Haar Cascade Classifier Research Articles

Self-Driving robotic car utilizing image processing and machine learning

The major goal of this paper is to build and represent a prototype of a fully autonomous car that employs computer vision to detect lanes and traffic signs without human intervention using limited computing capacity. The project contains an embedded system represented by a Raspberry Pi 3 which serves as the image processing and machine learning unit. This method requires a stream of images as input for the computer vision using OpenCV2 library with C++ programming language along with Haar Cascade Classifier for the detection of traffic signs. The Raspberry Pi will send binary signals to the Arduino UNO which is responsible for merging those signals with the ones from the ultrasonic sensor and producing new signals which are sent to the motor driver to control the direction and speed of the dc motors. The system was able to detect the lane and respond to changes in lane direction, as well as to detect traffic signs and give appropriate responses.

COMPARISON OF EIGENFACE AND FISHERFACE METHODS FOR FACE RECOGNITION

Abstract— Biometric information systems have been widely used in the fields of government, shopping centers, education and even security, which offer biological authentication so that the system can recognize its users more quickly. The parts of the human body are identified by a biometric system that has unique and specific characteristics, one of which is the face. Adjustment of facial image deals with objects that are never the same, due to the parts that can change. These changes are caused by facial expressions, light intensity, shooting angle, or changes in facial accessories. With this, the same object with several differences must be recognized as the same object. In this study, the data used were 388 face images and the sata test consisted of 30 face images. Before the face is tested, preprocessing and feature extraction are carried out using the Haar Cascade Classifier and then detected using Eigenface and Fisherface. Based on the research results, the Fisherface method is an algorithm that is accurate and efficient compared to the Eigenface algorithm. The Fisherface algorithm has an accuracy of 88%. while the Eigenface method has an accuracy rate of 76%.
 Keywords – Haar Cascade Classifier, Eigenface, Fisherface,.

Mask Covered Face Recognition Using Haar Cascade Classifier and Fuzzy Logic

Right from the beginning of the COVID-19 outbreak, everyone is aware of the havoc caused by the pandemic. To curb its spread, every healthcare agency and civic body around the globe has been advising to wear masks. However, this necessary practice has posed a significant challenge for the modern-day Facial Recognition technology. Face recognition finds significant application in the security domain that demands speed and accuracy both simultaneously. This requires the system to be highly optimized and efficient. Through this paper, we present a novel approach using Haar cascade classifier for face detection with Local Binary Patterns Histograms (LBPH) face recognizer. This work further goes on to address the various problems that occur when the user wears a mask that covers a different area and percentage coverage of the face resulting in inaccuracies as various tests come with false negatives or false positives. This problem is addressed by making use of a fuzzybased system that decides the “threshold confidence score” needed to pass the authentication dynamically. Our proposed model for masked Face recognition achieves an accuracy of 86% when a Haar-feature-based cascade classifier with LBPH face recognizer is used standalone which further improves to around 97% when used in conjunction with a fuzzy system

Detection of the Use of Mask to Prevent the Spread of COVID-19 Using SVM, Haar Cascade Classifier, and Robot Arm

In the effort to hold up the case spread of COVID-19’s growth rate by implementing health protocols such as the use of masks, supervision is needed especially for the people who have not or still have problems to wearing masks. In this research, the system utilizes the robotic power to identify visitors whether they are wearing masks or not, and automatically distribute masks if the user is detected as not wearing a mask. The user face detection process uses the Haar Cascade Classifier algorithm and SVM (Support Vector Machine) to classify users who wear masks or not. For the user who is detected as not wearing masks, myCobot-Pi with the support of suction pump will distribute masks to users. The use of myCobot-Pi as a raspberry pi based robotic arm allows the application of the system on devices that are minimal in terms of specifications and size. Through trials by taking 41 examples of detection cases, 29 cases were found that managed to detect the correct use of masks. In addition, in this study we use PP sheet plastic protector to replace the packaging of the mask because it can be carried by the suction pump properly.

Implementation of facial mask detection and verification of vaccination certificate using Jetson Xavier kit

Computer Vision (CV) based object detection system is used for locating and recognizing objects of interest in the digital images or videos/live streams. In this work, real-time face mask detection and verification of vaccination certificates are implemented using a Jetson Xavier kit. Initially, live stream video is captured through a Universal Serial Bus (USB) High Definition (HD) video camera. Then, pre-processed using the standard Python CV library to grab a frame for further processing. Using Haar-Cascade Classifier and face encoding, recognition of the face from the frame is performed and labeled appropriately, then a pre-trained Deep Learning (DL) model is used to locate the target class (mask) on the frame. The Quick Response (QR) code embedded in the vaccination certificate is also analyzed from the frame. The beneficiary and vaccination details are obtained from the website of the concerned authority. Finally, the unique QR code with the relevant information is stored in the Jetson Xavier kit for detection and retrieval during the next instance. The system’s performance is verified and found to be good in terms of accuracy compared to other methods considered in this work. The implementation of this work will cater to the local community and help achieve the nation’s sustainable development goal.

Survey Paper for Real Time Car Driver Drowsiness Detection using Machine Learning Approach

Abstract: Tiredness is a way wherein one level of mindfulness is decreased due to lacking of rest or weariness and it might cause the driver fall into rest unobtrusively. A Drowsy or lethargic driver can't decide while he/she can have a crazy rest. Nod off pulverizes are exceptionally outrageous as far as injury.Recent insights gauge that every year 1,200 passings and 76,000 wounds can be caused to fatigue or sluggishness related crashes. Over 25% of interstate car crashes are caused as consequence of driver weariness. Decrease the gamble of a mishap by advance notice the driver of his/her tiredness. This undertaking is fundamentally founded on four parts 1) Face and Eye identification: Performs scale invariant location utilizing Haar Cascade Classifier perform through a webcam. 2) Eye highlight extraction: Eye highlights are removed utilizing Hough Circle and 3) Extract single eye 4) Edge location and perform tiredness discovery on it. In the proposed technique, following the face identification step, the facial parts those are more significant and considered as the best for tiredness, are removed and followed in video succession outlines. The framework has been tried and carried out in a genuine climate. The commitment work is when tiredness recognized, after it will give alert admonition sign to the driver.

Face Authentication in Chat Application using Cryptography

Data plays a vital role in the world of computer network. There are many platforms to share data between a group of devices or individual or a group of peoples. Every person in the community needs him Information to keep safe from a third party. Here we have created a simple chat application with face authentication that can be used once to view end-to-end encryption authentication and shared message. This takes credentials to register and requires Face snap to authenticate the user. There are face snaps. Trained using the KNN Classifier. These face snaps are classified using the Haar cascade classifier algorithm for detection. User's face can be identified and logged in only if the user matches the credentials and can send messages. The receiving user is identified and the message is viewed.

Deteksi Pelanggaran Social Distancing Menggunakan Raspberry Pi Kamera Berbasis IOT

Social distancing is one of the methods used to keep yourself and others safe during a pandemic and prevent the transmission of Covid-19. With social distancing, the spread of Covid-19 can be minimized. By utilizing two tools, namely the Raspberry Pi to classify using the haar cascade classifier algorithm from data obtained from the pi camera module as a visual raspberry pi used to detect a person's body. Raspberry pi camera will start detecting using crontab unit time. If the Raspberry Pi camera detects a person who is less than one meter away, that person is said to have violated social restrictions and vice versa, the data will be stored on the Raspberry Pi which also becomes a web server. The website is used as a system user interface, Raspberry Pi camera communication uses the machine to machine (M2M) protocol which is applied to a website-based application.

AUTO-ASSISTANCE SYSTEM FOR VISUALLY IMPAIRED PEOPLE USING DEEP LEARNING

World Health Organization (WHO) outlined that there are two eighty-five million visually disabled worldwide. Among them thirty-nine million people are completely blind. One of the difficult activities that could be conducted by visually impaired is object detection which could be implemented using Machine Learning. It is an approach towards Artificial Intelligence that provides system the capacity for natural learning and development from experience without specifically programmed. It provides computer vision to the system which make decisions based on training algorithms. The chief goal of this research paper is to develop an object detection system to assist totally blind individual to manage their activities independently. Paper also compares different object detection algorithms like Haar Cascade and Convolutional Neural Network with yolov5(CNN). Haar Cascade classifier is a basic face detection algorithm which could also be trained to detect different objects whereas convolutional neural network falls under deep learning approach which could be employed for object recognition. The custom dataset is created with 2300 images consisting of 3 different classes. This comparison is being executed to find the yolov5 as a suitable algorithm for this system from the aspect of accuracy for real time scenario. Keywords- Object Detection, Computer Vision, Deep learning, Feature Extraction and Recognition , Convolution neural network.

Smart Voting System Using Deep Learning and Computer Vision

Abstract: Since, India being the largest democracy in the world. Still uses voting machines to hold elections, which comes with high costs and manual labour. But voting plays a crucial role in the election of high-ranking government officials and reflects our view of how a governing body should be formed. Investigations are conducted from time to time to troubleshoot the central voting system, increasing anonymity, credibility and security while preventing all types of fraud. In this paper, there is an implementation of a method that uses deep learning techniques which develops a smart voting system. Due to its advanced features, most researchers follow and use the Boosted Cascade framework. The enhanced Cascade Framework features help you calculate and build a classifier that works more accurately. However, this accuracy rate requires a large number of Cascade Stages to help reduce similar performance with the detection and recognition accuracy. This provides protection in the sense that the most secure voter password is verified before a vote is received on the main database of the Indian Electoral Commission and voters can verify that his or her vote has reached to correct participant of election. The votes counting is done automatically, thus saving a lot of time and the results can be announced in a very short time by the Indian Electoral Commission. The user verification process is enhanced by adding a face recognition to the app that will determine whether the voter is a certified user or not. Keywords: Online voting system, smart voting, face recognition, face detection, security, user authentication, deep learning, haar cascade classifier, computer vision.

Accident Prevention by Detecting Drivers Fatigueness

Abstract: Various studies indicate that fatigueness in drivers leads to road accidents. It can lead to serious injuries like brain damage or it can lead to death. Therefore preventing people from harming the countermeasure device is necessary as a reliable solution. This study therefore came up with a new way of explaining the driver's fatigueness. This example uses the Haar Cascade algorithm, next to the OpenCV library to keep an eye on the real-time video of the driver and critique the driver's eyes. The Eye Aspect Ratio (EAR) is used in this measurement device to determine if the eyes are open or closed. The Mouth Aspect Ratio (MAR) is also used as an important element while this model describes the driver's fatigueness as the driver begins to yawn just before the driver feels fatigue. If the driver is found to be fatigue, a warning signal is issued. Keywords: OpenCV Automated algorithms, database, Image recognition, face detection, frontal postures, extraction phase, OpenCV, data models, HaarCascade classifier, training set.

Face Recognition Door Lock System Using Raspberry PI

Abstract: Face recognition is a technique for recognizing or validating an individual's identification by looking at their face. Face recognition software can identify persons in photographs, videos, or in real time. During police stops, officers may use mobile devices to identify persons. Computer vision is a method for understanding how photos and movies are stored, as well as manipulating and retrieving data from them. Artificial Intelligence relies on or is mostly based on computer vision. Self-driving cars, robotics, and picture editing apps all rely heavily on computer vision. The software used is OpenCV Library which is useful to display and process the image produced by webcam. In this paper, we employ Haar Cascade Classifier in an image processing of user face to render the face detection with high accuracy. Keywords: OpenCV, Raspberry pi 4, Pi-Cam, Image Processing, Intelligent door lock.

SURVEY PAPER ON REAL TIME SIGN LANGUAGE INTERPRETER

Deaf and Dumb people use sign language to communicate. There are various sign recognition techniques that produce output in the form of words or identified signs. The suggested method focuses on Sign Language Interpretation in correct sentences .In addition to sign identification several NLP (Natural Language Processing) techniques are applied .Input is given as video of sign language followed by framing and segmentation on video. Deaf people get isolated as result of this. However, if an android application can be developed to convert sign language into written and audio format, the gap between normal people and the deaf community can be narrowed. The HAAR CASCADE classifier is used to identify signs. The continuous words for each sign are sent as input to the POS (Part of Speech) tagging module after sign recognition. It is utilized a word net POS tagger with its own word net dictionary. Finally, the sentence is framed using the LALR Parser. In this approach the suggested sign language Interpreter model produces intelligible sentence. This sentence is again converted into audio using gTTS API .We will be developing an android application which will scan the sign languages and interpret it into specific sentence and audio using which a normal person can understand the language of Deaf and Dumb People.

Human Face Identification Using Haar Cascade Classifier and LBPH Based on Lighting Intensity

The problem in implementing online learning during the Covid-19 era is the lack of internet access for video streaming, especially in small towns or villages. The solution idea is to minimize the video bandwidth quota by only showing emoticons. The first step of the process is the system must be able to lock the face area to be translated. This study aims to identify areas of the human face based on camera captures. The research was conducted using the Haar cascade classifier algorithm to recognize the facial area of the captured image. Then the Local Binary Pattern Histogram algorithm will recognize the identity of the face. The lighting scenario will be used as a distracting effect on the image. The results showed that based on 30 sets of images tested in bright conditions, the system was able to recognize facial identities up to 62%, normal conditions 51% and dark conditions 46%.

Investigation of Methods to Create Future Multimodal Emotional Data for Robot Interactions in Patients with Schizophrenia: A Case Study.

Rapid progress in humanoid robot investigations offers possibilities for improving the competencies of people with social disorders, although this improvement of humanoid robots remains unexplored for schizophrenic people. Methods for creating future multimodal emotional data for robot interactions were studied in this case study of a 40-year-old male patient with disorganized schizophrenia without comorbidities. The qualitative data included heart rate variability (HRV), video-audio recordings, and field notes. HRV, Haar cascade classifier (HCC), and Empath API© were evaluated during conversations between the patient and robot. Two expert nurses and one psychiatrist evaluated facial expressions. The research hypothesis questioned whether HRV, HCC, and Empath API© are useful for creating future multimodal emotional data about robot–patient interactions. The HRV analysis showed persistent sympathetic dominance, matching the human–robot conversational situation. The result of HCC was in agreement with that of human observation, in the case of rough consensus. In the case of observed results disagreed upon by experts, the HCC result was also different. However, emotional assessments by experts using Empath API© were also found to be inconsistent. We believe that with further investigation, a clearer identification of methods for multimodal emotional data for robot interactions can be achieved for patients with schizophrenia.

Survey on Vision Based Hand Gesture Interface for Controlling Multimedia Player

As computers become additional pervasive in society, facilitating natural human–computer interaction (HCI) can have a positive impact on their use. Hence, there has been growing interest within the development of recent approaches and technologies for bridging the human-computer barrier. the last word aim is to bring HCI to a regime wherever interactions with computers are going to be as natural as associate degree interaction between humans, and to the current finish, incorporating gestures in HCI is a crucial analysis space. Gestures have long been thought-about as associate degree interaction technique that may doubtless deliver additional natural, creative, and intuitive strategies for human activity with our computers. Hand gesture recognition is one amongll|one amongst|one in every of} the systems that may notice the gesture of the hand in a period of time video. The gesture of hand is classed inside a definite space of interest. during this study, planning hand gesture recognition is one among the difficult jobs that involves 2 major issues. foremost is that the detection of the hand. Another drawback is to form an indication that's appropriate to be used one hand at a time. This project concentrates on however a system might notice, acknowledge and interpret hand gesture recognition through computer vision with the difficult factors that variability within the create, orientation, location, and scale. To perform well for developing this project, differing kinds of gestures like numbers and sign languages got to be created during this system. The image taken from the period of time video is analyzed via Haar-cascade Classifier to notice the gesture of hand before the image process is finished or in different words to notice the looks of hand in a very frame. during this project, the detection of hand are going to be done mistreatment the theories of Region of Interest (ROI) via Python programming. the reason of the results are going to be targeted on the simulation half since the distinction for the hardware implementation is that the ASCII text file to scan the period of time input video. the event of hand gesture recognition mistreatment Python, OpenCV, and YOLO V3 will be enforced by applying the theories of hand segmentation and also the hand detection system that uses the Haar-cascade classifier.

A Hybrid Algorithm for Face Recognition System Based Smart Attendance System

Face recognition aims to detect, track, identify, or verify person from video or images using recognition system. Its application varies from security, authentication, social media, and so on. With the advancement of technologies, still it impedes in accuracy. The challenges involved in face recognition are variations in facial expression involve diverging light, facial changes interfered with facial hair, pose, and so on. This research work presents a new hybrid approach, which improves the accuracy in identification of a person. The accuracy of the proposed work clearly depicts that it can be best suited in real time environments for tracking persons. It can be used for applications like authentication, security, and so on. The proposed work is validated by tracking the student presence. It captures the image of the student and detect the face by pre-processing, extracting the features and recognizing the person. For recognition it uses Local Binary Pattern Histogram (LBPH) algorithm and Haar Cascade classifier. LBP Algorithm uses Histogram, which improves the accuracy of the face recognition. It is integrated with Haar classifier, which uses machine learning algorithm. Adaboost Learning algorithm, which selects small features from large set, which improves the efficiency of recognition.

FACE RECOGNITION BASED ATTENDANCE MONITORING SYSTEM

In this digital era, face recognition system plays a significant role in nearly each sector. Face recognition is one in all the principally used bioscience. It will used for security, authentication, identification, and has more blessings. Despite of getting low accuracy when put next to iris recognition and fingerprint recognition, it's being wide used because of its contactless and non- invasive method. what is more, face recognition system also can be used for attending marking in colleges, colleges, offices, etc.This method aims to make a category attending system that uses the construct of face recognition as existing manual attending system is time consuming and cumbersome to take care of. And there could also be possibilities of proxy attending. Thus, the requirement for this method will increase.This method consists of 4 phases- info creation, face detection, face recognition, attending change. info is formed bythe pictures of thestudents in school. Face detection and recognition is performed exploitation Haar-Cascade classifier and native Binary Pattern bar chart algorithmic rule severally. Faces square measure detected and recognized from live streaming video of the room.attending are going to be mail-clad to the various college at the top of the session. keywords—Automatic Attendance, Face Detection, Face Recognition, Image Enhancement, Enrollment,

Systematic development of real-time driver drowsiness detection system using deep learning

Advancements in globalization have significantly seen a rise in road travel. This has also led to increased car accidents and fatalities, which become a global cause of concern. Driver's behavior, including drowsiness, contributes to many of the road deaths. The main objective of this study is to develop a system to diminish mishaps caused by the driver's drowsiness. Recently deep convolutional neural networks have been used in multiple applications, including identifying and anticipate driver drowsiness. However, limited studies investigated the systematic optimization of convolutional neural networks (CNNs) hyperparameters, which could lead to better anticipation of driver drowsiness. To bridge this gap, a holistic approach based on the deep learning method is proposed in this paper to anticipate the drivers' drowsiness and provide an alerting mechanism to prevent drowsiness related accidents. To ensure optimal performance achievement by the system, a database of real-time images preprocessed via Haar cascade's classifiers is used to systematically optimize the CNN model's hyperparameters. Different metrics, including accuracy, precision, recall, F1-score, and confusion matrix, are used to evaluate the performance of the model. The training evaluation results of the optimal model achieved an accuracy of 99.87%, while the testing results accurately classify the drowsy driver with 97.98%.

Face detection and recognition in organic video: A comparative study for sport celebrities database

The present paper reports on an experimental study carried out under the applicative field of organic video processing, and relates to the possibility of identifying sport celebrities (soccer players) in video content. In contrast to common state-of-the-art studies, a special attention is paid on the cases in which the face is not completely included in the frame (lateral views, partial occlusions, etc.) and/or in which arbitrarily lighting conditions occur. To this aim, we consider two conventional types of face detection algorithms (Haar Cascade Classifier, and MMOD - Max-Margin object detection) coupled to two conventional face recognition models (LSBH - Local binary pattern histogram, and CNN-based Pruned ResNet). The experimental work consists of evaluating the end-to-end performances of the four possible combinations among the above-mentioned two face detection and two face recognition methods. A database of 20 video sequences of about 3 minutes each is organized. As an overall conclusion, we brought to light that the MMOD coupled to a Pruned ResNet model seems to better suit the organic video processing use-case constraints, being able to reach a recognition rate of 98%.