponding author: E-mail Address: [email protected] College of Chemistry and Environmental Engineering, Shenzhen University, Shenzhen 518073 Shenzhen University General Hospital, Shenzhen University Clinical Medical Academy, Shenzhen 518055 https://doi.org/10.31635/ccschem.021.202101408 SectionsAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail Metallo-supramolecular chemistry based on the reversible, directional, and predictable noncovalent interaction of metal–ligand coordination has been widely applied in constructing a myriad of two- and three-dimensional supramolecules with sophisticated structures. The development of metallo-supramolecular chemistry highly relies on facile characterization methods. For advancement of structural complexity with desired functions, however, conventional crystallographic and spectroscopic characterizations are limited by crystal growth, sample purity, and severe signal overlaps. With high sensitivity and low requirement for purity, multidimensional mass spectrometry (MS), including electrospray ionization-MS (ESI-MS), tandem MS (MS2), and ion mobility-MS (IM-MS), is emerging as an essential analytical tool for characterization of the chemical compositions, connectivity, stability, shape, and size of metallo-supramolecules, as well as kinetic and thermodynamic features of self-assembly processes. Thus, this review mainly focuses on the recent development of multidimensional MS in metallo-supramolecular characterization and discusses how ESI-MS, MS2, and IM-MS assist the development of metallo-supramolecular chemistry. Download figure Download PowerPoint Introduction Supramolecular self-assembly, inspired by the precise organization of bio-macromolecules in living systems, allows for the spontaneous formation of ordered complexes from a chaos state.1 Among various noncovalent driving forces, metal–ligand coordination plays a vital role in the self-assembly of structurally and functionally sophisticated metallo-supramolecules, benefitting from the high directionality and predictable feature of coordination bonds.2–6 As a result, metallo-supramolecular chemistry has witnessed an explosion in constructing a myriad of artificial supramolecular architectures, including polygons,7–11 knots,12–15 links,16–19 fractals,20–22 as well as Platonic,23–27 Archimedean,28–34 Goldenberg polyhedrons,35,36 and so on. And these complexes show great potentials among different fields, including catalysis,37–46 nano-electronics,47,48 stabilization of air- or water-sensitive molecules,49–51 disease diagnosis and therapy,52–57 and so on. Furthermore, various delicate assembling approaches, inter alia, directional bonding,2 symmetry interactions,58 molecular paneling,59 and sub-components,60 have been developed to assist the exploration of metallo-supramolecular chemistry. Compared to the flourish of self-assembly strategies, however, the lack of efficient characterization methods always limits the investigation of assembled complexes at equilibrium and the kinetic details of self-assembly processes in solution because of the weak interaction and dynamic nature of metallo-supramolecules. Among the conventional characterizations of metallo-supramolecules, single-crystal X-ray diffraction (SC-XRD) is a powerful and direct method that allows for detailed structural information at the atomic level.61 However, it remains a formidable challenge to obtain high-quality single crystals, particularly for metallo-supramolecules with large voids filled by disordered solvent molecules or counterions. In addition, the refinement of the crystal structures for large metallo-supramolecules is difficult because the current crystal refinement approaches were mainly developed for small molecules. Apart from those issues the structural information obtained through SC-XRD in solid phase may not correlate with the study of metallo-supramolecular systems in solution or gas phase. Another well-established characterization method is NMR spectroscopy, whereby various nuclei in metallo-supramolecular systems, such as 1H, 13C, 19F, 31P, and so on can be detected, allowing the chemical environment of these nuclei in metallo-supramolecules to be elucidated.62,63 Through two-dimensional (2D) NMR spectroscopic techniques, for example, 1H–1H correlation spectroscopy (COSY), nuclear Overhauser effect spectroscopy (NOESY), and diffusion ordered spectroscopy (DOSY), diverse structural information about the metallo-supramolecules can be further obtained, including metal–ligand coordinations, geometric distributions of nuclei, and details of the size and shape of assembled architectures. However, the application of NMR is also limited in analyzing samples with low purity, severe overlapping signals, or those containing paramagnetic metal ions. As complementary characterization techniques, various microscopic methods, including atomic force microscopy (AFM), scanning tunneling microscopy (STM), and transmission electron microscopy (TEM), directly image the analytes. AFM and STM, in which tips probe the surface of the target molecules on substrates, reveal the size and shape of an individual molecule64 and distinguish the metal ions by using scanning tunneling spectroscopy (STS).65 However, high image resolution relies on sharp tips, as well as high-quality samples without impurities, severe aggregations, or dissociations of the target supramolecules. TEM uses electrons as the "probe" and can offer atomic resolution of the entire molecular structure assisted by computer-assisted three-dimensional (3D) reconstruction, selected area electron diffraction (SAED), and energy dispersive X-ray spectroscopy (EDX) techniques.66 Unfortunately, in addition to the challenges of preparing a high-quality sample, TEM also suffers from the intrinsic instability of metallo-supramolecules under electron beam or during sample preparation. The mass-to-charge ratio (m/z) information of the intact metallo-supramolecular ions obtained by mass spectrometry (MS) measurements in the gas phase enables the determination of the chemical composition of analytes. Furthermore, MS characterization provides higher sensitivity but a lower sample purity requirement than crystallographic analysis and NMR. Such features along with high analytical speed make it possible to monitor the dynamic process of self-assembly before the system in question reaches equilibrium. Thus, MS is emerging as an essential analytical tool for determining structures and monitoring reaction processes. However, traditional MS approaches also suffer from the fragmentation of metallo-supramolecules bound by weak coordination interaction even when a soft ionization method, that is, electrospray ionization (ESI), is applied. Cold-spray ionization (CSI),67 equipped with an ESI source cooled by liquid nitrogen, was used to overcome the problem by lowering the internal energy gained during the ionization. Due to the labile connectivity, matrix-assisted laser desorption/ionization (MALDI) could cause very severe fragmentation and thus is not preferred for the characterization of metallo-supramolecules. In addition to the m/z information obtained from MS, tandem MS (MS2) and ion mobility-MS (IM-MS) can elucidate more detailed information of the analytes. First, as a controllable fragmentation technique, MS2 offers deeper insight into the analytes, such as the information about the stability and connectivity of the structures. Furthermore, as an orthogonal in situ separation method, IM-MS can identify different species with the same m/z; more importantly, it provides structural information in terms of the shape and size of selected ions. During the past two decades, multidimensional MS with the combination of MS, IM-MS, and MS2 has played an important role in the development of metallo-supramolecular chemistry. Here, we discuss how ESI-MS, MS2, and IM-MS can be applied in (1) determining chemical compositions, connectivity, shape, and size of metallo-supramolecules, (2) distinguishing the isomers and by-products present in a dynamic combinatorial library, and (3) investigating kinetic and thermodynamic features of self-assembly processes. More importantly, we anticipate that multidimensional MS would not only act as an analytical method, but also serve as a powerful tool to assist the design of next-generation metallo-supramolecules, and even advance the development of new synthetic strategies. ESI-MS Studies Charged coordination complexes are ideal analytes for ESI-MS studies because of the easy-to-ionize feature of these supramolecules in which the counterions are readily stripped off, especially in polar and volatile solvents. Another advantage of ESI-MS is its intrinsic capability of reducing the signals corresponding to the multiply charged analytes with high molar mass (<104 Da) to the mass range of the conventional mass analyzers, since it only measures the m/z values of these ensembles. Consequently, a series of peaks is usually observed, representing different charge states of the same intact analyte through successive loss of counterions. To minimize the fragmentation during ionization, it is crucial to ionize the fragile analytes under mild conditions, including careful selection of the focusing and ionization voltages, low operation temperature, and fast ion extraction process through increasing source pressure (or a low vacuum).68 Additionally, highly concentrated metallo-supramolecular samples are often required to inhibit the potential entropy-driven dissociation in solution, although the ESI source needs more extensive cleaning after each measurement. Without annoying fragmentation, ESI-MS is suitable for detecting the chemical compositions of the ionic species in the metallo-supramolecular analytes, as well as providing information regarding the following aspects of the self-assembly: (1) fast screening the assembled structures; (2) monitoring the dynamic processes during the self-assembly; and (3) detecting the minor products to guide further design. Acting as a fast-screening characterization approach for the self-assembly Fast screening of the analytes' structural information by ESI-MS does not require extensive purification processes that are otherwise inevitable for NMR measurement and single-crystal growth. In most cases, after simply removing the inorganic salts to reduce ion suppression, the resulting sample solution is ready for direct measurement. With careful control of stoichiometry, the solution containing the self-assembled species can be directly subjected to ESI-MS without any further purification. Additionally, its capability of detecting m/z values enables the characterization of multiple assemblies with different numbers of building blocks, paramagnetic metallo-supramolecular analytes, and host–guest adducts. Therefore, ESI-MS could serve as a complementary or even superior method to perform structural characterization compared to NMR. The first example showcasing the fast screening by ESI-MS is the characterization of different generations of metallo-dendrimers S1/ S2 (Figure 1a) comprised of polyazaaromatic Ru(II) complexes reported by Kirsch-De Mesmaeker et al.69 ESI-MS directly generated clear and simple patterns that allowed the unambiguous determination of the chemical formula, whereas the other spectroscopic characterization just provided nonconclusive data (Figure 1b). Second, in the characterization of metallo-macrocycles reported by Newkome and co-workers,701H NMR gave very similar chemical shifts for the hexagon, octagon, and decagon ( S3- S5) assembled from the Ru(II)-connected bitopic 2,2′:6′,2″-terpyridyl (TPY) subunit L3 (Figure 1c) because of the similar symmetry and chemical environments of the detected protons. In contrast, these analogs with distinct molar mass were unambiguously identified by ESI-MS, where three distinct sets of peaks were detected (Figures 1d and 1e).70 Figure 1 | ESI-MS acting as a fast-screening approach for the preparation of (a and b) metallo-dendrimers comprised of polyazaaromatic Ru(II) complexes S2 and its corresponding ESI-MS spectrum; (c–e) TPY-Ru(II)/Fe(II)-based metallo-hexagon S3, -octagon S4, and -decagon S5 with corresponding 1H NMR spectra and ESI-MS spectra; (f–h) self-assembly of supramolecular fractals S6 based on the metal complexation between a TPY-based ligand and paramagnetic metal ions with corresponding 1H NMR and ESI-MS spectrum of the Co(II) complex (S6-Co). (a and b) Adapted with permission from ref 69. Copyright 1996 American Chemical Society. (c–e) Adapted with permission from ref 70. Copyright 2011 Wiley-VCH. (f–h) Adapted with permission from ref 71. Copyright 2020 American Chemical Society. Download figure Download PowerPoint Metallo-supramolecules containing paramagnetic transition metal ions with unpaired electrons remain a formidable challenge for NMR characterization. In contrast, ESI-MS has no problem characterizing these samples. For example, Shionoya et al.25 constructed a series of metallo-supramolecular octahedra by mixing a tritopic pyridinyl building block with different transition metal ions. ESI-MS spectra of the paramagnetic analytes containing Cu(II), Ni(II), Co(II), and Mn(II) ions clearly showed a dominant set of peaks ascribed to the designed structures; whereas, NMR spectra only displayed broad peaks with limited information.25 Similarly, with the help of ESI-MS characterization, coordination-driven self-assembly based on a TPY-paramagnetic metal ion complex was applied to construct a series of supramolecular fractals reported by Li et al.,71 which not only expanded the library of metal ions for assembling TPY-based architectures but also set up an ideal platform for studying the kinetic and thermodynamic features of these metallo-supramolecules (Figures 1f–1h). When characterizing the paramagnetic Co(II) and Ni(II) cages assembled using the sub-component self-assembly strategy, the reliable ESI-MS method also played a vital role in identifying the formation of target structures.72,73 Investigation of host–guest complexation is an important subfield of metallo-supramolecular chemistry, which is related to the fundamental study of molecular recognition and the potential applications of these metal–organic containers. NMR spectroscopy is extensively applied to provide evidence for the interaction with diagnostic chemical shifts of the protons on guest molecules. However, it cannot always accurately address how many guest molecules are encapsulated in the containers. ESI-MS can properly examine the exact change in mass upon the encapsulation of the guest molecules, thus, providing an accurate m/z value for deducing the number of guest molecules. In a self-assembled M8L6 (M = the metal ion, L = the ligand) cube with porphyrin faces, Nitschke et al.74 described that large guests such as coronene and fullerene can be selectively encapsulated (Figure 2a). Compared with the complicated NMR spectra that only provided a 2:1 ratio of guests in two different environments, ESI-MS spectrum of the coronene-encapsulated cube directly provided the number of the guest molecules via deconvolution of the dominant set of peaks (Figures 2b and 2c). The ESI-MS spectrum also clearly confirmed that the cube allowed for selective encapsulation of C70, C76, C82, and C84 over C60 from the fullerene soot, as evidenced by the absence of any discernible peaks assignable to the C60-encapsulated or guest-free cube. Figure 2 | ESI-MS-based fast screening for studying the guest encapsulation of metallo-supramolecular cages. (a–c) Self-assembly of porphyrin-faced cube S8 for encapsulation of three coronene with corresponding characterization by 1H NMR and ESI-MS; (d and e) a PBI-edged tetrahedron capable of encapsulating two molecules of C60S9 and the corresponding ESI-MS spectrum; (f–h) self-assembly of a metallo-supramolecular peanut S10 for simultaneous encapsulation of one diamantane and two phenanthrenes (S11) with corresponding characterization by 1H NMR and ESI-MS. (a–c) Adapted with permission from ref 74. Copyright 2011 Wiley-VCH. (d and e) Adapted with permission from ref 76. Copyright 2013 American Chemical Society. (f–h) Adapted with permission from ref 83. Copyright 2017 Springer Nature. Download figure Download PowerPoint Thus, ESI-MS is also a feasible and rapid method for qualitatively studying the encapsulation selectivity of a certain metallo-supramolecule with different guests. In another example of selective encapsulation, when mixing a metallo-supramolecular tetragonal prism with a mixture of C60 and its derivatives, the selectivity could be determined qualitatively by comparing the relative abundance of the peaks corresponding to the host–guest complexes containing different guests. Among them, C60 and PCBM-C60 have higher affinity with the host than N-methylpyrrolidine-C60.75 However, it is still challenging to quantitatively determine the selectivity by ESI-MS because of the different ionization efficiency of those complexes, although the host molecules are the same. Ideally, quantitative data could be obtained if the ionization efficiencies of different complexes are calibrated. In addition to encapsulating one fullerene molecule, Würthner et al.76 applied ESI-MS to unequivocally prove the encapsulation of one or two molecules of C60 inside the cavity of perylene-bisimide (PBI) edged tetrahedron S9 (Figures 2d and 2e). Note that the C60 guest does not bear any protons so that it can only be identified by 13C NMR. Although the 13C NMR could confirm the encapsulation of the C60, it was unable to determine the number of guest molecules because conventional 13C NMR does not allow for quantitative analysis as opposed to 1H NMR. Metallo-supramolecular trigonal prisms, wherein C3-symmetrical tritopic pyridyl ligands, linear ditopic pyridyl ligands, and an end-capped Pd(II) moiety served as the base faces, pillars, and vertices, respectively, were found to be an excellent molecular container capable of encapsulating a variety of guest molecules, including, coronene,77 triphenylene,78 pyrene-4,5-dione,79,80 and metal–organic complexes.81,82 More intriguingly, the number of the encapsulated guest molecules can be precisely manipulated via tuning the length of the organic pillars. The accurate number of encapsulated guests could be unambiguously determined by ESI-MS. A more complicated situation was reported by Yoshizawa et al.83 The self-assembled metallo-supramolecular peanut S10 heteroleptically encapsulated two different types of guests, that is, one diamantane and two phenanthrenes, inside its two cavities (Figure 2f). Without an SC-XRD result or clear NMR spectrum (Figure 2g), such an unusual encapsulation was identified by ESI-MS (Figure 2h), where peaks with m/z values of the expected host–guest complex were found. Monitoring the dynamic processes for unveiling self-assembly details Metallo-supramolecules bonded via labile coordination interactions exhibit dynamic features in solution due to the dissociation and reassociation of the weak bonds. Thus, it is hard to investigate the kinetic processes by traditional characterization techniques in solution. Nevertheless, in an ESI-MS measurement, each complex ion is a separated species in gas phase. Therefore, the dynamic process is "frozen" to display the kinetic instead of thermodynamic feature of the system. As one type of dynamic process, ligand exchange involves the disassociation of a ligand from the assembled supramolecule and subsequent replacement by another ligand. Such a process is related to the coordination bond strength and the activation energies during the exchange processes. If the time scales of such dynamic exchanges are comparable for MS measurement, one should be able to study the exchange process of two kinds of pre-assembled supramolecules with distinct molar masses of ligands by MS. Once the exchange occurs, the m/z values corresponding to each possible re-assembled species as well as the initial ensembles would be different, all of which can be specifically recorded by ESI-MS measurement. For example, Fujita et al.84 reported the studies of ligand exchange between two M12L24 spheres via mixing two topologically identical spheres bearing different length of alkyl chains (–OC3H7 vs –OC6H13) and then monitoring under time-dependent ESI-MS (Figures 3a and 3b). Due to the difference in mass, all the peaks corresponding to the initial two spheres and the exchanged ones were easily identified in the spectra. Time-dependent measurements revealed a quite remarkable kinetic stability of the spheres regarding ligand exchange. The half-time for the first ligand exchange was measured to be approximately 20 days. After evaluating the ionization efficiency of two spherical supramolecules, the estimated kinetic ligand exchange rate was found to be 2 × 10−2 M−1·s−1, which is far less than the exchange rate between a free ligand with a sphere (half-time of 23 min). Therefore, the ESI-MS studies semi-quantitatively revealed that the fully closed framework slowed down the dynamic exchange processes through a cooperative effect.84 A similar method was also utilized to carefully study the dynamic nature of metallo-polygons assembled from organoplatinum motifs and pyridyl ligands with 1H/2D isotope labeling.85 In addition to the quantitative determination of the ligand exchange rates, further understanding about the kinetic processes in coordination-driven self-assembly was provided by time-dependent ESI-MS, which showed the ligand exchange rate highly depended on the experimental conditions such as temperature, solvent, and counterions. Figure 3 | ESI-MS of ligand exchange. (a) Ligand exchange between two structurally identical spheres (Pd12L724 and Pd12L824) with different alkyl chain lengths (-n-C3H7 vs -n-C6H13), and (b) the time-dependent ESI-MS spectra; (c) self-assembly of four types of supramolecular snowflakes (S12–S15) with different lengths of the ligands, and time-dependent ESI-MS spectra of the mixture of and (a and b) Adapted with permission from ref Copyright American Chemical Society. and e) with permission from ref Copyright Wiley-VCH. with permission from ref Copyright 2017 American Chemical Society. Download figure Download PowerPoint In addition to the pyridyl ESI-MS also provided insight into the ligand exchange processes of metallo-supramolecular systems based on analyzing the peaks to the complexes in ESI-MS it was revealed that the exchange processes a certain of to the size and shape of the ligand via the of the structure when the Through a time-dependent ESI-MS measurement, such a was estimated in the metallo-supramolecular snowflakes with different length of ligand The exchange rate was found to directly correlate with that is, the the the ligand exchange. When = the exchange equilibrium = required for whereas = in no exchange (Figures Furthermore, the ligand exchange rate was also found to correlate with the coordination metal ion in TPY-based To quantitatively these exchange rates, different metal ions (M = were applied. of these metal ions was used to two structures with the same molecular size and shape but different molar After mixing the of pre-assembled supramolecules, time-dependent ESI-MS was utilized to the relative abundance of each After the ionization efficiency of these supramolecules, ESI-MS data provided exchange of these metallo-supramolecules at with a of Mn(II) Co(II), whereas the and fractals were found to be to the ligand exchange. The exchange rate of fractals was determined to be × M−1·s−1, which is higher than that of fractals × Monitoring the in the self-assembly processes is also a kinetic investigation that can be carefully by ESI-MS. When using NMR to monitor the complexation of ligands with or Ni(II) shifts of the peaks were observed, the of an before the formation of However, the limited information provided by NMR spectra not elucidate the ESI-MS was then used to the and the that is, linear by the m/z value of the intact Thus, ESI-MS can reveal the reaction via monitoring the Such capability was also used in monitoring the formation of a supramolecular wherein ligand and self-assembly were in a The time-dependent the of the coordination and processes (Figures and Figure | (a) The formation of supramolecular via a ligand and self-assembly (b) time-dependent ESI-MS monitoring the (b) Adapted with permission from ref Copyright American Chemical Society. Download figure Download PowerPoint With m/z values of the more details about the self-assembly processes could be allowing for deeper understanding of the assembling For in the of ligands with ions to and an with metal ion in the ligand with instead of was through ESI-MS by increasing ligand or metal Such an that the complexation to the A similar of complexation was also in other systems by ESI-MS, its ESI-MS of the species during the formation of prism with a ligand revealed the of the The ESI spectra the of self-assembly and showed the formation of the base with a Such a the design of a building block with two types of two distinct geometric that is, as base and respectively, allowing for the self-assembly of a structures to further design With the reliable m/z ESI-MS can identify the desired before other and characterization. For based on the directional an tritopic ligand was designed to metallo-supramolecules by coordination with or to the assembled could be or Unfortunately, NMR characterization because of the Nevertheless, ESI-MS provided evidence for the formation of a which the of the ligand to the of the complex for structural on

This paper proposes a high-precision chip positioning and defect detection technology based on machine vision. Divided into low-high double precision detection process, the probe are roughly aligned with the initial test point at low magnification, the fine alignment and feature detection at high magnification. First, the chip image was pre-processed at a low magnification to obtain the MER, and the orientation of the initial test point was obtained by the template matching method, the rough position of the probe tip was obtained by Harris detection. At high magnifications, an improved positioning algorithm is used to obtain the sub-pixel coordinates of the center of the test point and the probe tip. Combined with the clustering method, the row and column spacing of the test point array is obtained. At the same time, the characteristics of test points are extracted in the test point detection process and compared with the standard database to detect defects.KeywordsMachine visionChip detectionPositioning accuracyDefect detection

The paper deals with the feature detection problem in aircraft machine vision systems. We developed a fast detection algorithm. We determined feature locations in an image sequence depicting fast-moving objects as registered by aircraft machine vision systems. We analysed existing feature detection algorithms. The input image frame sequence undergoes a geometric transformation, that is, rotation. We simulated how our algorithm processed a sequence of frames depicting the target. We compared efficiency and speed of feature detection in the well-known SIFT algorithm with that of our algorithm. We demonstrate that the algorithm developed ensures faster feature detection that is more resistant to geometric transformations of the target image as compared to the SIFT algorithm.

Presently, cognitive Internet of Things (CIoT) with cloud computing (CC) enabled intelligent healthcare models are developed, which enables communication with intelligent devices, sensor modules, and other stakeholders in the healthcare sector to avail effective decision making. On the other hand, Alzheimer disease (AD) is an advanced and degenerative illness which injures the brain cells, and its earlier detection is necessary for suitable interference by healthcare professional. In this aspect, this paper presents a new Oriented Features from Accelerated Segment Test (FAST) with Rotated Binary Robust Independent Elementary Features (BRIEF) Detector (ORB) with optimal artificial neural network (ORB-OANN) model for AD diagnosis and classification on the CIoT based smart healthcare system. For initial pre-processing, bilateral filtering (BLF) based noise removal and region of interest (RoI) detection processes are carried out. In addition, the ORB-OANN model includes ORB based feature extractor and principal component analysis (PCA) based feature selector. Moreover, artificial neural network (ANN) model is utilized as a classifier and the parameters of the ANN are optimally chosen by the use of salp swarm algorithm (SSA). A comprehensive experimental analysis of the ORB-OANN model is carried out on the benchmark database and the obtained results pointed out the promising outcome of the ORB-OANN technique in terms of different measures.

Cloud computing is a more advanced technology for distributed processing, e.g., a thin client and grid computing, which is implemented by means of virtualization technology for servers and storages, and advanced network functionalities. However, this technology has certain disadvantages such as monotonous routing for attacks, easy attack method, and tools. This means that all network resources and operations are blocked all at once in the worst case. Various studies such as pattern analyses and network-based access control for infringement response based on Infrastructure as a Service, Platform as a Service and Software as a Service in cloud computing services have therefore been recently conducted. This study proposes a method of integration between HTTP GET flooding among Distributed Denial-of-Service attacks and MapReduce processing for fast attack detection in a cloud computing environment. In addition, experiments on the processing time were conducted to compare the performance with a pattern detection of the attack features using Snort detection based on HTTP packet patterns and log data from a Web server. The experimental results show that the proposed method is better than Snort detection because the processing time of the former is shorter with increasing congestion.

Forecasting of congestion in traffic with Neural Network is an innovative and new process of identification and detection of chaotic features in time series analysis. With the use of Duffing Holmes Oscillator, we estimate the emergence of traffic flow congestion when the traffic load on a specific section of the road and in a specific time period is close to exceeding the capacity of the road infrastructure. The orientated model is validated in six locations with a specific requirement. The paper points out the issue of importance of traffic flow forecasting and simulations for preventing or rerouting possible short term traffic flow congestions.

Big Data is defined as an emerging paradigm that includes complex and large-scale information beyond the processing capability of conventional tools. Traditional data analytics methods have been commonly used for many applications, such as text classfication, image recognition, and video tracking. For analysis purposes, these data often need to be represented as vectors. However, many other types of data objects in real-world applications contain rich feature vectors and structure information, such as chemical com pounds in bio-pharmacy, brain regions in brain networks and users in social networks. Unfortunately, vector representa tions are very simple features that do not inherently contain the object’s structure information. In reality, objectsmay have complicated characteristics depending on how the objects are assessed and characterized. Data may also reside in het erogeneous domains, such as traditional tabular-based data, sequential patterns, social networks, time series information, and semi-structured data. As a result, novel data analytics methods are needed to discover meaningful knowledge in advanced applications from objects with large-scale com plex characteristics. In total, there were 104 submissions to this IEEE ACCESS Special Section, with 39 accepted after rigorous peer review.

Sentence embeddings from transformer models encode much linguistic information in a fixed-length vector.We investigate whether structural information -specifically, information about chunks and their structural and semantic properties -can be detected in these representations.We use a dataset consisting of sentences with known chunk structure, and two linguistic intelligence datasets, whose solution relies on detecting chunks and their grammatical number, and respectively, their semantic roles.Through an approach involving indirect supervision, and through analyses of the performance on the tasks and of the internal representations built during learning, we show that information about chunks and their properties can be obtained from sentence embeddings.

The aim of this study was to evaluate (a) the effect of aging on spectral and temporal resolution, as measured both behaviorally and electrophysiologically, and (b) the contributions of spectral and temporal resolution and cognition to speech perception in younger and older adults. Eighteen younger and 18 older listeners with normal hearing or no more than mild-moderate hearing loss participated in this cross-sectional study. Speech recognition was assessed with the QuickSIN test and six-band noise-vocoded sentences. Frequency discrimination, temporal interval discrimination, and gap detection thresholds were obtained using a three-alternative forced-choice task. Cortical auditory evoked potentials were recorded in response to tonal frequency changes and to gaps in noise. Cognitive testing included nonverbal reasoning, vocabulary, working memory, and processing speed. There were age-related declines on many outcome measures, including speech perception in noise, cognition (nonverbal reasoning, processing speed), behavioral gap detection thresholds, and neural correlates of spectral and temporal processing (smaller P1 amplitudes and prolonged P2 latencies in response to frequency change; smaller N1-P2 amplitudes and longer P1, N1, P2 latencies to temporal gaps). Hearing thresholds and neural processing of spectral and temporal information were the main predictors of degraded speech recognition performance, in addition to cognition and perceptual learning. These factors accounted for 58% of the variability on the QuickSIN test and 41% of variability on the noise-vocoded speech. The results confirm and extend previous work demonstrating age-related declines in gap detection, cognition, and neural processing of spectral and temporal features of sounds. Neural measures of spectral and temporal processing were better predictors of speech perception than behavioral ones. https://doi.org/10.23641/asha.28883711.

Currently, the contextual bandit (CB) has been applied in the field of slow-moving small target detection on sea surface, to solve the performance decline problem of feature detection method under less coherent pulse number. However, the existing method only gives a brief description on its implementation details and also has two shortcomings: 1) the algorithms used for estimating target location (TL) lack theoretical support and 2) the adopted policy is idealized. For the first issue, we assess the rationality and robustness of the initial TL estimation method, the online update algorithm for estimated TL, and the suppression algorithm for successive false alarms, both theoretically and experimentally. Then, we summarize the above three parts and propose a real-time TL estimation method. For the second issue, by modeling a nonlinear relationship between the expected reward of action and its observed context, the kernelized upper confidence bound (KernelUCB) policy is introduced. Subsequently, a k-nearest neighbor (KNN)-based approach is proposed to reduce its algorithm complexity. Furthermore, we propose designing an alternative action, which is enabled once main action chosen by the policy is invalid, thereby improving the performance of policy. By combining the above improvements with the CB process modeled from the classical feature detection process, the proposed method is obtained, which can realize the autonomous optimized selection for alternative feature detectors on each frame. The sufficient experiments demonstrate that the proposed method can integrate the performance advantages of various feature detectors while meeting the real-time detection requirement.

In this article we deal with one of the fundamental problems in the area of the 3D reconstruction for objects with homogeneous surface such as, inter alia, human body or sculptures. The interest point detection on typical photos with many differing elements and changing intensities is already well-solved issue. Considerable difficulty and novelty is the interest point detection for homogeneous surfaces. To reconstruct such surfaces from images we have to artificially produce as many elements on surface as needed to allow proceed with the 3D coordinate's extraction process with desired density. Four methods were selected. The first, definitely the best documented was the Harris corner detector. Next was the Nobel's version of auto-correlation, the other was the minimum eigenvalue method known as the Kanade-Tomasi algorithm and the last tested method was the fast radial feature detector known as the Loy-Zelinsky algorithm. Chosen methods are well-known on the 3D reconstruction theatre, well implemented and documented, efficient in the terms of computational complexity. Also some image enhancements were utilized before feature extraction to improve the detection process. It was shown that the best choice was the Nobel's version of auto-correlation function and a very interesting candidate for further research is the Loy-Zelinsky method.

Although direct volume rendering is established as a powerful tool for the visualization of volumetric data, efficient and reliable feature detection is still an open topic. Usually, a tradeoff between fast but imprecise classification schemes and accurate but time-consuming segmentation techniques has to be made. Furthermore, the issue of uncertainty introduced with the feature detection process is completely neglected by the majority of existing approaches.In this paper we propose a guided probabilistic volume segmentation approach that focuses on the minimization of uncertainty. In an iterative process, our system continuously assesses uncertainty of a random walker-based segmentation in order to detect regions with high ambiguity, to which the user's attention is directed to support the correction of potential misclassifications. This reduces the risk of critical segmentation errors and ensures that information about the segmentation's reliability is conveyed to the user in a dependable way. In order to improve the efficiency of the segmentation process, our technique does not only take into account the volume data to be segmented, but also enables the user to incorporate classification information. An interactive workflow has been achieved by implementing the presented system on the GPU using the OpenCL API. Our results obtained for several medical data sets of different modalities, including brain MRI and abdominal CT, demonstrate the reliability and efficiency of our approach.

For the implementation of stereo image-based visual servoing algorithm in the eye-in-hand robotics applications, one of the main concerns is the accurate point feature detection and matching algorithm. Since the visual servoing is carried out in the textureless environment, the feature detection process is even more challenging. To fulfill the requirement of a robust and reliable point feature detection process, in this paper we present the novel deep learning-based algorithm. The approach based on convolutional neural networks and algorithm for detection of manufacturing entities is proposed and detected regions of interest are utilized for the improvement of the point feature detection algorithm. The proposed algorithm is experimentally evaluated in real-world settings by using wheeled nonholonomic mobile robot RAICO equipped with stereo vision system. The experimental results show the improvement of 58% in the accuracy of matched point features in the images obtained during the visual servoing process. Moreover, with the implementation of the proposed deep learning-based approach, the number of successful experimental runs has increased by 80%.

While shadow can give useful information about size and shape of objects, it can pose problems in feature detection and object detection, thereby, it represents one of the major perturbator phenomenons frequently occurring on images and unfortunately, it is inevitable. “Shadows may lead to the failure of image analysis processes and also cause a poor quality of information which in turn leads to problems in implementation of algorithms.” (Mahajan and Bajpayee, 2015). It also affects multiple image analysis applications, whereby shadow cast by buildings deteriorate the spectral values of the surfaces. Therefore, its presence causes a deterioration in the visual image's quality and limits the information that the former could give. Ignoring the existence of shadows in images may cause serious problems in various visual processing applications such as false objects detection. In this context, many researches have been conducted through years. However, it is still a challenge for analysts all over the world to find a fully automated and efficient method for shadow removal from images.