Pakistan sign language recognition: leveraging deep learning models with limited dataset

Lightweight Deep Learning Framework for Speech Emotion Recognition

Data Sensing Devices (DSD’s) have gained lot of traction for various use cases like border control, vehicle tracking. Data Sensing device network (DSDN) is shaped with the aid of combining lot of DSD’s across a random area. Like this multiple groups are formed. In each of the group the specific DSD is elected which is responsible for communication between two independent groups. Each of the group head has multiple attributes with first attribute based on distance, the second attribute based on remaining energy. These attributes will be input for the group head selection based on machine learning, The entire DSD’s inside a group are classified into HIGH, MEDIUM and LOW. The first priority will be given to HIGH followed by others for the primary group head selection. LEACH is a classical method used for transmission of chunks to the control center in a DSDN network. The selection of head DSD by LEACH will happen by making use of the random selection of DSD in each group using random probability selection mechanism. During the data chunk deliver the scanning process will happen from the initiator DSD to head DSD and from there the link is established with the base station (BS), the BS will then scan each group until the destination DSD is reached. The selection of head DSD by LEACH causes more holes in the DSDN because there are chances that the non-performer DSD can become a head DSD. Secondly for the transmission of chunks there is lot of back-and-forth propagation between the BS and the normal DSDs which reduces the battery level of the DSD by a large amount. The Energy based LEACH is modified on top of LEACH by measuring the energy of the DSDs and then selecting the group heads but suffers from multiple group head maintenance as well as more number of links. The proposed method will improve this by reducing the links used for end-to-end communication. In the proposed system the communication will happen based on initiator DSD, primary DSDs in different groups and then destination DSD which will avoid overhead compared to existing methods namely E-LEACH and LEACH. The proposed method is compared with LEACH and E-LEACH with respect to time taken, link count, energy consumption, residual energy measure, lifetime and overhead.

pdf/a-novel-modified-energy-and-throughput-efficient-loyal-un-1jlmkuxk8i.pdf

A Novel Modified Energy and Throughput Efficient LOYAL UN-Supervised LEACH Method for Wireless Sensor Networks

There is massive call for the Packet Sender Device Network (PSDN) primarily based on tracking of areas, figuring out the consequences of climate, detection of enemy vehicles. The PSDN could have useful Packet Sender Devices (PSD’s) which can study the vicinity and then send the data from initial to receiver PSD. There are numerous constraints which have restrictions on the feature like battery, memory, and range. There is hierarchical community wherein the PSDs are spread on more than one area with every vicinity having their very own PSD’s whilst communique has to manifest among PSDs of various areas then it requires chief PSD in every vicinity which have to be elected primarily based on higher battery degree, distance to base station in addition to mobility of the PSD over a duration of time. LEACH suffers from side-to-side propagation among base station and PSDs. LEACH elects the pinnacle PSD primarily based on random possibility which has consequences of Good-Bad Ratio due to the fact there are possibilities that the PSD with low battery degree be selected as a head PSD. The proposed (HTBH) method will select the head PSD based on battery level, distance between base station and PSD, the mobility of PSD. In order to deliver the data to the processing lab PSDs of the area along with head PSD are used. The packets are classified into high and low priority so that preferences can be used during the data delivery process. Along with sending the current data packets even the data packets reside within the PSD which have to send towards the processing centre. Two policies namely single threshold and dual threshold are used to evict the packets within the PSD and then moving them towards processing centre. This study proposes a higher choice of head PSD with the aid of using considering computation of element which takes into consideration battery, distance, and mobility. The communique among the detection vicinity to receiver vicinity takes place with the assist of PSD’s and head PSD there with the aid of using decreasing range of hops. HTBH method is compared to LEACH and E-LEACH with respect various performance parameters. 

A Novel Buffer Packet Delivery Strategy for High Throughput and Better Health (HTBH) Method in Wireless Sensor Networks

Human activity recognition aims to work out the activities performed by someone in a picture or video. Examples of actions are running, sitting, sleeping, and standing. Complex movement patterns and harmful occurrences like falling may be a part of these activities. The suggested ConvLSTM network can be created by successively combining fully connected layers, long immediate memory (LSTM) networks, and convolutional neural networks (CNN). The acquisition system will pre calculate skeleton coordinates using human detection and pose estimation from the image/video sequence. The ConvLSTM model builds new controlled features from the raw skeleton coordinates and their distinctive geometric and kinematic properties. Raw skeleton coordinates are utilized to generate geometric and kinematic properties supported by relative joint position values, joint differences, and their angular velocities. By utilizing a multi-player trained CNN-LSTM combination, novel spatiotemporal directed features can be obtained. The classification head with completely connected layers is then utilized. The suggested model was tested using the KinectHAR dataset, which consists of 130,000 samples with 81 attribute variables and was compiled using the Kinect (v2) sensor Experimental data is used to compare the performance of independent CNN and LSTM networks.

human activity recognition using CNN and long term short term memory

Sentiment analysis Sentiment analysis is the process of extracting information from the text and it is considered as opinion mining. Machine learning experiments have been shown in the study that involves review, blog, for some texts that are written in different languages including Dutch, English and French. Set of example sentence have been set that are manually labeled, neutral, positive or negative. The study covers the curiosity of the consumer regarding the specific consumption products. Categorization models have been developing that has been used in the study. Number of issues that includes noisy nature of the text has been discussed in the study. With an accuracy of roughly 83 percent for English texts, we can determine positive, negative, and neutral sentiments toward the subject under investigation using unigram features augmented with linguistic information. The role of active learning approaches in minimizing the number of instances that need to be manually annotated is discussed in this article. Our studies also give data on the transferability of learned models between domains and languages. Here, Sentiment analysis of a particular person has studied using a K-Means Clustering and SVM classifier to classify the sentiment of a person from the text

The Sentiment Analysis of Human Behavior on Products and Organizations using K-Means Clustering and SVM Classifier

Emotion recognition is an essential field of research in social robots, medical applications, neuromarketing, E-learning, monitoring, entertainment, law and games in human-to-machine interactions In face-to-face interactions, facial expressions convey non-verbal information among humans, which is an important task in human–machine interfaces Since the early 1990s, it has attracted attention from researchers Classical approaches to machine learning often involve complex processes of extraction of features and poor results Significant research has been carried out over the past decade in the development of different methods of extraction, tracking and analysis of facial features The analysis includes measuring facial feeling and recognition of expression by extracting options for classification victimization ancient machine learning algorithms and could be a complicated and long task in comparison to deep learning techniques As per the literature survey, using machine learning algorithms, the achieved accuracy rate ranges between 75 to 90% for mostly four to six emotions from the facial data set In this paper, a deep CNN, a recent advancement in deep learning, is employed, which might accurately interpret linguistics information out there in the face in an automatic manner while not hand-designing the feature descriptors to acknowledge seven completely different emotions, viz neutral, anger, fear, happy, disgust, surprise and sad Categorical cross-entropy loss and Adam optimizer algorithms are applied to coach D-CNN with a powerful classification power This D-CNN is applied on well-known JAFFE data set which consists of seven emotions and ten subjects and could achieve an accuracy of 9753% Compared to machine learning, the number of parameters in D-CNN decreases intensively, accelerating the overall performance improvement and making this network well suited for real-time applications

Emotion Recognition from Facial Biometric System Using Deep Convolution Neural Network (D-CNN)

In today’s world, emotions play a crucial role in human-to-human and human-to-machine interactions and that is important for both fair and good decisions. The display of emotions gives an important information of the mental state of a human being and this opened up a new research area with a fundamental goal of understanding and recovering desired emotions. In the past years, several biometrics have been used to classify different emotions (facial expressions, speech and physiological signals, etc.) of human beings. Some intrinsic advantages make speech signals became a good source for affective computing. There are several applications in real time for communicating with robots, and also in web-based e-learning and commercial applications to detect emotions from speech, etc. In this work, LPCC, MFCC, and the features from high-amplitude regions of each VOP (vowel onset points)-extracted syllable are selected and used to classify different emotions using the machine learning classifier (SVM). This study uses six emotions: anger, disgust, fear, happy, sad, and surprised.

Emotion Recognition from Speech Biometric System Using Machine Learning Algorithms

Communication is crucial in our day-to-day lives. It is the basis for all human interactions. It is how people pass on information to other people and receive information from them. A person with a speaking or hearing problem cannot communicate properly with others. A person who isdeaf or dumb usually uses Sign Language to communicate with others. Normal individuals who would not know Sign language, on the other hand, cannot communicate better with them. Indian Sign Language acts as a mode of communication for deaf and dumb people who constitute a hefty portion of the Indian population. Most people struggle to apprehend ISL gestures. As a result, communication gap is created between the hearing and speech impaired and those who do not understand ISL. Translation systems are required to help bridge the gap between the community of people withhearing and speechimpairment and the general population. This paper focuses on developing an end-to-end system that can recognize the spoken language and interpret the corresponding speech to animated sign language gestures and that is also capable of converting Indian Sign Language to speech.

Indian Sign Language Translator

In the Covid-19 age, we are becoming increasingly reliant on virtual interactions like as Zoom and Google meetings / Teams chat. The videos received from live webcamera in virtual interactions become great source for researchers to understand the human emotions. Due to the numerous applications in human-computer interaction, analysis of emotion from facial expressions has piqued the interest of the newest research community (HCI). The primary objective of this study is to assess various emotions using unique facial expressions captured via a live web camera. Traditional approaches (Conventional FER) rely on manual feature extraction before classifying the emotional state, whereas Deep Learning, Convolutional Neural Networks, and Transfer Learning are now widely used for emotional classification due to their advanced feature extraction mechanisms from images. In this implementation, we will use the most advanced deep learning models, MTCNN and VGG-16, to extract features and classify seven distinct emotions based on their facial landmarks in live video. Using the FER2013 standard dataset, we achieved a maximum accuracy of 97.23 percent for training and 60.2 percent for validation for emotion classification.

Real-Time Emotional Analysis from A Live Webcam Using Deep Learning

Handle everyday research tasks with reliable, citation-backed results

Your personal Research Agent to handle research tasks with citation-backed results

Popular Tasks used by Researchers

How can I help with your research?

Meet SciSpace

Get more enhanced response by uploading the PDFs you want me to reference.

No relevant PDFs in your library

SciSpace is the AI research assistant for academics. Run systematic literature reviews on 280M+ papers, and write papers with cited sources. Trusted by 1M+ students, PhDs & researchers.

SciSpace | AI for Research

Analyze PDFs

Code & Manuscripts

Funding & Grants

Literature & Patents

Medical & Clinical Data

Systematic Review

Visualize & Present

Web & Data

Build a Google Scholar-like website for your research.

Build a website

Create charts and images for your research

Create a Chart

Write a paper for submission to a journal

Draft a manuscript

Patent Search

Design eye-catching scientific posters in minutes.

Scientific Poster Generation

Systematic Literature Review

One task is running at the moment. Your messages will be shown right after.

Drag and drop or click here to browse

Loved by <highlight>1 million+</highlight> researchers

Extract a list of specific topics and their sources from unstructured text

Topics

Compare and analyze relevant papers that matches with your search

Papers

Get insights from PDFs and bookmarked papers from your library

My library

Recent searches

Try searching for:

Catch AI-generated content in scholarly and non-scholarly content

{ai} Detector

Ai Writer

Get PDF Summaries, highlighted text explanations 

Chat with PDF

Effortlessly create in-text citations and bibliographies in APA and 2,500 other formats

Citation generator

Get explanations, summaries, and answers on academic papers

Ease up your research workflow with {scispace}'s cohort of exciting AI tools

Elevate your academic writing skills and convey your ideas the way you want

Paraphraser

Explore our range of reading and writing tools

Your file is being prepared and should be ready in a few minutes. If it's a large file, it might take a bit longer. You can close this window, and we'll email you the file when it's done.

You have reached a maximum limit of <strong>{limit}</strong> columns in the table. Remove at least <strong>1</strong> column to add or create another one.

Chevella Anil kumar

Author Tools

Chat about Author