A virtual teleconferencing system based on face detection and 3D animation in a low‐bandwidth environment

In the field of pattern recognition, off-line handwriting recognition is one of the most intensive areas of study. This paper proposes an automatic off-line Thai language student name identification system which was built as a part of a completed off-line automated assessment system. There is limited work undertaken in developing off-line automatic assessment systems using handwriting recognition. To the authors' knowledge, none of the work on the proposed system has been performed on the Thai language. In addition the proposed system recognises each Thai name by using an approach for whole word recognition, which is different from the work found in the literature as most perform character-based recognition. In this proposed system, the Gaussian Grid Feature (GGF) and the Modified Direction Feature (MDF) extraction techniques are investigated on upper and lower contours, loops from full word contour images of each name sample, and artificial neural networks and support vector machine are used as classifiers. The encouraging recognition rates for both feature extraction techniques were achieved when applied on loop, upper and lower contour images (99.27% accuracy rate was achieved using MDF on artificial neural networks and 99.27% using GGF with a support vector machine classifier).

Academic conference was majorly face-to-face gathering for sharing new knowledge, disseminating research findings, establishing networks among team members and for social interaction. However, holding virtual meetings is not completely a new phenomenon but conference virtualisation was not in place before the year 2020. But the current widespread adoption of virtual conferences was facilitated by the global outbreak of the Coronavirus (Covid-19) pandemic, which struck almost the whole world and forced nearly all sectors to resort to online activities. This study seeks to examine the acceptable level of virtual conferences using the most popular web conferencing tools such as Zoom, Google Meet, and Microsoft Teams. Three research questions were raised with three corresponding hypotheses. The design adopted was a survey, the population comprised over 800 academic staff in Yaba College of Technology, Lagos and a sample size of 206 was selected. A structured questionnaire was used to collect data and was analysed with frequency distribution and Chi-square test of association. The finding of the study revealed high awareness about virtual conferences (88.1%), high attendance (76%), and high preference for virtual conference attendance (62%)and these stemmed fromthe opportunity to build networks, security concerns, time management, and cost-effectiveness. It also showed that Zoom was the most preferred virtual technology (50%).The study concluded that virtual conferencing has come to stay because of its high awareness, attendance and preference among academics and Zoom technology is the key driver as the world is moving towards a global community.

PurposeThough prior literature has shown that virtual conferences improve accessibility and provide a comparable educational experience, further research is required to characterize their educational value.MethodsIn this repeated cross-sectional study, demographic and survey data were compared between attendance perspectives for the in-person student-led internal medicine conference held in 2019 and subsequent virtual conference held in 2020.ResultsThere were 146 attendees at the in-person conference and 200 attendees at the online conference, in which 32 (22% response rate) and 52 responses (26% response rate) were gathered, respectively. Comparison of Likert Scale data via Mann-Whitney U Test revealed that learning objectives were better met in-person for the overall conference (p < 0.01) and didactic sessions (p < .05), but not for workshops, in which there was no significant difference. Survey takers noted the virtual conference to be more accessible on multiple factors, but felt as though their potential for interaction with other participants was more limited.ConclusionsResults indicate that though the virtual conference appeared more accessible to attendees, overall learning objectives for the conference and didactic sessions were better met in-person. Interestingly however, there was no observed difference in perceived educational value for small group workshops.

Face detection and tracking is one of the emerging research areas in the image analysis and computer vision systems. This face detection and tracking helps local security forces to investigate crime incidents. This paper describes a face tracking framework that is capable of tracking a face in real time rapidly frame by frame. Camshift algorithm and KLT algorithm implemented and a comparison study between these two algorithms has been described in this paper. A real time video is experimented using these two algorithms for face detection and tracking. The experimental results show that the KLT algorithm performance is better than the Camshift algorithm in detecting the face and tracking the face. In this article it has been shown how the KLT algorithm proved to be better tracking algorithm than Camshift algorithm.

The phenomenon of Fake news is experiencing a rapid and growing progress with the evolution of the means of communication and Social media. Fake news detection is an emerging research area which is gaining big interest. It faces however some challenges due to the limited resources such as datasets and processing and analysing techniques. In this work, we propose a system for Fake news detection that uses machine learning techniques. We used term frequency-inverse document frequency (TF-IDF) of bag of words and n-grams as feature extraction technique, and Support Vector Machine (SVM) as a classifier. We propose also a dataset of fake and true news to train the proposed system. Obtained results show the efficiency of the system. In this work, we propose a system for Fake news detection that uses machine learning techniques. We used term frequency-inverse document frequency (TF-IDF) of bag of words and n-grams as feature extraction technique, and Support Vector Machine (SVM) as a classifier. We propose also a dataset of fake and true news to train the proposed system. Obtained results show the efficiency of the system.

The current face detection in Microsoft HoloLens can only be achieved by remote call of face detection interface algorithm which is, however, restricted by network, resulting in slow detection and failing to meet real-time detection demand. This paper proposes an improved Viola-Jones algorithm of face detection based on HoloLens upgrading classical Viola-Jones face detection algorithm relying on Haar-like rectangle feature expansion to enhance detection efficiency, and accelerating detection building on two-dimensional convolution separation and image re-sampling technique. The detection efficiency of improved face detection algorithm is 12% on average higher than that of existing face detection interface algorithm, and its detection speed is four-folded. Moreover, HoloLens depth camera enables 3D face detection and location, and its unique gaze, voice, and gesture interacting techniques free the hands, thereby realizing easier and less-burdened man-computer interaction. HoloLens furnished with real-time video face detection algorithm as detailed in this paper can be applied in such fields as social contact, public security, and business management.

BackgroundInfectious diseases in plants, animals and humans are often transmitted indirectly between hosts (or between groups of hosts), i.e. via some route through the environment instead of via direct contacts between these hosts. Here we study indirect transmission experimentally, using transmission of Campylobacter jejuni (C. jejuni) between spatially separated broilers as a model system. We distinguish three stages in the process of indirect transmission; (1) an infectious “sender” excretes the agent, after which (2) the agent is transported via some route to a susceptible “receiver”, and subsequently (3) the receiver becomes colonised by the agent. The role of the sender and receiver side (stage 1 and stage 3) was studied here by using acidification of the drinking water as a modulation mechanism.ResultsIn the experiment one control group and three treatment groups were monitored for the presence of C. jejuni by taking daily cloacal swabs. The three treatments consisted of acidification of the drinking water of the inoculated animals (the senders), acidification of the drinking water of the susceptible animals (the receivers) or acidification of the drinking water of both inoculated and susceptible animals. In the control group 12 animals got colonised out of a possible 40, in each treatment groups 3 animals out of a possible 40 were found colonised with C. jejuni.ConclusionsThe results of the experiments show a significant decrease in transmission rate (β) between the control groups and treatment groups (p < 0.01 for all groups) but not between different treatments; there is a significant negative interaction effect when both the sender and the receiver group receive acidified drinking water (p = 0.01). This negative interaction effect could be due to selection of bacteria already at the sender side thereby diminishing the effect of acidification at the receiver side.

There are diverse approaches in face detection and recognition, highlighting algorithms, datasets, and practical applications. The Haar Cascade Classifier is a popular technique for fast and simple detection, though it is hindered by false positives and reduced accuracy in complex scenes. Advanced methods, such as combining machine learning algorithms with feature extraction techniques like Principal Component Analysis (PCA) and Three-Patch Local Binary Pattern (TPLBP), are examined for their higher recognition rates and improved accuracy. The use of datasets like ORL and Sheffield demonstrates robustness under various conditions, including changes in pose, lighting, and expression. Hybrid methods like Scale-Invariant Feature Transform (SIFT) and Principal Component Analysis integration, achieve remarkable accuracy on multiple datasets despite image variances. Notable contributions include the Dual Shot Face Detector algorithm for addressing challenges like scale variation and occlusion and the Viola-Jones algorithm, which is used for real-time detection with minimum computational overhead. The study also evaluates the performance of different algorithms across datasets and real-world applications, including automated attendance systems and emotion detection. This study conducts a comparative analysis of existing face detection methods, evaluating their accuracy, efficiency, and scalability across diverse environments. By identifying strengths and limitations, the study aims to provide insights and propose potential enhancements to guide the development of more robust and adaptable face detection solutions.

The Covid-19 pandemic has forced a change in the way people work, and the location that they work from. The impact has caused significant disruption to education, the work environment and how social interactions take place. Online user habits have also changed due to lockdown restrictions and virtual conferencing software has become a vital cog in team communication. In result, a spate in software solutions have emerged in order to support the challenges of remote learning and working. The conferencing software landscape is now a core communication solution for company-wide interaction, team discussions, screen sharing and face-to-face contact. Yet the number of existing platforms is diverse. In this article, a systematic literature review investigation on virtual conferencing is presented. As output from the analysis, 67 key features and 74 obstacles users experience when interacting with virtual conferencing technologies are identified from 60 related open-source journal articles from 5 digital library repositories.

I don't like going to conferences, mostly because I'm a nervous traveler. After standing in airport lines, chasing meals, and sleeping poorly, I usually come home from a several-day event somewhat disoriented. But the stress surrounding travel is usually worth it because most of the conferences I attend are excellent. And while the technical presentations can be useful and interesting, it's the networking that promotes professional growth and creates important friendships. Before COVID-19, conferences were all about making new connections, running into old friends, and being physically surrounded by colleagues, clients, and vendors—everyone there to give a presentation, hear an expert's talk, or ask around for opinions. When you're physically present, it feels like you can more easily identify the experts and find the solutions you need. Every part of this ultimately guides collective improvement because the lessons learned from large conferences are taken home and spread to neighboring communities as ideas pass between local and regional colleagues. Unfortunately, the best aspects of a conference are also the exact reasons we can't have them—large face-to-face gatherings may promote the sharing of good ideas, but we're also trying to not spread a deadly virus. While in-person opportunities may begin to come back in 2021 (I hope to see you at ACE21 in San Diego), consider making provisional space for yourself and your staff to stay connected through virtual conferences as much as possible. Although more screen time is the last thing we typically want, you can start by enjoying the fact that you don't need to go to an airport, you get to sleep in your own bed, and you can make your own coffee. Added benefits of virtual conferences include savings on travel and the fact that many events are recorded for on-demand viewing—together, this means more water professionals can attend. Smaller organizations that might not normally budget for travel to large events now have the same opportunity to hear about practical approaches and voice their experiences and boundary conditions. Water professionals from across the globe can truly come together to exchange ideas. And for the many who need certification or education credits, virtual conferences provide the answer. Whether it's in person, online, or both, water professionals must do their best to stay connected. Make sure young professionals in your organization participate. Go into events with specific goals, email questions to presenters, and make sure your organization's attendees share what they learned. While in-person connections are limited for the time being, it's up to all of us to keep the flame alive by attending and promoting online events and training in the water industry. This month's Journal AWWA features articles on water reuse, pandemic planning, and utility optimization, all topics you will be guaranteed to find in a virtual conference. If you are interested in submitting an article, contact me at journaleditor@awwa.org.

In this technologically accelerating era, wireless message passing is playing an exceptional role. These messages being transmitted over an open environment is vulnerable to various security threats. To hold back these impedances several cryptographic measures have been introduced. Computation complexities, key management etc are some of the common hindrances that the existing cryptographic techniques face. The proposed system, Cryptolight, uses visible light communication channels for transmission. The system is completely immune to radio frequency attacks. The data is encoded as LED blinks in the sender side and reacquired at the receiver side by the process of image processing. Key board, display device, PIC board and an LED strip constitutes the sender side. The receiver side incorporates a camera, Raspberry-pi and a display device. Filters are employed in order to diminish the effect of noise. The display device and the key board is connected to the LED strip with the PIC board as the interface. The input text is converted into LED blinks, and is captured using camera at the receiver side. These images are later processed in the Raspberry-pi to retain the text.

In this paper several face detection algorithms are compared on the basis of mathematical analysis to find out the most efficient algorithm. At first the mathematical model of different face detection algorithms (Camshift, AdaBoost, LBP and Viola Jones algorithms) are analyzed and compared to find out the most efficient one. Mathematical results show that Viola Jones performs best result to detect the face. But in case of Viola Jones, integral image integrates the non-face region pixels with face region pixels as a result, the pixel value redundancy is occurred which degrades its efficiency. To overcome this problem, a new face detection algorithm is proposed in this paper which is named as Break Point Value (BPV) algorithm. The mathematical model of our proposed method is derived where integral images are compared with Local Binary Pattern (LBP) and the compared value is suggested as test value. If the test value is less than or equal to the BPV then the region is a face region and if it is not, the region is a non-face region. Since there is a comparison between integral image value and LBP value of the same pixel region the redundant values are reduced. Furthermore, the use of BPV helps to find out more relevant frames. Thus the proposed method is more efficient face detection process as compared to the previous processes in the field of face detection system.

A dual-driven adaptive video system over WLAN

Objectives: The study presents a novel DH (Data Hiding) technique which is the combination of DH and context aware system (CAS) for developing/improving the conventional DH scheme. It ensures that the decoder/receiver authenticity before proceeding into the decode process. Methods/Statistical Analysis: This validation operation is performing in three different layers using the CAS properties. The decode action is performed based on the validation result by the proposed system at receiver side. In detail, if the decoder validation is success then he/she may permit to decode the encoded data by Modified Steganography for Image Decode (MSID) process else the system will give another chance to make their validation. If the validation is failed maximum of three times then the system will be deleted the encoded without the knowledge of the receiver. Findings: This feature has been added with the conventional DH system. The proposed system is divided into two different stages of DH and validation using CAS. The secret image is hiding inside of the known image using Modified Steganography for Image encode (MSIE) process and the validation parameters of the user name, location where the decode should be performed and time schedule are appended and it sent to the receiver in the sender side. He/ she must perform the validation process and the CAS will take a decision based on the validation result once the receiver/ decoder received the data. As the outcome of this system, the intruder can’t be access the encoded data 99%, the strengthen ratio is improved up to 95% than the existing scheme, the secret data is reconstructed without any loss, it can ensure the reliability, availability and confidentiality of the secret data up to 98%. Application/Improvements: It has implemented for the clandestine image/data transmission between sender and receiver using CAS. Keywords: Context Aware System, Data Hiding, Data Hiding Technique, Receiver, Secret Data, Sender

A GPU Based Implementation of Robust Face Detection System