The outbreak of the COVID-19 pandemic has changed the whole world scenario and made researchers innovate on the corona virus. Researchers are working on information that includes symptoms, Infection spreading, preventive measures, health and travel advisories, and help lines for further assistance. During this pandemic scenario, the health assistant Chatbot is a very useful conversation tool for COVID-19, which provides preliminary medical advice and preventive measure suggestions. The paper proposes an Artificial Intelligence-based Re-Co Chatbot to provide information about the corona virus and also assist with customer queries. The goal is to build a 24/7 COVID Chatbot capable of answering user questions and to emphasize and stress the concept of contextual semantic search and Knowledge Graph to serve as the FAQ for Corona information. Natural Language Processing (NLP) is used to process the user question and the SpaCy library is used for text processing. Once the question is processed, entities (the subject of question) and relations (predicate of the question) are recognized and extracted. The Chatbot is designed for about 100 question-answers pairs in the CSV file and will create about 575 relationships in the Knowledge Graph.

Any smart water management system should include the optimal use of the water by effective quality monitoring, controlling, distribution, and consumption. For a county like India, smart water management is a challenging issue because people consider water can be available forever, but it is not the case. Unlike electricity generation using wind and solar, electricity generated by water is more difficult due to the absence of a metering system, the bill would be the common flat rate for all residents irrespective of the varying amounts of water consumed by each household. This chapter proposes the real-time measurement of consumption, monitoring of leakages, ability to control the water supply if there is leakage, a completely automated platform for societies, and apartment complexes to set up their billing system. The solution consists of a flow sensor meter installed in the main water inlet pipe to pick up water usage data and communicates through the Wi-Fi network to iOS and Android-compatible applications. Two sets of modules Home 1 and Home 2 metering system have been designed and tested. The test results show that the bill generated is 13.33 and 22.133 for flow rate of 17.688 and 26.667, respectively.

The development in the healthcarePatgar, Tanuja sector improves people life more digitized day by day. The adaptation of electronic devices makes patient’s normal activities become more easier without physical movement. Assistive technology is considered as one such excellent healthcare solution to disable person to support the growth of ecosystem. Tetraplegia is a paralysis condition where a patient cannot move parts below the neck and in some cases the patient may even become dumb. A new framework called robust integrated tetraplegia assistive (RITA) is proposed in this work for detectionPatel, Ripal and tracking the eye movement and makes easier communication between patient and caretaker. Further, the real value of eye movement and blink of the patient is recorded, processed and converted into corresponding voice output. The voice output can use the system for device automation of controlling fan, lights and basic needs. Test results show 100% detection accuracy for a distance of 13 in. Eye aspect ratio is almost in the range 0.32–0.36 when the eye is opened and value rapidly drops to a value in the range 0.09–0.13 and then raises again, indicating that blink has taken place. To improve the accuracy of detection and blinking for a distance equal to 13 in., artificial light is used as add-on service. Artificial light is used to improve the accuracy of detection as well as blinking for a distance equal to 13 in.

Structure health monitoring of concrete structures has gained more attention in the recent years due to advancement in the technology. Different methods like acoustic, ultrasonic and image processing based inspection methods have been deployed to carry out an assessment of concrete structure. In this paper, work has been carried out to monitor the health of laboratory scale concrete objects using vision-based inspection. The objective is to provide a modified image pre-processing algorithms for accurate concrete crack detection. Different image processing based algorithms reviewed from existing literature were implemented and tested to detect cracks on the surface of a 15 × 15 × 15 cm concrete cube. Due to random unevenness on the surface of concrete blocks, designing of an accurate and robust algorithm becomes difficult and challenging. Developed algorithm was applied to different images of concrete cubes. Receiver operating characteristics analysis and computation time analysis along with result images were discussed in the paper. In order to validate the applicability of developed algorithm, test results of crack detection on practical crack images are presented. Python was used to develop algorithm along with OpenCV library for image processing functions.

This paper presents the novel algorithm for automatic fire detection from still images and video sequences. Proposed technique has been using the color cue and flame flicker for detecting fire. This paper proposes a combination of two algorithms to detect fire from video clips. Firstly, the algorithm defines the method to detect fire in static images which can be called as color feature technique. Secondly, the algorithm defines to detect the fire in video sequences, which can be called as flicker technique. Furthermore, the final result is generated from the combination of both the results to get the exact fire output eliminating noise regions. Hence, in results, exact fire region is detected which satisfies both algorithms. Lastly, the buzzer is attached to our system; it will ring when the fire is detected by our algorithm. Results have been evaluated for still images from benchmark dataset. The overall performance of the proposed technique is appropriate.

Effective modeling of the human action using different features is a critical task for human action recognition; hence, the fusion of features concept has been used in our proposed work. By fusing several modalities, features, or classifier decision scores, we present six different fusion models inspired by the early fusion schemes, late fusion schemes, and intermediate fusion schemes. In the first two models, we have utilized early fusion technique. The third and fourth models exploit intermediate fusion techniques. In the fourth model, we confront a kernel-based fusion scheme, which takes advantage of kernel basis of classifiers i.e. Support Vector Machine (SVM). In the fifth and sixth models, we have demonstrated late fusion techniques. The performance of all models is evaluated with ASLAN and UCF11 benchmark dataset of action videos. We obtained significant improvements with the proposed fusion schemes relative to the usual fusion schemes relative state-of-the-art methods. (C) 2016 Elsevier Ltd. All rights reserved.

With the increase in the number of thefts, robberies and encroaching in the world, the existing security system is not sufficient. Hence to circumvent this problem the demand of bio-metric systems is increased, as they provide more dependable and effective means of identity confirmation. One such bio-metric security that has seen an uproar in the recent years is the gait identification. Gait recognition targets fundamentally to address this problem by recognizing people based on the way they walk. First of all, silhouette of the persons is extracted. Moreover, step size of the person is considered as unique feature for representing gait and effectively classify the person based on gait. Finally, features are feed to neural network for classification. The proposed approach gives better accuracy.

Handwritten character recognition has been vigorous and tough task in the field of pattern recognition. Considering its application to various fields, a lot of work is done and is being continuing to improve the results through various methods. In this paper we have proposed a system for individual handwritten character recognition using multilayer feed-forward neural networks. For the experimental purpose we have taken 15 samples of lower & upper case handwritten English alphabets in scanned image format i.e. 780 different handwritten character samples. There are two methods of feature extraction are used to construct the pattern vectors for training set. This training set is presented to the six different feed-forward neural networks namely newff, newfit, newpr, newgrnn, newrb and newrbe. The test pattern set is used to evaluate the performance of these neural networks models. The results are compared to find the accuracy in recognition of the respective models. The number of hidden layer, number of neurons in hidden layer, validation checks and gradient factors of the neural networks models are taken into consideration during the training.

The voice recognized word counter is a system that is used to count the particular word. It recognizes the isolated spoken words that we want to count. It then performs the counting operations, and displays the final answer on display. Voice recognition systems have a very strong probability of becoming a necessity in the workplace in the future. Such systems would be able to improve productivity and would be more convenient to use. The idea of a hardware that can recognize any person's voice without the training time involved in currently employed systems is a very promising one, and possibly a marketable one too. KeywordMatlab, voice reorganization, noise removable, word extraction, cross correlation, Filtering, thresholding technique

This paper includes a proposed technique for the Estimation of Skew present in the image of Gujarati Script Document using the Hough Transform technique. It includes simple pre-processing tasks like the Dilation, Erosion, and Thinning. Once these processes are applied the Final image is gone through Hough Transform and a quietly close angle is achieved. It provides promising results when applied on a wide range of images and also at different Skew Angles. This method provides less complexity in the Optical Character Recognition for the Gujarati Script. We obtain 44 % accuracy in this method.

Nowadays, Face Recognition is one of the most popular topics in Image Processing and Computer Vision. This heightened popularity is because of its non-intrusiveness, userfriendliness and immense application in fraud detection, law enforcement, surveillance and other security purposes. In this paper, we present four approaches for Facial Detection. The first approach we have exhibited is Normalized Cross Correlation which also has applications in pattern recognition, cryptanalysis, single particle analysis, and neurophysiology. For reliability, the output of correlation should be sharply peaked. The second approach, Peak to Side lobe Ratio (PSR) is used to measure the peak sharpness. The third approach uses one of the most important features of face i.e. eyes. The distance between the eyes being variable helps in classifying a person. The fourth approach Principal Component Analysis (PCA) is one of the traditional methods implemented for Face Recognition. Experimental results on GTAV database and Yale database shows that these approaches show sufficiently good results and is robust to illumination variation. Keywords— Face recognition, PCA, Normalize cross correlation, Eye distance approach, and Feature extraction.

In this paper, we have proposed approach for skew detection and correction of handwritten and printed Gujarati document using Linear Regression method/technique. Skew detection and correction is important for any recognition system as it directly affects the recognition process of characters/documents. The proposed method work involves linear regression formula for detecting angle of rotation and correcting it for printed and handwritten document/characters. With this approach for skew detection and correction we get up to 59.63% of accuracy for printed and 45.58% of accuracy for handwritten document/characters. This proposed method is simple and fast for detecting angle of rotation as well as it corrects the skewed image fast. Keywords— Character recognition system, Handwritten character recognition, Optical character recognition, Skew detection and correction, Linear regression.

The IEEE 802.15.3 is an ad hoc MAC layer suitable for multimedia WPAN applications and a PHY capable of data rates in excess of 20 Mbps. In 2.4 GHz unlicensed band, IEEE 802.15.3 specifies data rates up to 55 MBPS. It basically employs an ad hoc PAN topology, with roles for “master” and “slave” devices.In this work, it needs to determine the delay in discovering the devices by Pico Net Controller (PNC) and the most efficient antenna pattern so as to achieve maximum throughput and as a result increase in the QoS. At the beginning of each super-frame, a network beacon is transmitted which carries WPAN-specific parameters, including power management, and information for new devices to join the ad hoc network. PNC sends beacons to various devices and the device, which accepts the beacon, sends an acknowledgment frame to PNC. When the devices are connected to PNC, the piconet is formed through which PNC and devices can communicate with each other. For the performance assessment of the wireless networks using antennas, baseline models of directional antenna and isotropic antennas have been developed using the simulator OPNET Modeler 12.0. The code developed is to be such that, it can be easily extended to a real time implementation. From the results of the simulation it is observed that although the models with isotropic antennas have good throughput, the maximum is achieved by using directional antenna models. Moreover, use of directional antennas at transmitter and receiver decreases in end-to-end delay. For simulation, licence version of OPNET Modeler 12.0 is used. Keywords— PNC, QoS, piconets, DEVs, throughput

Detection of vehicles in images represents an important step towards achieving automated roadway monitoring capabilities. The challenge lies in being able to reliably and quickly detect multiple small objects of interest against a cluttered background which usually consists of road signs, trees and buildings. To this end we present a Proof of Concept Traffic monitoring application. The application counts the number of cars passing in either direction. Car detection is done using a boosted cascade of Haar features and is combined with the pyramidal KLT tracker to achieve a fast monitoring system.

In this paper, we have present the face recognition method based on partial Hausdorff distance.Normally face recognition algorithm gives poor results against pose and illumination variation.But the algorithm we have presented is robust to those conditions.We have applied transformation on face image which is robust todifferent face pose and illumination variations.Then the partial Hausdorff distance is calculated for matching after that the performance of face recognition is evaluated on different database.

—The main objective of this paper is to design a moving object detection model which can handle quick changes in illumination conditions. Two approaches were used for this purpose. First approach divide distortion in an image as color and brightness distortion then pixel having color distortion above a certain threshold value are classified as foreground pixels and rest as background. Second approach combines GMM with the fact that edges and gradient information remains invariant during illumination change. First model gives appreciable results while second model doesn't show much improvement over GMM.

This paper addresses the problem of counting thenumber of object in an image frame.This paper presents ahuman detection model, that is designed to work with object.The system proposed does learning through templates.The modelmakes use of Haar based features to form templates performsmatching of Haar-transformed images.Object can be detectedirrespective of the texture and color of their clothing as well asorientation.

Texture analysis is significant field in image processing and computer vision. Shape and texture has groovy correlation and texture can be defined by shape descriptor. Three individual approach Zernike moment, which is orthogonal shape signifier, Gabor features and Haralick features are utilized for texture analysis. Another approach is applied by aggregating all the features for texture analysis. Texture is defined by features which are extracted using Gabor filter, GLCM and Zernike moments. Classification of texture are done using back-propagation neural network. Individual approach is applied on texture images and accuracy is determined. By combining all approaches overall result is improved.

ABSTRACT In recent days, the need of biometric security system is heightened for providing safety and security against terrorist attacks, robbery, etc. The demand of biometric system has risen due to its strength, efficiency and easy availability. One of the most effective, highly authenticated and easily adaptable biometric security systems is facial feature recognition. This paper h a s covered almost all the techniques for face recognition approaches. It also covers the relative analysis between all the approaches which are useful in face recognition. Consideration of merits and demerits of all techniques is done and recognition rates of all the techniques are also compared. General Terms Image Processing, Computer Vision and Pattern Recognition. Keywords Still Face Recognition, Video Face Recognition, Biometric System. 1. INTRODUCTION In recent advance in computer vision, pattern recognition and image processing, face recognition is one of the most popular research topics. The reason behind this is that among the various biometric security systems based on finger print, iris, voice or speech, signature, etc., face recognition seems to be the most universal, non-intrusive, and accessible system. It is easy to use, can be used efficiently for mass scanning which is quite difficult in case of other biometrics, and also increases user-friendliness in human-computer interaction. Moreover, its wide range of surveillance, access control and law enforcement applications and availability of executable technologies after vigorous research in last few decades has made it gain significant attention. This paper provides the techniques used for face recognition in last few decades, its present scenario, and comparison of these techniques. Finally this paper concludes by proposing the possible future advancements.