Recognition Methodology Research Articles

Method of Digitally Presented Texts Extraction from the Image of a Document and their Examination

In this work, the problems of extracting textual information from a document image are solved. It is important to note that the work includes complex use of special knowledge in three areas: technical and forensic examination of documents; phototechnical expertise and computer-technical expertise. The article considers and demonstrates in detail the information for detecting changes in textual information (ink, toner and writing instrument pastes) from digital images. The commission of crimes in the economic sphere, as a rule, is associated with the falsification (forgery) of documents, and the increase in the document flow in the regulation of legal relations, which has been noticeable in recent years, has led to a significant increase in the number of forged documents. This is confirmed by statistical data on forensic handwriting and technical forensic examinations. Since the document is the basis of all human activity, and it contains data drawn up in a certain order and having legal significance. The problem has got many interesting solutions thanks to practical applications such as text recognition, translation, forensics and etc. Despite the fact that this problem has been dealt with for a long time, it is still missing, i.e. there is no well-developed methodology for image recognition of documents. The image of text information depends on the type of scene, type of digital camera, amount and direction of illumination, type of substrate and etc. It is important to note here that when obtaining information (from the document under study) in modern conditions, it is possible to fix the document using any device (camera, smartphone, scanner and etc.) and in any conditions (day/night, street, radiation source) that are unknown in advance. Proceeding from this, and this is a problem in modern conditions, in the study of electronic digital information of text documents. In this work, a study to solve this problem was carried out.

Iris Recognition System Based on Zak-Gabor Wavelet Packets

The paper proposes a new iris coding method based on Zak-Gabor wavelet packet transform. The essential component of the iris recognition methodology design is an effective adaptation of the transformation parameters that makes the coding sensitive to the frequencies characterizing ones eye. We thus propose to calculate the between-to-within class ratio of weakly correlated Zak-Gabor transformation coefficients allowing for selection the frequencies the most suitable for iris recognition. The Zak-Gabor-based coding is non-reversible, i.e., it is impossible to reconstruct the original iris image given the iris template. Additionally, the inference about the iris image properties from the Zak-Gabor-based code is limited, providing a possibility to embed the biometric replay attack prevention methodology into the coding. We present the final prototype system design, including the hardware, and evaluate its performance using the database of 720 iris images.

AGORA: An intelligent system for the anonymization, information extraction and automatic mapping of sensitive documents

Public institutions, such as law enforcement agencies or health centers, have a vast volume of unstructured text documents, e.g. police reports. Currently, before this data can be shared (e.g. with research institutions), it must go through a lengthy and costly human anonymization procedure.This paper addresses this issue by presenting AGORA, a cutting-edge tool that automatically identifies key entities and anonymizes sensitive data in text documents. AGORA has been developed in partnership with the Spanish National Office Against Hate Crimes and validated in the police and medical domains. This tool allows to export both anonymized texts and identified entities to structured files, thus, simplifying its exploitation for analysis purposes. Also, AGORA is capable of plotting the location entities identified in the documents, as well as obtaining and displaying relevant information from their geographical surroundings. Thus, it simplifies the task of generating comprehensive datasets for subsequent data analysis or predictive tasks. Its main goal is to foster cooperation between public institutions and research centers by easing document sharing as well as serving as a foundation for addressing succeeding phases in data science.The paper conducts a comprehensive assessment of the literature on Named Entity Recognition methodologies and technologies. Followed by extensive computational experiments to identify the best configuration for the NER models embedded in AGORA which include both successful state-of-the-art model setups and novelly proposed ones. Finally, the methodology, conclusions and software provided can be easily reused in similar application scenarios.

Revealing Intramolecular Isotope Effects with Chemical-Bond Precision.

Isotope substitution of a molecule not only changes its vibrational frequencies but also changes its vibrational distributions in real-space. Quantitatively measuring the isotope effects inside a polyatomic molecule requires both energy and spatial resolutions at the single-bond level, which has been a long-lasting challenge in macroscopic techniques. By achieving ångström resolution in tip-enhanced Raman spectroscopy (TERS), we record the corresponding local vibrational modes of pentacene and its fully deuterated form, enabling us to identify and measure the isotope effect of each vibrational mode. The measured frequency ratio νH/νD varies from 1.02 to 1.33 in different vibrational modes, indicating different isotopic contributions of H/D atoms, which can be distinguished from TERS maps in real-space and well described by the potential energy distribution simulations. Our study demonstrates that TERS can serve as a non-destructive and highly sensitive methodology for isotope detection and recognition with chemical-bond precision.

"Lighting up" fluorescence precise recognition of Al3+ with an effective fluorescence detection using a Bisphenol A-based sensor.

Although aluminum is a ubiquitous metal in the ecosystem and has numerous critical roles in both the medicinal and biological fields, human daily life is seriously threatened by its assorted harmful influences. By this virtue, tracking the amount of aluminum byrapid sensitive and selective recognition methodologies is of great importance. Based on this, a novel fluorescent chemosensor 4,4'-(propane-2,2-diyl)bis(2-(((-2-hydroxybenzylidene) hydrazineylidene)-methyl)phenol) (BFASA) capable of recognizing Al3+ in a medium was constructed via an easy Schiff-base reaction between bisphenol A-containing molecule and the salicylaldehyde. The metal-binding studies of BFASA indicated a drastically enhanced emission with color alteration from colorless to green establishing the utility of BFASA against monitoring of Al3+ and only Cu2+/Al3+ significantly enhanced the absorbance intensity of the probe solution at 433 and 406, respectively. Its ability to selectively sense Al3+ demonstrated "switch-on" fluorescence responses for Al3+ with a low detection limit (LOD) of 0.56μM and good selectivity, and pH adaptation range (5-8). The stoichiometric ratio of BFASA against the Al3+ was verified by the Job's plot and TOF-MS analysis and determined as 1:2. To make the recognition process inexpensively, viable and straightforward, Smartphone application of BFASA was effectively applied to Al3+ sensing, which could benefit the on-site Al3+ recognition. In the fluorescence bio-imaging aspect, the BFASA could effectively monitor Al3+ in living cells.

Sign Language Detection Using Deep Learning

Abstract: People with hearing impairments are found worldwide; therefore, the development of effective local level sign language recognition (SLR) tools is essential. We conducted a comprehensive review of automated sign language recognition based on machine/deep learning methods and techniques published between 2014 and 2021 and concluded that the current methods require conceptual classification to interpret all available data correctly. Thus, we turned our attention to elements that are common to almost all sign language recognition methodologies. This paper discusses their relative strengths and weaknesses, and we propose a general framework for researchers. This study also indicates that input modalities bear great significance in this field; it appears that recognition based on a combination of data sources, including vision-based and sensorbased channels, is superior to a unimodal analysis. In addition, recent advances have allowed researchers to move from simple recognition of sign language characters and words towards the capacity to translate continuous sign language communication with minimal delay. Many of the presented models are relatively effective for a range of tasks, but none currently possess the necessary generalization potential for commercial deployment. However, the pace of research is encouraging, and further progress is expected if specific difficulties are resolved.

Recognition of Children’s Facial Expressions Using Deep Learned Features

The emotional well-being of a child is crucial for their successful integration into society as a productive individual. While technology has made significant strides in enabling machines to decipher human emotional signals, current research in emotion recognition primarily prioritizes adults, disregarding the fact that children develop emotional awareness at an early stage. This highlights the need to explore how machines can recognize facial expressions in children, although the absence of a standardized database poses a challenge. In this study, we propose a system that employs Convolutional-Neural-Network (CNN)-based models, such as VGG19, VGG16, and Resnet50, as feature extractors, and Support Vector Machine (SVM) and Decision Tree (DT) for classification, to automatically recognize children’s expressions using a video dataset, namely Children’s Spontaneous Facial Expressions (LIRIS-CSE). Our system is evaluated through various experimental setups, including 80–20% split, K-Fold Cross-Validation (K-Fold CV), and leave one out cross-validation (LOOCV), for both image-based and video-based classification. Remarkably, our research achieves a promising classification accuracy of 99% for image-based classification, utilizing features from all three networks with SVM using 80–20% split and K-Fold CV. For video-based classification, we achieve 94% accuracy using features from VGG19 with SVM using LOOCV. These results surpass the performance of the original work, which reported an average image-based classification accuracy of 75% on their LIRIS-CSE dataset. The favorable outcomes obtained from our research can pave the way for the practical application of our proposed emotion recognition methodology in real-world scenarios.

A critical assessment of a protected area conflict analysis based on secondary data in the age of datafication

Recently, a global trend towards a broader use of secondary data in social sciences has been reinforced by the COVID-19 pandemic. This evoked doubts about the validity of the results unless restrictive assessment procedures are implemented. To address this need in the field of protected area (PA) conflict analysis, we propose a three-fold approach (theory-, method-, and cross-scale simulation-driven) to assess the usefulness of the utilized state register dataset and the indicator analysis methodology for the multi-level recognition of PA conflict determinants. With the ultimate aim to inform case study selection, we processed 187 relevant indicators from the official Statistics Poland register for a Lesser Poland region. We distinguished five types of PA conflict determinants in Lesser Poland (‘urbanity’, ‘agriculture’, ‘tourism’, ‘small-scale entrepreneurship’, and ‘sprawl’) and respective groups of 15 clusters comprising local-level units. For one cluster, we juxtaposed the obtained results with secondary data from another source (Internet content) and for a specific PA (Tatra National Park). Although the reported conflict issues corresponded to the indicator-derived descriptors of the cluster, in the theory-driven phase of the assessment, the state register failed to address the key prerequisites of PA conflicts. We have demonstrated that, in crisis conditions such as COVID-19, the proposed method can serve as a proxy for a multi-level recognition of PA conflict potentials, provided that it synthesises the results of different methodological approaches, followed by in-person interviews in the selected case studies.

Named Entity Recognition for English Language Using Deep Learning Based Bi Directional LSTM-RNN

The NER has been important in different applications like data Retrieval and Extraction, Text Summarization, Machine Translation, Question Answering (Q-A), etc. While several investigations have been carried out for NER in English, a high-accuracy tool still must be designed per the Literature Survey. This paper suggests an English Named Entities Recognition methodology using NLP algorithms called Bi-Directional Long short-term memory-based recurrent neural network (LSTM-RNN). Most English Language NER systems use detailed features and handcrafted algorithms with gazetteers. The proposed model is language-independent and has no domain-specific features or handcrafted algorithms. Also, it depends on semantic knowledge from word vectors realized by an unsupervised learning algorithm on an unannotated corpus. It achieved state-of-the-art performance in English without the use of any morphological research or without using gazetteers of any sort. A little database group of 200 sentences includes 3080 words. The features selection and generations are presented to catch the Name Entity. The proposed work is desired to forecast the Name Entity of the focus words in a sentence with high accuracy with the benefit of practical knowledge acquisition techniques.

Evaluation of interpretability for deep learning algorithms in EEG emotion recognition: A case study in autism

Current models on Explainable Artificial Intelligence (XAI) have shown a lack of reliability when evaluating feature-relevance for deep neural biomarker classifiers. The inclusion of reliable saliency-maps for obtaining trustworthy and interpretable neural activity is still insufficiently mature for practical applications. These limitations impede the development of clinical applications of Deep Learning. To address, these limitations we propose the RemOve-And-Retrain (ROAR) algorithm which supports the recovery of highly relevant features from any pre-trained deep neural network. In this study we evaluated the ROAR methodology and algorithm for the Face Emotion Recognition (FER) task, which is clinically applicable in the study of Autism Spectrum Disorder (ASD). We trained a Convolutional Neural Network (CNN) from electroencephalography (EEG) signals and assessed the relevance of FER-elicited EEG features from individuals diagnosed with and without ASD. Specifically, we compared the ROAR reliability from well-known relevance maps such as Layer-Wise Relevance Propagation, PatternNet, Pattern-Attribution, and Smooth-Grad Squared. This study is the first to bridge previous neuroscience and ASD research findings to feature-relevance calculation for EEG-based emotion recognition with CNN in typically-development (TD) and in ASD individuals.

QSLRS-CNN: Qur'anic sign language recognition system based on convolutional neural networks

ABSTRACT Deaf and dumb Muslims face educational barriers. They can't read, recite, or comprehend the Holy Qur'an, hence they can't practise Islamic ceremonies. This study proposes a CNN-based Qur'anic sign language recognition methodology. First, photos are used to train for dynamic and static gesture recognition. Second, preparing images diversifies datasets. Finally, CNN-based deep learning models extract and classify features. To teach the deaf and dumb Islamic ceremonies, the programme recognises Arabic sign language hand motions referring to dashed Qur'anic letters. Only 24,137 photos of the Holy Qur'an's 14 dashed letters were used in the trials from ArSL2018, a huge Arabic sign language collection. SMOTE raises training and testing accuracy to 98.31% and 97.67%, respectively, whereas the proposed model reaches 98.05% and 97.13%. RMU obtains 98.66% and 97.52% training and testing accuracy, whereas RMO achieves 98.37% and 97.36%.

Sign Language Recognition using Deep Learning through LSTM and CNN

This study presents the application of using deep learning to detect, recognize and translate sign language. Understanding sign language is crucial for communication between the deaf and mute people and the general society. This helps sign language users to easily communicate with others, thus eliminating the differences between both parties. The objectives of this thesis are to extract features from the dataset for sign language recognition model and the formulation of deep learning models and the classification performance to carry out the sign language recognition. First, we develop methodology for an efficient recognition of sign language. Next is to develop multiple system using three different model which is LSTM, CNN and YOLOv5 and compare the real time test result to choose the best model with the highest accuracy. We used same datasets for all algorithms to determine the best algorithm. The YOLOv5 has achieved the highest accuracy of 97% followed by LSTM and CNN with 94% and 66.67%.

Premature Ventricular Contraction Recognition using a Fuzzy Maximum Approaching Degree

This work presents a new methodology for ventricular premature contraction arrhythmias recognition using a set of geometrical attributes recently proposed and a fuzzy maximum approaching degree. Pattern models based on triangular and trapezoidal membership functions are proposed and a committee comprising these functions is composed using some statistical data, beyond a mechanism for manual selection of attributes and automatic weighting for each attribute. The obtained results show the efficiency and validity of the proposed approach, with 99.07%, 98.36% and 99.79% of accuracy, sensibility and specificity, respectively, as good as the ones obtained by the state-of-art methods.

CNV Deep Learning based Methodology for Recognition

Background: Copy number variants (CNVs) account for a significant amount of genetic variation. Since many CNVs include genes that result in differential levels of gene expression, substantial normal phenotypic variation can be explained. Current efforts are directed toward a more comprehensive characterization of CNVs that will provide the basis for determining how genomic diversity impacts biological function, evolution and common diseases in human as well as plants. Methods: The analytical variability in next generation sequencing (NGS) and artifacts in coverage data along with lack of robust bioinformatics tools for CNV detection have limited the utility of targeted NGS data to identify CNVs. Literature has the evidence of development of deep learning-based pipeline that incorporates a machine learning component to identify CNVs from targeted NGS data. Result: It is believed that combining this with clinical “gold standard” (e.g. FISH) information, the CNV detection could be more accurate. This would lead to a new research direction, supplementing the existing NGS methods.

Multimodal emotion recognition using cross modal audio-video fusion with attention and deep metric learning

In the last few years, the multi-modal emotion recognition has become an important research issue in the affective computing community due to its wide range of applications that include mental disease diagnosis, human behavior understanding, human machine/robot interaction or autonomous driving systems. In this paper, we introduce a novel end-to-end multimodal emotion recognition methodology, based on audio and visual fusion designed to leverage the mutually complementary nature of features while maintaining the modality-specific information. The proposed method integrates spatial, channel and temporal attention mechanisms into a visual 3D convolutional neural network (3D-CNN) and temporal attention into an audio 2D convolutional neural network (2D-CNN) to capture the intra-modal features characteristics. Further, the inter-modal information is captured with the help of an audio-video (A-V) cross-attention fusion technique that effectively identifies salient relationships across the two modalities. Finally, by considering the semantic relations between the emotion categories, we design a novel classification loss based on an emotional metric constraint that guides the attention generation mechanisms. We demonstrate that by exploiting the relations between the emotion categories our method yields more discriminative embeddings, with more compact intra-class representations and increased inter-class separability. The experimental evaluation carried out on the RAVDESS (The Ryerson Audio-Visual Database of Emotional Speech and Song), and CREMA-D (Crowd-sourced Emotional Multimodal Actors Dataset) datasets validates the proposed methodology, which leads to average accuracy scores of 89.25% and 84.57%, respectively. In addition, when compared to state-of-the-art techniques, the proposed solution shows superior performances, with gains in accuracy ranging in the [1.72%, 11.25%] interval.

Autonomous Underwater Environment Perceiving and Modeling: An Experimental Campaign With FeelHippo AUV for Forward Looking Sonar-Based Automatic Target Recognition and Data Association

Seabed inspection is one of the most sought-after applications for autonomous underwater vehicles (AUVs). Acoustical sensors, such as side-scan sonars and forward-looking sonars (FLSs), are commonly favored over optical cameras to carry out such a task. Indeed, sonars are not influenced by illumination conditions and can provide high-range data. However, due to the lack of features and low resolution, acoustical images are often hard to interpret with conventional automatic techniques, forcing human operators to analyze thousands of collected images to identify the so-called objects of potential interest (OPIs). In this article, we report the development of an automatic target recognition (ATR) methodology to identify and localize OPIs in FLS imagery. Such detections have been then exploited to realize a virtual world model with the probabilistic multiple hypothesis anchoring data association and model tracking algorithm. Distinct models of convolutional neural networks have been trained with a data set acquired in May 2019 at the Naval Support and Experimentation Centre (Centro di Supporto e Sperimentazione Navale—CSSN) basin in La Spezia, Italy. The ATR strategy has been successfully validated offline with the data gathered in October 2019 in the same site where the seabed targets were replaced and relocated. As regards the world modeling technique, it has been preliminarily tested on a simulated scenario built upon unmanned underwater vehicle Simulator. Finally, both the ATR and world modeling systems were on-field tested in October 2020 at the CSSN basin in a multivehicle architecture by employing an acoustical channel between FeelHippo AUV and an autonomous moving buoy.

FaceScan: Unmasking Identity Beyond the Surface

Biometrics, focusing on human behavior and features, encompasses face recognition as a vital technique. This paper offers a survey of face recognition methods, analyzing algorithms and approaches. Face recognition, crucial for security, is explored, emphasizing its computer-based recognition mechanisms. The paper provides an overview of face recognition methodology, recent techniques, and their efficacy under various conditions. Investigating potential biases in current deep learning systems, particularly towards children, reveals performance disparities. Strategies to mitigate this bias are evaluated, identifying effective fusion techniques. Recent advancements in face recognition, fueled by technological progress, have made automated systems practical and widely applicable. This review assesses the strengths, weaknesses, and current/future applications of face recognition technology, including IBM's system and its domains of application.

Multi-Scale Receptive Fields Convolutional Network for Action Recognition

Extracting good action representations from video frames is an intricate challenge due to the presence of moving objects of various sizes across current action recognition datasets. Most of the current action recognition methodologies have paid scant attention to this characteristic and have relied on deep learning models to automatically solve it. In this paper, we introduce a multi-scale receptive fields convolutional network (MSRFNet), which is fashioned after the pseudo-3D residual network architecture to mitigate the impact of scale variation in moving objects. The crux of MSRFNet is the integration of a multi-scale receptive fields block, which incorporates multiple dilated convolution layers that share identical convolutional parameters, but feature different receptive fields. MSRFNet leverages three scales of receptive fields to extract features from moving objects of diverse sizes, striving to produce scale-specific feature maps with a uniform representational power. Through visualization of the attention of MSRFNet, we analyze how the model re-allocates its attention to moving objects after implementing the multi-scale receptive fields approach. Experimental results on the benchmark dataset demonstrate that MSRFNet achieves improvement of 3.2% on UCF101, improvement of 5.8% on HMDB51, and improvement of 7.7% on Kinetics-400 compared with the baseline. Compared with state-of-the-art techniques, MSRFNet gets comparable or superior results, thereby affirming the effectiveness of the proposed approach.

An Optimum Classifier Model with Fuzzy C-Means for Fire Detection Technology

Flames recognition methodology is most important for completely diminishing the flame losses in different fired environmental conditions. However, there is delayed detection and lower accuracy in the various common detection methods. Thus, optimum image/video fire detection technology is proposed in this paper based on a support vector machine (SVM) with the fuzzy c-mean, discrete wavelet transform (DWT), and gray level co-occurrence matrices (GLCM) feature extraction for the detection of fires. This algorithm has been tested on various fire and non-fire images for classification accuracy. A performance evaluation of the proposed classifier algorithm and existing algorithms is compared, showing that the accuracy and other metrics of the proposed classifier algorithm are higher than other algorithms. Furthermore, simulation results show that the proposed classifier model is improved the forecast detection accuracy of fires.

Development and Evaluation of a Small-Scale Apple Sorting Machine Equipped with a Smart Vision System

One of the most important matters in international trades for many local apple industries and auctions is accurate fruit quality classification. Defect recognition is a key in online computer-assisted apple sorting machines. Because of the cavity structure of the stem and calyx regions, the system tends to mistakenly treat them as true defects. Furthermore, there is no small-scale sorting machine with a smart vision system for apple quality classification where it is needed. Thus, the current study focuses on a highly accurate and feasible methodology for stem and calyx recognition based on Niblack thresholding and a machine learning technique using k-nearest neighbor (k-NN) classifiers associated with a locally designed small-scale apple sorting machine. To find an appropriate mode, the effects of different numbers of k and metric distances on stem and calyx region detection were evaluated. Results showed the effectiveness of the value of k and Euclidean distances in recognition accuracy. It is found that the 5-nearest neighbor classifier and the Euclidean distance using 80 training samples produced the best accuracy rates, at 100% for stem and 97.5% for calyx. The significance of the result is very promising in fabricating an advanced small-scale and low-cost sorting machine with a high accuracy for the horticultural industry.