Search results for: data recognition

Authors: Ferah Tesfaye Admasu

Abstract:

Autism Spectrum Disorder (ASD) often presents significant communication challenges, particularly for non-verbal individuals who struggle to express their needs and emotions effectively. Assistive technologies (AT) have emerged as vital tools in enhancing communication abilities for this population. Recent advancements in artificial intelligence (AI) hold the potential to revolutionize the design and functionality of these technologies. This study explores the application of AI in developing intelligent, adaptive, and user-centered assistive technologies for non-verbal individuals with autism. Through a review of current AI-driven tools, including speech-generating devices, predictive text systems, and emotion-recognition software, this research investigates how AI can bridge communication gaps, improve engagement, and support independence. Machine learning algorithms, natural language processing (NLP), and facial recognition technologies are examined as core components in creating more personalized and responsive communication aids. The study also discusses the challenges and ethical considerations involved in deploying AI-based AT, such as data privacy and the risk of over-reliance on technology. Findings suggest that integrating AI into assistive technologies can significantly enhance the quality of life for non-verbal individuals with autism, providing them with greater opportunities for social interaction and participation in daily activities. However, continued research and development are needed to ensure these technologies are accessible, affordable, and culturally sensitive.

Keywords: artificial intelligence, autism spectrum disorder, non-verbal communication, assistive technology, machine learning

Procedia PDF Downloads 19

Authors: Aurelija Burinskienė, Dalė Dzemydienė, Arūnas Miliauskas

Abstract:

This research aims at developing the approach for more effective freight delivery and transportation process management. The road congestions and the identification of causes are important, as well as the context information recognition and management. The measure of many parameters during the transportation period and proper control of driver work became the problem. The number of vehicles per time unit passing at a given time and point for drivers can be evaluated in some situations. The collection of data is mainly used to establish new trips. The flow of the data is more complex in urban areas. Herein, the movement of freight is reported in detail, including the information on street level. When traffic density is extremely high in congestion cases, and the traffic speed is incredibly low, data transmission reaches the peak. Different data sets are generated, which depend on the type of freight delivery network. There are three types of networks: long-distance delivery networks, last-mile delivery networks and mode-based delivery networks; the last one includes different modes, in particular, railways and other networks. When freight delivery is switched from one type of the above-stated network to another, more data could be included for reporting purposes and vice versa. In this case, a significant amount of these data is used for control operations, and the problem requires an integrated methodological approach. The paper presents an approach for providing e-services for drivers by including the assessment of the multi-component infrastructure needed for delivery of freights following the network type. The construction of such a methodology is required to evaluate data flow conditions and overloads, and to minimize the time gaps in data reporting. The results obtained show the possibilities of the proposing methodological approach to support the management and decision-making processes with functionality of incorporating networking specifics, by helping to minimize the overloads in data reporting.

Keywords: transportation networks, freight delivery, data flow, monitoring, e-services

Procedia PDF Downloads 125

Authors: Silvia Moreno, Banu Iyisan, Hannes Gumz, Brigitte Voit, Dietmar Appelhans

Abstract:

Polymersomes are materials which are considered as artificial counterparts of natural vesicles. The nanotechnology of such smart nanovesicles is very useful to enhance the efficiency of many therapeutic and diagnostic drugs. Those compounds show a higher stability, flexibility, and mechanical strength to the membrane compared to natural liposomes. In addition, they can be designed in detail, the permeability of the membrane can be controlled by different stimuli, and the surface can be functionalized with different biological molecules to facilitate monitoring and target. For this purpose, this study demonstrates the formation of multifunctional and pH sensitive polymersomes and their functionalization with different reactive groups or biomolecules inside and outside of polymersomes´ membrane providing by crossing the membrane and docking/undocking processes for biomedical applications. Overall, they are highly versatile and thus present new opportunities for the design of targeted and selective recognition systems, for example, in mimicking cell functions and in synthetic biology.

Keywords: multifunctionalized, pH stimulus, controllable release, cellular uptake

Procedia PDF Downloads 318

Authors: Bency Jacob, S.B. Waykar

Abstract:

Degradations in historical document images appear due to aging of the documents. It is very difficult to understand and retrieve text from badly degraded documents as there is variation between the document foreground and background. Thresholding of such document images either result in broken characters or detection of false texts. Numerous algorithms exist that can separate text and background efficiently in the textual regions of the document; but portions of background are mistaken as text in areas that hardly contain any text. This paper presents a way to overcome these problems by a robust binarization technique that recovers the text from a severely degraded document images and thereby increases the accuracy of optical character recognition systems. The proposed document recovery algorithm efficiently removes degradations from document images. Here we are using the ostus method ,local thresholding and global thresholding and after the binarization training and recognizing the characters in the degraded documents.

Keywords: binarization, denoising, global thresholding, local thresholding, thresholding

Procedia PDF Downloads 342

Authors: Hafiza Javaria Ashraf, Xinghong Wang, Zhanghong Shi, Youming Hou

Abstract:

Insect’s innate immunity depends on a variety of defense responses for the recognition of invading pathogens. Pathogen recognition involves particular proteins known as pattern recognition receptors (PRRs). These PRRs interact with pathogen-associated molecular patterns (PAMPs) present on the surface of pathogens to distinguish between self and non-self. C-type lectins (CTLs) belong to a superfamily of PPRs which involved in insect immunity and defense mechanism. Rhynchophorus ferrugineus Olivier is a devastating pest of Palm cultivations in China. Although studies on R. ferrugineus immune mechanism and host defense have conducted, however, the role of CTL in immune responses of R. ferrugineus remains elusive. Here, we report RfCTL, which is a secreted protein containing a single-CRD domain. The open reading frame (ORF) of CTL is 226 bp, which encodes a putative protein of 168 amino acids. Transcript expression analysis revealed that RfCTL highly expressed in immune-related tissues, i.e., hemolymph and fat body. The abundance of RfCTL in the gut and fat body dramatically increased upon Staphylococcus aureus and Escherichia coli bacterial challenges, suggesting a role in defense against gram-positive and gram-negative bacterial infection. Taken together, we inferred that RfCTL might be involved in the immune defense of R. ferrugineus and established a solid foundation for future studies on R. ferrugineus CTL domain proteins for better understanding of insect immunity.

Keywords: biological invasion, c-type lectin, insect immunity, Rhynchophorus ferrugineus Oliver

Procedia PDF Downloads 157

Authors: R. Chinnaiyaprabhu, S. Bharanidharan, V. Dharsana, Rajalavanya

Abstract:

The concept of the Metaverse represents a potential evolution in the realm of cyberspace. In the early stages of Web 2.0, we observed a proliferation of online pseudonyms or 'nyms,' which increased the prevalence of fake accounts and made it challenging to establish unique online identities for various roles. However, in the era of Web 3.0, particularly in the context of the Metaverse, an individual's digital identity is intrinsically linked to their real-world identity. Consequently, actions taken in the Metaverse can carry significant consequences in the physical world. In light of these considerations, we propose the development of an innovative authentication system known as 'Metasec.' This system is designed to enhance security for digital assets, online identities, avatars, and user accounts within the Metaverse. Notably, Metasec operates as a password less authentication solution, relying on a multifaceted approach to security, encompassing device attestation, facial recognition, and pattern-based security keys.

Keywords: metaverse, multifactor authentication, security, facial recognition, patten password

Procedia PDF Downloads 65

Authors: Sonali Banrjee, Swarup Kumar Mitra, Aritra Acharyya

Abstract:

A piano note recognition method has been proposed by the authors in this paper. The authors have used a comprehensive method for onset detection of each note present in a piano piece followed by segmented short-time Fourier transform (STFT) for the identification of piano notes. The performance evaluation of the proposed method has been carried out in different harsh noisy environments by adding different levels of additive white Gaussian noise (AWGN) having different signal-to-noise ratio (SNR) in the original signal and evaluating the note detection error rate (NDER) of different piano pieces consisting of different number of notes at different SNR levels. The NDER is found to be remained within 15% for all piano pieces under consideration when the SNR is kept above 8 dB.

Keywords: AWGN, onset detection, piano note, STFT

Procedia PDF Downloads 159

Authors: Kevin Wongleedee

Abstract:

This research paper was aimed to examine the reasons associated with hotel employees’ loyalty. This was a case study of 200 hotel employees in Bangkok, Thailand. The population of this study included all hotel employees who were working in Bangkok during January to March, 2014. Based on 200 respondents who answered the questionnaire, the data were complied by using SPSS. Mean and standard deviation were utilized in analyzing the data. The findings revealed that the average mean of importance was 4.40, with 0.7585 of standard deviation. Moreover, the mean average can be used to rank the level of importance from each factor as follows: 1) salary, service charge cut, and benefits, 2) career development and possible advancement, 3) freedom of working, thinking, and ability to use my initiative, 4) training opportunities, 5) social involvement and positive environment, 6) fair treatment in the workplace and fair evaluation of job performance, and 7) personal satisfaction, participation, and recognition.

Keywords: hotel employees, loyalty, reasons, case study

Procedia PDF Downloads 401

Authors: M. Mueller, M. Kuehn, M. Voelker

Abstract:

In small and medium-sized enterprises (SMEs), the challenge is to create a well-grounded and reliable basis for process analysis, optimization and planning due to a lack of data. SMEs have limited access to methods with which they can effectively and efficiently analyse processes and identify cause-and-effect relationships in order to generate the necessary database and derive optimization potential from it. The implementation of digitalization within the framework of Industry 4.0 thus becomes a particular necessity for SMEs. For these reasons, the abstract presents an analysis methodology that is subject to the objective of developing an SME-appropriate methodology for efficient, temporarily feasible data collection and evaluation in flexible production and logistics systems as a basis for process analysis and optimization. The overall methodology focuses on retrospective, event-based tracing and analysis of material flow objects. The technological basis consists of Bluetooth low energy (BLE)-based transmitters, so-called beacons, and smart mobile devices (SMD), e.g. smartphones as receivers, between which distance data can be measured and derived motion profiles. The distance is determined using the Received Signal Strength Indicator (RSSI), which is a measure of signal field strength between transmitter and receiver. The focus is the development of a software-based methodology for interpretation of relative movements of transmitters and receivers based on distance data. The main research is on selection and implementation of pattern recognition methods for automatic process recognition as well as methods for the visualization of relative distance data. Due to an existing categorization of the database regarding process types, classification methods (e.g. Support Vector Machine) from the field of supervised learning are used. The necessary data quality requires selection of suitable methods as well as filters for smoothing occurring signal variations of the RSSI, the integration of methods for determination of correction factors depending on possible signal interference sources (columns, pallets) as well as the configuration of the used technology. The parameter settings on which respective algorithms are based have a further significant influence on result quality of the classification methods, correction models and methods for visualizing the position profiles used. The accuracy of classification algorithms can be improved up to 30% by selected parameter variation; this has already been proven in studies. Similar potentials can be observed with parameter variation of methods and filters for signal smoothing. Thus, there is increased interest in obtaining detailed results on the influence of parameter and factor combinations on data quality in this area. The overall methodology is realized with a modular software architecture consisting of independently modules for data acquisition, data preparation and data storage. The demonstrator for initialization and data acquisition is available as mobile Java-based application. The data preparation, including methods for signal smoothing, are Python-based with the possibility to vary parameter settings and to store them in the database (SQLite). The evaluation is divided into two separate software modules with database connection: the achievement of an automated assignment of defined process classes to distance data using selected classification algorithms and the visualization as well as reporting in terms of a graphical user interface (GUI).

Keywords: event-based tracing, machine learning, process classification, parameter settings, RSSI, signal smoothing

Procedia PDF Downloads 129

Authors: Silvia Boschetti, Jakub Binter, Robin Kopecký, Lenka PříPlatová, Jaroslav Flegr

Abstract:

A strict dichotomy was present between religion and science, but recently, cognitive science focusses on the impact of supernatural beliefs on cognitive processes such as pattern recognition. It has been hypothesized that cognitive and perceptual processes have been under evolutionary pressures that ensured amplified perception of patterns, especially when in stressful and harsh conditions. The pattern detection in religious and non-religious individuals after induction of negative, anxious mood shall constitute a cornerstone of the general role of anxiety, cognitive bias, leading towards or against the by-product hypothesis, one of the main theories on the evolutionary studies of religion. The apophenia (tendencies to perceive connection and meaning on unrelated events) and perception of visual patterns (or pateidolia) are of utmost interest. To capture the impact of culture and upbringing, a comparative study of two European countries, the Czech Republic (low organized religion participation, high esoteric belief) and Italy (high organized religion participation, low esoteric belief), are currently in the data collection phase. Outcomes will be presented at the conference. A battery of standardized questionnaires followed by pattern recognition tasks (the patterns involve color, shape, and are of artificial and natural origin) using an experimental method involving the conditioning of (controlled, laboratory-induced) stress is taking place. We hypothesize to find a difference between organized religious belief and personal (esoteric) belief that will be alike in both of the cultural environments.

Keywords: culture, esoteric belief, pattern perception, religiosity

Procedia PDF Downloads 184

Authors: Tiee-Jian Wu, Chih-Yuan Hsu

Abstract:

Mode estimation is an important task, because it has applications to data from a wide variety of sources. We propose a semi-parametric approach to estimate the mode of an unknown continuous multivariate density function. Our approach is based on a weighted average of a parametric density estimate using the Box-Cox transform and a non-parametric kernel density estimate. Our semi-parametric mode estimate improves both the parametric- and non-parametric- mode estimates. Specifically, our mode estimate solves the non-consistency problem of parametric mode estimates (at large sample sizes) and reduces the variability of non-parametric mode estimates (at small sample sizes). The performance of our method at practical sample sizes is demonstrated by simulation examples and two real examples from the fields of climatology and image recognition.

Keywords: Box-Cox transform, density estimation, mode seeking, semiparametric method

Procedia PDF Downloads 283

Authors: Wentian Shi, Daming Shi, Maysam Orouskhani, Feng Tian

Abstract:

Whereas currently the most prevalent deep learning methods require a large amount of data for training, few-shot learning tries to learn a model from limited data without extensive retraining. In this paper, we present a loss function based on triplet loss for solving few-shot problem using metric based learning. Instead of setting the margin distance in triplet loss as a constant number empirically, we propose an adaptive margin distance strategy to obtain the appropriate margin distance automatically. We implement the strategy in the deep siamese network for deep metric embedding, by utilizing an optimization approach by penalizing the worst case and rewarding the best. Our experiments on image recognition and co-segmentation model demonstrate that using our proposed triplet loss with adaptive margin distance can significantly improve the performance.

Keywords: few-shot learning, triplet network, adaptive margin, deep learning

Procedia PDF Downloads 167

Authors: Hong Im Shin

Abstract:

Previous research demonstrated that self-objectification (women view themselves as objects for use) is related to system-justification. Three studies investigated whether cultural disposition as its system-justifying function could have an impact on self-objectification and dehumanization of sexualized women and men. Study 1 (N = 91) employed a survey methodology to examine the relationship between cultural disposition (collectivism vs. individualism), trait of system-justification, and self-objectification. The results showed that the higher tendency of collectivism was related to stronger system-justification and self-objectification. Study 2 (N = 60 females) introduced a single category implicit association task (SC-IAT) to assess the extent to which sexually objectified women were associated with uniquely human attributes (i.e., culture) compared to animal-related attributes (i.e., nature). According to results, female participants associated sexually objectified female targets less with human attributes compared to animal-related attributes. Study 3 (N = 46) investigated whether priming to individualism or collectivism was associated to system justification and sexual objectification of men and women with the use of a recognition task involving upright and inverted pictures of sexualized women and men. The results indicated that the female participants primed to individualism showed an inversion effect for sexualized women and men (person-like recognition), whereas there was no inversion effect for sexualized women in the priming condition of collectivism (object-like recognition). This implies that cultural disposition plays a mediating role for rationalizing the gender status, implicit dehumanization of sexualized females and self-objectification. Future research directions are discussed.

Keywords: cultural disposition, dehumanization, implicit test, self-objectification

Procedia PDF Downloads 237

Authors: Rui Mao, Heming Ji, Xiaoyu Wang

Abstract:

Keystroke dynamics are widely used in identity recognition. It has the advantage that the individual typing rhythm is difficult to imitate. It also supports continuous authentication through the keyboard without extra devices. The existing keystroke dynamics authentication methods based on machine learning have a drawback in supporting relatively complex scenarios with massive data. There are drawbacks to both feature extraction and model optimization in these methods. To overcome the above weakness, an authentication model of keystroke dynamics based on deep learning is proposed. The model uses feature vectors formed by keystroke content and keystroke time. It ensures efficient continuous authentication by cooperating attention mechanisms with the combination of CNN and Bi-LSTM. The model has been tested with Open Data Buffalo dataset, and the result shows that the FRR is 3.09%, FAR is 3.03%, and EER is 4.23%. This proves that the model is efficient and accurate on continuous authentication.

Keywords: keystroke dynamics, identity authentication, deep learning, CNN, LSTM

Procedia PDF Downloads 153

Authors: Bahaa Eltahawy, Mikko Ylihärsilä, Reino Virrankoski, Esko Petäjä

Abstract:

Sheet metal processing is automated, but the step from product models to the production machine control still requires human intervention. This may cause time consuming bottlenecks in the production process and increase the risk of human errors. In this paper we present a system, which automatically recognizes features from the CAD-model of the sheet metal product. By using these features, the system produces a complete model of the particular sheet metal product. Then the model is used as an input for the sheet metal processing machine. Currently the system is implemented, capable to recognize more than 11 of the most common sheet metal structural features, and the procedure is fully automated. This provides remarkable savings in the production time, and protects against the human errors. This paper presents the developed system architecture, applied algorithms and system software implementation and testing.

Keywords: feature recognition, automation, sheet metal manufacturing, CAD, CAM

Procedia PDF Downloads 353

Authors: Adian Fatchur Rochim, Abdul Muis, Riri Fitri Sari

Abstract:

H-index has been widely used as a performance indicator of researchers around the world especially in Indonesia. The Government uses Scopus and Google scholar as indexing references in providing recognition and appreciation. However, those two indexing services yield to different H-index values. For that purpose, this paper evaluates the difference of the H-index from those services. Researchers indexed by Webometrics, are used as reference’s data in this paper. Currently, Webometrics only uses H-index from Google Scholar. This paper observed and compared corresponding researchers’ data from Scopus to get their H-index score. Subsequently, some researchers with huge differences in score are observed in more detail on their paper’s publisher. This paper shows that the H-index of researchers in Google Scholar is approximately 2.45 times of their Scopus H-Index. Most difference exists due to the existence of uncertified publishers, which is considered in Google Scholar but not in Scopus.

Keywords: Google Scholar, H-index, Scopus, performance indicator

Procedia PDF Downloads 272

Authors: Chien-Hung Chen, Yan-Chun Hsing, Yung-Chun Chang

Abstract:

The Internet has grown into a powerful medium for information dispersion and social interaction that leads to a rapid growth of social media which allows users to easily post their emotions and perspectives regarding certain topics online. Our research aims at using natural language processing and text mining techniques to explore the public emotions expressed on Twitter by analyzing the sentiment behind tweets. In this paper, we propose a composite kernel method that integrates tree kernel with the linear kernel to simultaneously exploit both the tree representation and the distributed emotion keyword representation to analyze the syntactic and content information in tweets. The experiment results demonstrate that our method can effectively detect public emotion of tweets while outperforming the other compared methods.

Keywords: emotion recognition, natural language processing, composite kernel, sentiment analysis, text mining

Procedia PDF Downloads 216

Authors: Agoston Restas

Abstract:

Introduction: In case of managing disasters decision makers can face many times such a special situation where any pre-sign of the drastically change is missing therefore the improvised decision making can be required. The complexity, ambiguity, uncertainty or the volatility of the situation can require many times the improvisation as decision making. It can be taken at any level of the management (strategic, operational and tactical) but at tactical level the main reason of the improvisation is surely time pressure. It is certainly the biggest problem during the management. Methods: The author used different tools and methods to achieve his goals; one of them was the study of the relevant literature, the other one was his own experience as a firefighting manager. Other results come from two surveys that are referred to; one of them was an essay analysis, the second one was a word association test, specially created for the research. Results and discussion: This article proves that, in certain situations, the multi-criteria, evaluating decision-making processes simply cannot be used or only in a limited manner. However, it can be seen that managers, directors or commanders are many times in situations that simply cannot be ignored when making decisions which should be made in a short time. The functional background of decisions made in a short time, their mechanism, which is different from the conventional, was studied lately and this special decision procedure was given the name recognition-primed decision. In the article, author illustrates the limits of the possibilities of analytical decision-making, presents the general operating mechanism of recognition-primed decision-making, elaborates on its special model relevant to managers at tactical level, as well as explore and systemize the factors that facilitate (catalyze) the processes with an example with fire managers.

Keywords: decision making, disaster managers, recognition primed decision, model for making decisions in emergencies

Procedia PDF Downloads 257

Authors: Zeki Oralhan, Mahmut Tokmakçı

Abstract:

SSVEP based brain computer interface (BCI) systems have been preferred, because of high information transfer rate (ITR) and practical use. ITR is the parameter of BCI overall performance. For high ITR value, one of specification BCI system is that has high accuracy. In this study, we investigated to recognize SSVEP with shorter time and lower error rate. In the experiment, there were 8 flickers on light crystal display (LCD). Participants gazed to flicker which had 12 Hz frequency and 50% duty cycle ratio on the LCD during 10 seconds. During the experiment, EEG signals were acquired via EEG device. The EEG data was filtered in preprocessing session. After that Canonical Correlation Analysis (CCA), Multiset CCA (MsetCCA), phase constrained CCA (PCCA), and Multiway CCA (MwayCCA) methods were applied on data. The highest average accuracy value was reached when MsetCCA was applied.

Keywords: brain computer interface, canonical correlation analysis, human computer interaction, SSVEP

Procedia PDF Downloads 265

Authors: Shaikh Farhad Hossain, Liakot Ali

Abstract:

Technology advances accelerate the quality and type of services provided for health care and especially for monitoring health conditions. Sensors have turned out to be more effective to detect diverse physiological signs and can be worn on the human body utilizing remote correspondence modules. An assortment of programming devices have been created to help in preparing a difference rundown of essential signs by examining and envisioning information produced by different sensors. In this proposition, we presented a Health signs and Activity acknowledgment monitoring system. Utilizing off-the-rack sensors, we executed a movement location system for identifying five sorts of action: falling, lying down, sitting, standing, and walking. The framework collects and analyzes sensory data in real-time, and provides different feedback to the users. In addition, it can generate alerts based on the detected events and store the data collected to a medical server.

Keywords: ADL, SVM, TRIL , MEMS

Procedia PDF Downloads 393

Authors: Laicheng Cao, Xiangqian Su, Youxiao Wu

Abstract:

Feature selection is one of the important problems in network security, pattern recognition, data mining and other fields. In order to remove redundant features, effectively improve the detection speed of intrusion detection system, proposes a new feature selection method, which is based on the invasive weed optimization (IWO) algorithm and tabu search algorithm(TS). Use IWO as a global search, tabu search algorithm for local search, to improve the results of IWO algorithm. The experimental results show that the feature selection method can effectively remove the redundant features of network data information in feature selection, reduction time, and to guarantee accurate detection rate, effectively improve the speed of detection system.

Keywords: intrusion detection, feature selection, iwo, tabu search

Procedia PDF Downloads 528

Authors: Nikat Parveen, M. Ananthi

Abstract:

Big data is one of the emerging technology, which collects the data from various sensors and those data will be used in many fields. Data retrieval is one of the major issue where there is a need to extract the exact data as per the need. In this paper, large amount of data set is processed by using the feature selection. Feature selection helps to choose the data which are actually needed to process and execute the task. The key value is the one which helps to point out exact data available in the storage space. Here the available data is streamed and R-Center is proposed to achieve this task.

Keywords: big data, key value, feature selection, retrieval, performance

Procedia PDF Downloads 339

Authors: Jia-Hong Lee, Mei-Yi Wu, Hsien-Tsung Kuo

Abstract:

Texture is an important characteristic in real and synthetic scenes. Texture analysis plays a critical role in inspecting surfaces and provides important techniques in a variety of applications. Although several descriptors have been presented to extract texture features, the development of object recognition is still a difficult task due to the complex aspects of texture. Recently, many robust and scaling-invariant image features such as SIFT, SURF and ORB have been successfully used in image retrieval and object recognition. In this paper, we have tried to compare the performance for texture classification using these feature descriptors with k-means clustering. Different classifiers including K-NN, Naive Bayes, Back Propagation Neural Network , Decision Tree and Kstar were applied in three texture image sets － UIUCTex, KTH-TIPS and Brodatz, respectively. Experimental results reveal SIFTS as the best average accuracy rate holder in UIUCTex, KTH-TIPS and SURF is advantaged in Brodatz texture set. BP neuro network works best in the test set classification among all used classifiers.

Keywords: texture classification, texture descriptor, SIFT, SURF, ORB

Procedia PDF Downloads 367

Authors: Michael Karnes, Alper Yilmaz

Abstract:

This paper presents the application of the omni-modeler towards pedestrian redetection. The pedestrian redetection task creates several challenges when applying deep neural networks (DNN) due to the variety of pedestrian appearance with camera position, the variety of environmental conditions, and the specificity required to recognize one pedestrian from another. DNNs require significant training sets and are not easily adapted for changes in class appearances or changes in the set of classes held in its knowledge domain. Pedestrian redetection requires an algorithm that can actively manage its knowledge domain as individuals move in and out of the scene, as well as learn individual appearances from a few frames of a video. The Omni-Modeler is a dynamically learning few-shot visual recognition algorithm developed for tasks with limited training data availability. The Omni-Modeler adapts the knowledge domain of pre-trained deep neural networks to novel concepts with a calculated localized language encoder. The Omni-Modeler knowledge domain is generated by creating a dynamic dictionary of concept definitions, which are directly updatable as new information becomes available. Query images are identified through nearest neighbor comparison to the learned object definitions. The study presented in this paper evaluates its performance in re-identifying individuals as they move through a scene in both single-camera and multi-camera tracking applications. The results demonstrate that the Omni-Modeler shows potential for across-camera view pedestrian redetection and is highly effective for single-camera redetection with a 93% accuracy across 30 individuals using 64 example images for each individual.

Keywords: dynamic learning, few-shot learning, pedestrian redetection, visual recognition

Procedia PDF Downloads 75

Authors: Stephen L. Green, Alexander N. Gorban, Ivan Y. Tyukin

Abstract:

Within the past decade, using Convolutional Neural Networks (CNN)’s to create Deep Learning systems capable of translating Sign Language into text has been a breakthrough in breaking the communication barrier for deaf-mute people. Conventional research on this subject has been concerned with training the network to recognize the fingerspelling gestures of a given language and produce their corresponding alphanumerics. One of the problems with the current developing technology is that images are scarce, with little variations in the gestures being presented to the recognition program, often skewed towards single skin tones and hand sizes that makes a percentage of the population’s fingerspelling harder to detect. Along with this, current gesture detection programs are only trained on one finger spelling language despite there being one hundred and forty-two known variants so far. All of this presents a limitation for traditional exploitation for the state of current technologies such as CNN’s, due to their large number of required parameters. This work aims to present a technology that aims to resolve this issue by combining a pretrained legacy AI system for a generic object recognition task with a corrector method to uptrain the legacy network. This is a computationally efficient procedure that does not require large volumes of data even when covering a broad range of sign languages such as American Sign Language, British Sign Language and Chinese Sign Language (Pinyin). Implementing recent results on method concentration, namely the stochastic separation theorem, an AI system is supposed as an operate mapping an input present in the set of images u ∈ U to an output that exists in a set of predicted class labels q ∈ Q of the alphanumeric that q represents and the language it comes from. These inputs and outputs, along with the interval variables z ∈ Z represent the system’s current state which implies a mapping that assigns an element x ∈ ℝⁿ to the triple (u, z, q). As all xi are i.i.d vectors drawn from a product mean distribution, over a period of time the AI generates a large set of measurements xi called S that are grouped into two categories: the correct predictions M and the incorrect predictions Y. Once the network has made its predictions, a corrector can then be applied through centering S and Y by subtracting their means. The data is then regularized by applying the Kaiser rule to the resulting eigenmatrix and then whitened before being split into pairwise, positively correlated clusters. Each of these clusters produces a unique hyperplane and if any element x falls outside the region bounded by these lines then it is reported as an error. As a result of this methodology, a self-correcting recognition process is created that can identify fingerspelling from a variety of sign language and successfully identify the corresponding alphanumeric and what language the gesture originates from which no other neural network has been able to replicate.

Keywords: convolutional neural networks, deep learning, shallow correctors, sign language

Procedia PDF Downloads 99

Authors: Sultana Afrin Nipa

Abstract:

Life on Earth cannot thrive or survive without water. Water is intimately tied with community, culture, spirituality, identity, socio-economic progress, security, self-determination, and livelihood. Thus, access to water is a United Nations recognized human right due to its significance in these realms. However, there is often conflict between those who consider water as the spiritual and cultural value and those who consider it an economic value thus being threatened by economic development, corporate exploitation, government regulation, and increased privatization, highlighting the complex relationship between water and culture. The Colorado River basin is home to over 29 federally recognized tribal nations. To these tribes, it holds cultural, economic, and spiritual significance and often extends to deep human-to-non-human connections frequently precluded by the Westphalian regulations and settler laws. Despite the recognition of access to rivers as a fundamental human right by the United Nations, tribal communities and their water rights have been historically disregarded through inter alia, colonization, and dispossession of their resources. Law of the River such as ‘Winter’s Doctrine’, ‘Bureau of Reclamation (BOR)’ and ‘Colorado River Compact’ have shaped the water governance among the shareholders. However, tribal communities have been systematically excluded from these key agreements. While the Winter’s Doctrine acknowledged that tribes have the right to withdraw water from the rivers that pass through their reservations for self-sufficiency, the establishment of the BOR led to the construction of dams without tribal consultation, denying the ‘Winters’ regulation and violating these rights. The Colorado River Compact, which granted only 20% of the water to the tribes, diminishes the significance of international legal frameworks that prioritize indigenous self-determination and free pursuit of socio-economic and cultural development. Denial of this basic water right is the denial of the ‘recognition’ of their sovereignty and self-determination that questions the effectiveness of the international law. This review assesses the international legal frameworks concerning indigenous rights and water justice and aims to pinpoint gaps hindering the effective recognition and protection of Indigenous water rights in Colorado River Basin. This study draws on a combination of historical and qualitative data sets. The historical data encompasses the case settlements provided by the Bureau of Reclamation (BOR) respectively the notable cases of Native American water rights settlements on lower Colorado basin related to Arizona from 1979-2008. This material serves to substantiate the context of promises made to the Indigenous people and establishes connections between existing entities. The qualitative data consists of the observation of recorded meetings of the Central Arizona Project (CAP) to evaluate how the previously made promises are reflected now. The study finds a significant inconsistency in participation in the decision-making process and the lack of representation of Native American tribes in water resource management discussions. It highlights the ongoing challenges faced by the indigenous people to achieve their self-determination goal despite the legal arrangements.

Keywords: colorado river, indigenous rights, law of the river, water governance, water justice

Procedia PDF Downloads 31

Authors: Gina M. Newton

Abstract:

Australia’s Environment Protection and Biodiversity Conservation Act (EPBC Act) is the premiere, national law under which species and 'ecological communities' (i.e., like ecosystems) can be formally recognised and 'listed' as threatened across all jurisdictions. The listing process involves assessment against a range of criteria (similar to the IUCN process) to demonstrate conservation status (i.e., vulnerable, endangered, critically endangered, etc.) based on the best available science. Over the past decade in Australia, there’s been a transition from almost solely terrestrial to the first aquatic threatened ecological community (TEC or ecosystem) listings (e.g., River Murray, Macquarie Marshes, Coastal Saltmarsh, Salt-wedge Estuaries). All constitute large areas, with some including multiple state jurisdictions. Development of these conservation and listing advices has enabled, for the first time, a more forensic analysis of three key factors across a range of aquatic and coastal ecosystems: -the contribution of invasive species to conservation status, -how to demonstrate and attribute decline in 'ecological integrity' to conservation status, and, -identification of related priority conservation actions for management. There is increasing global recognition of the disproportionate degree of biodiversity loss within aquatic ecosystems. In Australia, legislative protection at Commonwealth or State levels remains one of the strongest conservation measures. Such laws have associated compliance mechanisms for breaches to the protected status. They also trigger the need for environment impact statements during applications for major developments (which may be denied). However, not all jurisdictions have such laws in place. There remains much opposition to the listing of freshwater systems – for example, the River Murray (Australia's largest river) and Macquarie Marshes (an internationally significant wetland) were both disallowed by parliament four months after formal listing. This was mainly due to a change of government, dissent from a major industry sector, and a 'loophole' in the law. In Australia, at least in the immediate to medium-term time frames, invasive species (aliens, native pests, pathogens, etc.) appear to be the number one biotic threat to the biodiversity and ecological function and integrity of our aquatic ecosystems. Consequently, this should be considered a current priority for research, conservation, and management actions. Another key outcome from this analysis was the recognition that drawing together multiple lines of evidence to form a 'conservation narrative' is a more useful approach to assigning conservation status. This also helps to addresses a glaring gap in long-term ecological data sets in Australia, which often precludes a more empirical data-driven approach. An important lesson also emerged – the recognition that while conservation must be underpinned by the best available scientific evidence, it remains a 'social and policy' goal rather than a 'scientific' goal. Communication, engagement, and 'politics' necessarily play a significant role in achieving conservation goals and need to be managed and resourced accordingly.

Keywords: aquatic ecosystem conservation, conservation law, ecological integrity, invasive species

Procedia PDF Downloads 130

Authors: Rita Chattopadhyay, Vivek Anand Thoutam

Abstract:

Speech is the most acceptable means of communication where we can quickly exchange our feelings and thoughts. Quite often, people can communicate orally but cannot interact or work with computers or devices. It’s easy and quick to give speech commands than typing commands to computers. In the same way, it’s easy listening to audio played from a device than extract output from computers or devices. Especially with Robotics being an emerging market with applications in warehouses, the hospitality industry, consumer electronics, assistive technology, etc., speech-based human-machine interaction is emerging as a lucrative feature for robot manufacturers. Considering this factor, the objective of this paper is to design the “Audio-Visual Co-Data Processing Pipeline.” This pipeline is an integrated version of Automatic speech recognition, a Natural language model for text understanding, object detection, and text-to-speech modules. There are many Deep Learning models for each type of the modules mentioned above, but OpenVINO Model Zoo models are used because the OpenVINO toolkit covers both computer vision and non-computer vision workloads across Intel hardware and maximizes performance, and accelerates application development. A speech command is given as input that has information about target objects to be detected and start and end times to extract the required interval from the video. Speech is converted to text using the Automatic speech recognition QuartzNet model. The summary is extracted from text using a natural language model Generative Pre-Trained Transformer-3 (GPT-3). Based on the summary, essential frames from the video are extracted, and the You Only Look Once (YOLO) object detection model detects You Only Look Once (YOLO) objects on these extracted frames. Frame numbers that have target objects (specified objects in the speech command) are saved as text. Finally, this text (frame numbers) is converted to speech using text to speech model and will be played from the device. This project is developed for 80 You Only Look Once (YOLO) labels, and the user can extract frames based on only one or two target labels. This pipeline can be extended for more than two target labels easily by making appropriate changes in the object detection module. This project is developed for four different speech command formats by including sample examples in the prompt used by Generative Pre-Trained Transformer-3 (GPT-3) model. Based on user preference, one can come up with a new speech command format by including some examples of the respective format in the prompt used by the Generative Pre-Trained Transformer-3 (GPT-3) model. This pipeline can be used in many projects like human-machine interface, human-robot interaction, and surveillance through speech commands. All object detection projects can be upgraded using this pipeline so that one can give speech commands and output is played from the device.

Keywords: OpenVINO, automatic speech recognition, natural language processing, object detection, text to speech

Procedia PDF Downloads 78

Authors: Insiya Bhalloo

Abstract:

It is quite common in the Muslim faith for non-Arabic speakers to be able to convert written Arabic, especially Quranic Arabic, into a phonological code without significant semantic or syntactic knowledge. This is due to prior experience learning to read the Quran (a religious text written in Classical Arabic), from a very young age such as via enrolment in Quranic Arabic classes. As compared to native speakers of Arabic, these Arabic readers do not have a comprehensive morpho-syntactic knowledge of the Arabic language, nor can understand, or engage in Arabic conversation. The study seeks to investigate whether mere phonological experience (as indicated by the Arabic readers’ experience with Arabic phonology and the sound-system) is sufficient to cause phonological-interference during word recognition of previously-heard words, despite the participants’ non-native status. Both native speakers of Arabic and non-native speakers of Arabic, i.e., those individuals that learned to read the Quran from a young age, will be recruited. Each experimental session will include two phases: An exposure phase and a test phase. During the exposure phase, participants will be presented with Arabic words (n=40) on a computer screen. Half of these words will be common words found in the Quran while the other half will be words commonly found in Modern Standard Arabic (MSA) but either non-existent or prevalent at a significantly lower frequency within the Quran. During the test phase, participants will then be presented with both familiar (n = 20; i.e., those words presented during the exposure phase) and novel Arabic words (n = 20; i.e., words not presented during the exposure phase. ½ of these presented words will be common Quranic Arabic words and the other ½ will be common MSA words but not Quranic words. Moreover, ½ the Quranic Arabic and MSA words presented will be comprised of nouns, while ½ the Quranic Arabic and MSA will be comprised of verbs, thereby eliminating word-processing issues affected by lexical category. Participants will then determine if they had seen that word during the exposure phase. This study seeks to investigate whether long-term phonological memory, such as via childhood exposure to Quranic Arabic orthography, has a differential effect on the word-recognition capacities of native Arabic speakers and Arabic readers; we seek to compare the effects of long-term phonological memory in comparison to short-term phonological exposure (as indicated by the presentation of familiar words from the exposure phase). The researcher’s hypothesis is that, despite the lack of lexical knowledge, early experience with converting written Quranic Arabic text into a phonological code will help participants recall the familiar Quranic words that appeared during the exposure phase more accurately than those that were not presented during the exposure phase. Moreover, it is anticipated that the non-native Arabic readers will also report more false alarms to the unfamiliar Quranic words, due to early childhood phonological exposure to Quranic Arabic script - thereby causing false phonological facilitatory effects.

Keywords: modern standard arabic, phonological facilitation, phonological memory, Quranic arabic, word recognition

Procedia PDF Downloads 355

Authors: Alexander Wachtel, Michael T. Franzen, Walter F. Tichy

Abstract:

Programming requires years of training. With natural language and end user development methods, programming could become available to everyone. It enables end users to program their own devices and extend the functionality of the existing system without any knowledge of programming languages. In this paper, we describe an Interactive Spreadsheet Processing Module (ISPM), a natural language interface to spreadsheets that allows users to address ranges within the spreadsheet based on inferred table schema. Using the ISPM, end users are able to search for values in the schema of the table and to address the data in spreadsheets implicitly. Furthermore, it enables them to select and sort the spreadsheet data by using natural language. ISPM uses a machine learning technique to automatically infer areas within a spreadsheet, including different kinds of headers and data ranges. Since ranges can be identified from natural language queries, the end users can query the data using natural language. During the evaluation 12 undergraduate students were asked to perform operations (sum, sort, group and select) using the system and also Excel without ISPM interface, and the time taken for task completion was compared across the two systems. Only for the selection task did users take less time in Excel (since they directly selected the cells using the mouse) than in ISPM, by using natural language for end user software engineering, to overcome the present bottleneck of professional developers.

Keywords: natural language processing, natural language interfaces, human computer interaction, end user development, dialog systems, data recognition, spreadsheet

Procedia PDF Downloads 310