Pattern Recognition Applications Research Articles

Ship trajectory segmentation and semisupervised clustering via geospatial background knowledge

Cluster analysis is a common application of maritime traffic pattern recognition. However, most current methods are unsupervised clustering algorithms that rely solely on the features of the trajectories themselves for clustering, making them susceptible to uneven density distributions. To address this issue, we propose a semisupervised clustering algorithm that utilizes geospatial data to segment trajectories and generate clustering constraints. First, shoreline and anchorage information are employed for trajectory segmentation to enhance the consistency of local trajectory features. Second, cannot-link constraints derived from shoreline data are used to modify the distance between two trajectories, thereby preventing the clustering of trajectories separated by nonnavigable waters into a single cluster. Last, experimental verification is conducted using automatic identification system data, and the results are compared with those of existing algorithms. The results demonstrate more concentrated trajectory segmentation points, easier selection of clustering parameters, and significantly improved evaluation outcomes based on external clustering validity indices.

Lit me up: A reference free adaptive low light image enhancement for in-the-wild conditions

Images captured with different devices in uneven conditions (e.g., invariable lighting, low lighting, weather changes, exposure time, etc.) often lead to low image visibility and poor color and contrast, affecting the performance of computer vision and pattern recognition applications. The pre-trained convolutional neural networks (CNNs) solely rely on the training data and lack adaptation due to the uncertainty in the lighting conditions. Moreover, capturing large-scale datasets to train CNNs also raises the computational complexity and overall cost. This work integrates the knowledge and data and proposes a two-stage Uneven-to-Enliven network (U2E-Net), which rapidly learns to see in uneven conditions. A multiple-layered Uneven network learns to distinguish the reflection and illumination in the input images, and an encoder–decoder-based Enliven-Net contextualizes the illumination information. A key component in such ill-posed problems is to obtain information from priors and pairs; however, we present the compelling idea of information trade-off followed by decomposition consistency, thereby progressively improving the visual quality with the subsequent enhancement operations. To this end, we proposed a two-faceted framework that can work without depending on the data type. A novel color and contrast preservation strategy (CPS) is proposed following the decomposition of input data. CPS is integrated within the network to extract contrast in the darkest background regions.

Early detection of skin cancer using AI: Deciphering dermatology images for melanoma detection

This Review explores the transformative impact of artificial intelligence (AI) on the early detection of skin cancer, with a specific focus on melanoma, a potentially lethal form of the disease. Beginning with an overview of traditional diagnostic methods and their limitations, this paper delves into the evolution of AI within dermatology, emphasizing its application in image analysis and pattern recognition. A comprehensive examination of AI algorithms for melanoma detection, including machine learning and deep learning models, is provided. This Review critically assesses the performance metrics, training datasets, and comparative analyses with traditional methods. Addressing challenges such as data quality, interpretability, and ethical considerations, this paper outlines future directions, emphasizing ongoing research, algorithm improvements, and integration with clinical practices. Case studies and success stories highlight the real-world impact of AI in dermatology. This Review concludes by summarizing key findings and underlining the pivotal role of AI in revolutionizing early melanoma detection, with implications for personalized medicine and enhanced patient outcomes.

A novel knowledge-based similarity measure on intuitionistic fuzzy sets and its applications in pattern recognition

Similarity measure is a useful tool to determine the similarity of two intuitionistic fuzzy sets (IFSs). Theoretically, there are two free variables of membership and non-membership on IFSs, and the most fuzzy set is not solely. Therefore, it is impossible to establish a similarity measure satisfying the classical axiom on it by using only the distance between any given IFSs and the assumed most fuzzy set. The motivation and innovation of this study lies in the exploration and use of the two most fuzzy sets, and the weighted average distance between IFSs and the two most fuzzy sets is used to define the similarity measure. Based on this analysis, a knowledge-based similarity measure on IFSs is proposed and some applications of this proposed measure in pattern recognition are introduced. The advantage of the introduced similarity measure is that it calculates the dissimilarity between the complementary sets well. Furthermore, it proves that the novel similarity measure satisfies axioms of the similarity measure on IFSs, and overcomes the drawbacks of the existing measures. Finally, a series of examples in pattern recognition fields are introduced to demonstrate the effectiveness of the proposed similarity measure.

Neural Networks Based on Synthesized Training Data for the Automated Detection of Chemical Plumes in Passive Infrared Multispectral Images.

Automated detection of volatile organic compounds in the atmosphere can be achieved by applying pattern recognition analysis to passive infrared (IR) multispectral remote sensing data. However, obtaining analyte-active training data through field experiments is time-consuming and expensive. To address this issue, methodology has been developed for simulating radiance profiles acquired using a multispectral IR line-scanner mounted in a downward-looking position on a fixed-wing aircraft. The simulation strategy used Planck's radiation law and a radiometric model along with the laboratory spectrum of the target compound to compute the upwelling IR background radiance with the presence of the analyte within the instrumental field-of-view. By combining the simulated analyte-active radiances and field-collected analyte-inactive radiances, a synthetic training dataset was constructed. A backpropagation neural network was employed to build classifiers with the synthetic training dataset. Employing methanol as the target compound, the performance of the classifiers was evaluated with field-collected data from airborne surveys at two test fields.

Improving ellipse fitting via multi-scale smoothing and key-point searching

Fast and efficient fitting of accurate ellipses from data points has many applications in pattern recognition, machine vision, and robotics. However, the fitting accuracy may significantly degrade in the existence of outliers, such as the least-squares-based approaches. Despite robust methods attaining more accurate results than the least-squares manner under the contamination of outliers, they typically require the careful tuning of the hyper-parameters for good results. To mitigate the outlier disturbance, in this paper, we propose a conceptually simple yet quite useful preprocessing framework for high-precision ellipse fitting. Firstly, we leverage multi-scale operators to shrink the input image, by which a large number of outliers can be removed, followed by the smoothing of the sub-image to further improve the data quality. Then, we propose a key-point searching method to enhance the fitting precision via the analysis of the discrete pixel data in images. We prove that key-point-based ellipse fitting gives the upper bound of the approximation error generated by other sampled points with the same ellipse. Based on the key-point pairs inside and outside the ellipse, we further calculate their barycentric points and then perform fitting on these points to attain high-precision ellipses. We conduct extensive experiments on synthetic and real-world images to validate the proposed method and compare it with representative state-of-the-art approaches. Quantitative and qualitative results demonstrate that our method has more accurate and robust performance than competitors. Additionally, we employ the proposed method to compared approaches as a preprocessing step. Experiments demonstrate that our method is effective to significantly improve their fitting accuracy. Our source code is freely available at https://github.com/ChengQian09/MSKPF.

Population dynamic study of two prey one predator system with disease in first prey using fuzzy impulsive control

Abstract Objectives The prey-predator model provides a mathematical framework for understanding the population dynamics of interacting species, highlighting the delicate balance between predator and prey populations in ecological systems. The four-species predator-prey model extends the Lotka-Volterra framework to explore the dynamics of ecosystems with multiple interacting species. It provides a theoretical foundation for understanding how the populations of multiple prey and predator species influence each other over time. Apart from the traditional methods like direct approach for solving the non-linear system of equations, recent Fuzzy method approaches have been developed. The solution of non-linear systems using classical methods is not easy due to its non-linearity, analytical complexity, chaotic behavior, etc. and the T-S method is very much effective to analyze the non-linear models. Methods In this study, we considered an eco-epidemic model with two populations of prey and one population of predator, with the only infectious disease infecting the first prey population. The four-dimensional Lotka-Volterra predator-prey system’s model stability has been examined using the Takagi-Sugeno (T-S) impulsive control model and the Fuzzy impulsive control model. Following the formulation of the model, the global stability and the Fuzzy solution are carried out through numerical simulations and graphical representations with appropriate discussion for a better understanding the dynamics of our proposed model. Results The Takagi-Sugeno method has diverse applications in modeling, control, pattern recognition, and decision-making in systems where uncertainty and non-linearity play a significant role. Its ability to combine fuzzy logic with traditional mathematical models provides a powerful tool for addressing complex real-world problems. Conclusions The impulse control approach, what is considered within the foundation of fuzzy systems established on T-S model, is found to be suitable for extremely complex and non-linear systems with impulse effects.

Switchable Two-Dimensional AND and Exclusive OR Operation Based on Dual-Wavelength Metasurfaces.

Logic operation serves as the foundation and core element of computing networks; it will bring huge vitality to advanced information processing with its adaptation in the optical domain. As fundamental logic operations, AND and exclusive OR (XOR) operations serve a multitude of purposes, such as their ability to cooperate in enabling image processing and interpretation. Here, we propose and experimentally demonstrate a wavelength multiplexed AND and XOR function based on metasurfaces. By combining two cosine gratings with distinct frequencies and an initial phase difference of π/2, we extract the similarities and differences between two input images simultaneously by illuminating them with 445 and 633 nm wavelengths. Additionally, we explore its potential in information encryption, where overall security is enhanced by distributing distinct parts of initial information and encoded keys to different receivers. This design possesses the benefits of convenient mode switching and high-quality imaging, facilitating advanced applications in pattern recognition, machine vision, medical diagnosis, etc.

Novel similarity measure between hesitant fuzzy set and their applications in pattern recognition and clustering analysis

The extension of classical fuzzy sets are hesitant fuzzy sets (HFSs), in which each element has a possible value from [0,1]. Similarity and distance measures are useful implements for solving medical, clustering and pattern-recognition problems. Most of the researchers have suggested their ideas for HFSs using distance measures and extract the similarity measure from distance measure but most of them are getting inadequate results. Therefore, we proposed a new similarity measure to resolve these problems and also satisfied the properties of proposed measure for HFSs. Additionally, numerous examples are taken in consideration using HFS and compared the performance of existing measures with proposed measure for different cases. Furthermore, we have applied proposed measure for pattern recognition problems using three different examples and also calculate performance index (i.e., Degree of Confidence) to explore the behavior of different measures. Finally, we suggested MST based clustering algorithm using HF-environment and contrast the performance of proposed measure with existing ones. All these comparison illustrate that proposed measure is getting efficient and reasonable results and it also verified that proposed measure is not restricted to particular domain, it can be effectively applied for diverse field of application.

Nonlinear Dynamics of Neural Networks: Applications in Pattern Recognition

Nonlinear Dynamics of Neural Networks: Applications in Pattern Recognition

Development of a low-cost, compact, wireless, 16 - channel biopotential data acquisition, signal conditioning and arbitrary waveform stimulator.

The health and fitness of the human body rely heavily on physiological parameters. These parameters can be measured using various tools such as ECG, EMG, EEG, EOG, among others, to obtain real-time physiological data. Analysing the bio-signals obtained from these measurements can provide valuable information that can be used to improve health-care in terms of observation, diagnosis, and treatment. In bio-signal pattern recognition applications, more channels provide multiple information simultaneously. Different biosignal acquisition devices are available in the market, most of which are designed for specific signals like ECG, EMG, EEG etc The gain of the amplifiers and frequency of the filters are designed as per the targeted signals; due to which one device cannot be used for other signals. Also, most of the systems are wired system which is not comfortable for animal studies. In this paper, a low-cost, compact, wireless, 16 channel biopotential data acquisition system with integrated electrical stimulator is designed and implemented. There are several novel and flexible design approaches were incorporated in the proposed design like (1) It has user selectable digital filter in each channel based on the signal frequencies like ECG, EMG, EEG, EOG. The same system will be used to acquire different signals simultaneously. (2) It has variable gain with a configurable analog bandpass filter. (3) It can acquire signals from 4 patients simultaneously. (4) The system is capable to acquire signal from both two-electrode as well as three-electrode configurations. (5) It has integrated stimulator with trapezoidal, charge-balanced, biphasic stimulus output with near zero DC level and user selectable pulse duration or frequency of the stimulus. The developed system has the ability to acquire and transmit data wirelessly in real-time at a high transfer rate. To validate the performance of the system, tests were conducted on the acquired signals using a simulator.

On the Work of the Institute of Control Sciences of the Russian Academy of Sciences in the Field of Pattern Recognition Theory and Applications in the 20th Century

On the Work of the Institute of Control Sciences of the Russian Academy of Sciences in the Field of Pattern Recognition Theory and Applications in the 20th Century

The Use of Contrast and Gradient Features to Categorize Texture Images

Image texture is an important part of many types of images, for example medical images. Texture Analysis is the technique that uses measurable features to categorize complex textures. The main goal is to extract discriminative features that are used in different pattern recognition applications and texture categorization. This paper investigates the extraction of most discriminative features for different texture images from the “Colored Brodatz” dataset using two types of image contrast measures, as well as using the statistical moments on five bands (red, green, blue, grey, and black). The Euclidean distance measure is used in the matching step to check the similarity degree. The proposed method was tested on 112 classes of textures. The achieved results showed that the proposed method is accurate and fast concerning classification accuracy with low computational complexity. The achieved Recognition Rate (RR) was 100%.

Application of The Speed-Up Robust Features Method To Identify Signature Image Patterns On Single Board Computer

Through the development of a signature pattern recognition program on SBC Beagle-bone Black, this research seeks to determine how to differentiate between real and false signatures. Three techniques of gathering data were employed in this study: interviews, observations, and a review of the literature. The quick application development method is the approach that is applied. The rapid, efficient, and brief development cycle (RAD) is emphasized. This study uses a use-case diagram to illustrate the application's logic and data flow. In this study, OpenCV is used as a digital image processing library along with the C++ programming language and QT creator as an integrated development environment (IDE). This application was subjected to both accuracy and functional testing. The following conclusions are drawn from the findings of the investigation and testing that was done: Using the fast library approach for approximate nearest neighbors (FLANN) and the speeded-up robust features (SURF) feature extraction method, the signature pattern recognition program on the Beagle-bone black SBC can differentiate between real and fraudulent signatures. Through the processes of generating image scale space, feature localization, and feature description, the SURF approach extracts feature from signature images. This signature pattern recognition application is one of the digital image processing apps that can be run on the Beagle-bone Black single board computer. This indicates that the specifications of the SBC Beagle-bone Black for digital image processing are good.

Similarity Measure of Plithogenic Cubic Vague Sets: Examples and Possibilities

The crisp, fuzzy, intuitionistic fuzzy, and neutrosophic sets are the extension of the plithogenic set, in which elements are characterized by the number of attributes and each attribute can assume many values. To achieve more accuracy and precise exclusion, a contradiction or dissimilarity degree is specified between each attribute and the values of the dominating attribute. A plithogenic cubic vague set is a combination of a plithogenic cubic set and a vague set. The key tool for resolving problems with pattern recognition and clustering analysis is the similarity measure. In this research, we characterize and investigate the similarities between two Plithogenic Cubic Vague sets (PCVSs) for (z≡F), (z≡IF) and (z≡N). Also, examples are given to examine similarities in the pattern recognition application problems.

FPGA Realization of a High-Speed Spiking Neural Network with Modified LIF Neurons for Pattern Recognition

This paper presents the multiplier-less Modified Leaky Integrate and Fire Neuron Unit (MLIFNU) and proposed a modified Spiking Neural Network (SNN) architecture for Pattern Recognition application. Synthesis results of implementation on Field-Programmable Gate Array are presented. Accordingly, the maximum frequency of the MLIFNU and SNN are 361.991 MHz and 281.442 MHz, respectively.

Switching perspectives: Investigating the relationships and applications of q-fractional fuzzy sets with other fuzzy set classes in decision making

We explore switching techniques between q-fractional fuzzy sets (qFr sets) and various other classes of fuzzy sets to establish connections and provide a comprehensive framework. In particular, we examine the relationships between qFr sets and interval-valued fuzzy sets (IVFS), type 2 fuzzy sets(T2FS), intuitionistic fuzzy sets(IFS), Pythagorean fuzzy sets(PFS), q-rung orthopair fuzzy sets (q-ROFS), and linear diophantine fuzzy sets(LDFS). By examining these interconnections, we aim to understand better qFr sets and their applications in a wide range of fuzzy systems. It is possible to convert qFr sets into other fuzzy set models using the derived switching techniques, facilitating the utilization of existing methods and algorithms. The versatility of qFr sets, combined with the bridging techniques presented, holds promise for addressing complex problems in decision-making, pattern recognition, and other applications where uncertainty and imprecision play significant roles. Through case studies and practical applications, we illustrate the effectiveness and usefulness of the proposed switching techniques, showcasing their potential impact on real-world scenarios.

Evaluate the reliability of information sources using the non-parametric plausibility ReliefF algorithm for multi-source information fusion

How to evaluate the reliability of the information sources is a significant and open issue in the information fusion of Dempster-Shafer evidence theory (DST). We propose a new method, called the non-parametric plausibility ReliefF (NPReliefF) algorithm, to measure the reliability of information sources based on the criterion that sources providing more evidence aligned with “greater similarity within classes than between classes” are more reliable. The proposed method significantly improves the effect of normal BPAs and weakens the impact of abnormal BPAs on the fusion, even when abnormal BPAs are in the majority, thereby maximizing the correctness of the fusion results. Several numerical examples and an application of pattern recognition show the effectiveness and flexibility of the proposed methodology.

Coherent chord computation and cross ratio for accurate ellipse detection

This paper presents a new method for detecting ellipses in images, which has many applications in pattern recognition and robotic tasks. Previous approaches typically use sophisticated arc grouping strategies or calculate differential such as tangents, and thereby they are less efficient or more sensitive to noise. In this work, we present a novel ellipse detector, based on the simple yet effective chord computation, and on the projective invariant cross ratio, which achieves promising performance in both accuracy and efficiency. First, elliptical arcs are extracted by fast vector computations along with the removal of straight segments to speed up detection. Then, arcs from the same ellipse are grouped together according to the relative location and the intersecting chord constraints, both are on coherent chord computation without differential. Additionally, an efficient additive principle is applied to further accelerate the grouping process. Finally, a novel and robust verification by area-deduced cross ratio is introduced to pick out salient ellipses. Compared with predecessor methods, cross ratio is not only simple for computation, but also has invariant properties (used to discriminate ellipses). Extensive experiments on seven public datasets (including synthetic and real-world images) are implemented. The results highlight the salient advantages of the proposed method compared to state-of-the-art detectors: Easier to implementation, more robust against occlusion and noise, as well as attaining higher F-measure.

NNFSRR: Nearest Neighbor Feature Selection and Redundancy Removal Method for Nearest Neighbor Search in Microarray Gene Expression Data

INTRODUCTION: Gene expression data analysis is a critical aspect of disease prediction and classification, playing a pivotal role in the field of bioinformatics and biomedical research. High-dimensional gene expression datasets hold a wealth of information, but their effective utilization is hindered by the presence of irrelevant dimensions and noise. The challenge lies in extracting meaningful features from these datasets to enhance the accuracy of disease prediction and classification while maintaining computational efficiency.
 Feature selection is a crucial step in addressing these challenges, as it aims to identify and retain only the most informative characteristics from large high-dimensional microarray datasets. In the context of microarray gene expression data, characterized by its substantial dimensionality, selecting relevant features is essential for efficient nearest neighbor search, a fundamental component of various analytical tasks in bioinformatics and data mining.
 Existing feature selection methods in high-dimensional data often face issues related to the trade-off between search accuracy and computational efficiency. This paper introduces a novel approach, the Nearest Neighbor Feature Selection with Symmetrical Uncertainty-based Redundancy Removal (NNFSRR) method, designed to enhance the classification of microarray gene expression data through feature selection. The NNFSRR method focuses on reducing the dimensionality of the dataset by identifying and removing redundant features, allowing subsequent searches to operate solely on relevant dimensions.
 OBJECTIVES: The primary goal is to evaluate the NNFSRR method's effectiveness in improving nearest neighbor search in microarray gene expression datasets by reducing dimensionality. This method utilizes Symmetrical Uncertainty-based correlation between dimensions for feature selection and aims to enhance accuracy and efficiency compared to existing methods.
 METHODS: The NNFSRR method uses Symmetrical Uncertainty to identify and remove redundant features from microarray gene expression datasets. Reduced datasets are used for nearest neighbor search, improving accuracy and efficiency. Experiments are conducted using real-world datasets, and comparisons with existing methods are made based on search time and accuracy.
 RESULTS: The NNFSRR method demonstrates improved nearest neighbor search performance, outperforming basic brute force methods and existing feature selection techniques. Selected feature sets exhibit strong class associations while minimizing feature correlations, enhancing classification precision.
 CONCLUSION: In conclusion, the NNFSRR method presents a promising approach to address the challenges posed by high-dimensional gene expression data. It effectively reduces dimensionality, improves search accuracy, and enhances the efficiency of nearest neighbor search. Our experimental results demonstrate that this method outperforms existing techniques in terms of search time and accuracy, making it a valuable tool for applications in bioinformatics, data mining, pattern recognition, and biological information retrieval. The NNFSRR method holds the potential to advance our understanding of complex biological processes and support more accurate disease prediction and classification.