Local Binary Pattern Histogram Research Articles

Face Detection and Recognition Using Raspberry PI Computer

This paper presents a face detection and recognition system utilizing a Raspberry Pi computer that is built on a predefined framework. The theoretical section of this article shows several techniques that can be used for face detection, including Haar cascades, Histograms of Oriented Gradients, Support Vector Machine and Deep Learning Methods. The paper also provides examples of some commonly used face recognition techniques, including Fisherfaces, Eigenfaces, Histogram of Local Binary Patterns, SIFT and SURF descriptor-based methods and Deep Learning Methods. The practical aspect of this paper demonstrates use of a Raspberry Pi computer, along with supplementary tools and software, to detect and recognize faces using a pre-defined dataset.

AI-based face recognition system with telegram notification for room security on raspberry PI

This research is based on the importance of a security system in a room by implementing AI combined with the telegram notification system. The goal is that security information can be obtained quickly and in real-time. The methodology used is to design a hardware system consisting of input, process and output devices. The input device consists of a Logitech C270 camera mounted on 2 MG966R type servo motors so that the camera can rotate on the X and Y axes, then the results of the camera captures are processed using the Haar Cascade Classifier and Local Binary Pattern Histogram (LBPH) algorithms. Raspberry Pi 4 is used as a data processing center and push notification to telegrams in the form of images when faces are detected by a web camera. Only registered users may enter the room, by opening the door when a face is recognized. Our findings show that a room security system with an AI-based facial recognition application can be implemented, according to the planning and design results in this study. The door opening process produces an average result of 4.586 seconds, with the longest time being 4.981 seconds and the fastest time being 4.116 seconds. The door closing process produces an average result of 4.496 seconds, with the longest time being 4.966 seconds and the fastest time being 4.106 seconds. The average time of opening and closing the door is ideal and safe. From the results of the research that has been done, it can be concluded that the use of AI in this study aims to make decisions that only registered users can enter a room. In addition, the ability of the camera to move dynamically on the x and y axes is one of the system developments that did not exist before, so that the ability to take pictures besides being more accurate also becomes wider dynamic.

Artificial Intelligence Based Domotics Using Multimodal Security

All electronic devices in our cutting-edge technology world must be networked together via the Internet if users want to have remote access to them. As a result, it may raise a variety of serious security issues. This study suggests a remote access home automation security system that incorporates utilizing the Internet of Things (IoT), and Artificial Intelligence (AI) for ensuring the security of the house. For a highly efficient security system, Face recognition has been used to maneuver the door access. In case of power outage or for any technical issues, an alternative security PIN has been added which is only accessible by the owner. Moreover, individuals are able to monitor and control the door access along with other attributes of the house using an application. In this work, Face detection is performed using the Haar Cascade classifier, while face recognition is performed using the Local Binary Pattern Histogram (LBPH). 95.7% accuracy in recognizing faces has been achieved after evaluating the proposed system.

3D Road Lane Classification with Improved Texture Patterns and Optimized Deep Classifier.

The understanding of roads and lanes incorporates identifying the level of the road, the position and count of lanes, and ending, splitting, and merging roads and lanes in highway, rural, and urban scenarios. Even though a large amount of progress has been made recently, this kind of understanding is ahead of the accomplishments of the present perceptual methods. Nowadays, 3D lane detection has become the trending research in autonomous vehicles, which shows an exact estimation of the 3D position of the drivable lanes. This work mainly aims at proposing a new technique with Phase I (road or non-road classification) and Phase II (lane or non-lane classification) with 3D images. Phase I: Initially, the features, such as the proposed local texton XOR pattern (LTXOR), local Gabor binary pattern histogram sequence (LGBPHS), and median ternary pattern (MTP), are derived. These features are subjected to the bidirectional gated recurrent unit (BI-GRU) that detects whether the object is road or non-road. Phase II: Similar features in Phase I are further classified using the optimized BI-GRU, where the weights are chosen optimally via self-improved honey badger optimization (SI-HBO). As a result, the system can be identified, and whether it is lane-related or not. Particularly, the proposed BI-GRU + SI-HBO obtained a higher precision of 0.946 for db 1. Furthermore, the best-case accuracy for the BI-GRU + SI-HBO was 0.928, which was better compared with honey badger optimization. Finally, the development of SI-HBO was proven to be better than the others.

Evaluation Method of Real-Time Face Detection

Object detection is one of computer technology using image or video. Face detection is closely related Image processing and computer vision are used to detect several objects including human faces, landscapes, cars, etc. Face detection algorithm aims to confirm if an image has a face as the object in it. In this study, face detection uses several methods, namely the Eigenface method, the Fisherface method, and the Local Binary Pattern Histogram (LBPH) method. This study used 10 different subjects. The test was carried out 15 times using each face detection method with constant distance. The face detection process in this study was simulated using JupyterLab. The result showed that LBPH method obtained the highest level of accuracy in between comparison among Fisherface method and the Eigenface method. The accuracy of the LBPH method is 93.90%, while the Eigenface method is 85% and the Fisherface method is 53.33%. Differences in face detection accuracy were found due to the low level of lighting in the room and the use of accessories on the subject.

Face Recognition Based Attendance System

Abstract: If attendance is managed manually, it can be quite taxing on the teachers. Smart and automated attendance management systems are being used to address this issue. This architecture makes it simple to address the issue of proxies and students being counted as present even though they are not. The system implemented makes use of the LBPH face recognizer to instantly recognise a person's face. Light has an impact on both Eigen and Fisher faces, and real-world lighting conditions cannot always be ideal. This issue can be solved by a change in the LBPH faces. This method decides who is present and who is not by comparing the images from the test and training sessions. A system-updated excel sheet that contains the attendance information is kept in the database. The faculty can get a copy of the attendance sheet through mail by clicking auto mail button in the GUI. The face recognizer detects faces using Haar Cascade Classifier, and recognizes the faces using Local Binary Patterns Histograms (LBPH) algorithm.

Face Recognition Based Attendance System

In colleges, universities, organizations, schools, and offices, taking attendance is an important tasks that must be done on a daily basis. The majority of the time, it is done manually, such as by calling by name or by roll number. The goal of this project is to create a Face Recognition-based attendance system that will turn this manual process into an automated one making our task easy. This project meets the requirements for bringing modernization to the way attendance is handled, as well as the criteria for time management. This device is installed in the classroom, where and student's information, such as name, roll number, class, section, and photographs, is trained and is also available for teachers and staff attendance. The images are extracted using Open CV a library of python. Before the start of the corresponding class, the student can approach the machine, which will begin taking pictures and comparing them to the qualified dataset. The image is processed as follows: first, faces are identified using a Haarcascade classifier, then faces are recognized using the LBPH (Local Binary Pattern Histogram) Algorithm, histogram data is checked against an established dataset, and the device automatically labels attendance. An Excel sheet is developed, and it is updated every hour with the information from the respective class instructor.

Combining deep features and hand-crafted features for abnormality detection in WCE images

In this paper, a computer-aided method is proposed for abnormality detection in Wireless Capsule Endoscopy (WCE) video frames. Common abnormalities in WCE images include ulcers, bleeding, Angiodysplasia, Lymphoid Hyperplasia, and polyp. In this paper, deep features and Hand-crafted features are combined to detect these abnormalities in WCE images. There are not sufficient images available to train deep structures, therefore the ResNet50 pre-trained model is used to extract deep features. Hand-crafted features are associated with color, shape, and texture. We used a novel idea to reveal unexpected color changes in the background due to existing lesions as a color feature set. Histogram of gradient (HOG) and local binary pattern (LBP) were used respectively for shape and texture features. They are extracted from the region of interest (ROI), i.e. suspicious region. The expectation Maximization (EM) algorithm is used to extract more distinct areas in the background as ROI. The expectation Maximization (EM) algorithm is configured in a way that can extract areas with a distinct texture and color as ROI. The EM algorithm is also initialized with a new fast method which leads to an increase in the accuracy of the method. A large number of features are created by the method, so the minimum redundancy maximum relevance approach is used to select a subset of more effective features. These selected features are then fed to a Support Vector Machine for classification. The results show that the proposed approach can detect mentioned abnormalities in WCE frames with the accuracy of 97.82%

A Novel Approach Towards Development of Automated Intelligent Lifesaving Buoy

The main goal of this integrated application is to design an intelligently controlled and automated lifebuoy that can detect living people and save them from drowning. The article gives an overview of the development of a battery-powered remote-control boat for various applications. The Lifebuoy is a revolutionary upgrade of an invention created 300 years ago, including a remote control, thrusters, and a battery that can be sent to someone in the water in need. It's a clever lifesaver that gives these lifebuoys a U-shape to give a person who is drowning or injured and unable to swim a good grip to keep and stay in the U-shaped area of the buoy remain. The U-shaped lifebuoy can be moved through the water by remote control, allowing it to reach the victim quickly and bring them to safety if necessary. The U-shaped remote-controlled buoy can work in adverse conditions thanks to its navigation and guidance systems. The remote-controlled submarine buoy is a quick and effective way to save lives. Once it reaches the victim, the U-shaped buoy has enough power to carry it to safety, which can be very useful on large ships that would otherwise have to launch lifeboats. It also allows those on board to start the device without having to turn around, saving time at the beginning of a rescue operation. The waves can't stand up to the U-shaped buoy that you'll see racing by to reach your destination. Water is a dangerous place, and despite the fact that a remote-controlled U-shaped buoy is a lifesaver, people are easy to reach. Object tracking is an important task in many image-processing applications. Optical flow is one of the most widely used image processing and video analysis techniques. This article implements an object-tracking algorithm based on the optical flow method for computation on a Raspberry Pi microcomputer. The Lucas-Kanade method was used to compute the velocity vector of an object moving between two consecutive frames. This paper represents a face recognition mechanism carried out as part of the developing an intelligent lifebuoy. It utilises the technologies available in the Open Computer Vision (OpenCV) library and the methodology for their implementation with Python. Haar Cascades was used for face detection, and eigenfaces, fisherfaces, and local binary pattern histograms were used for face recognition. An experiment was performed to evaluate the robustness of the proposed algorithm against the new computing device. The results were encouraging for the use of the proposed real-time application in a variety of contexts.

Track A Missing Child Using Local Binary Pattern Histogram Based Face Recognition Method

In our society, a countless number of children are missing every-day. Even though missing cases are getting filed against them in nearby police stations but it is complicated to trace them through investigation as it requires lot of efforts and it is time consuming also it becomes quite challenging for police to find them and it’s really getting impossible to find them in most cases. So to make it convenient we planned to design a website for “Tracking the missing child using Face Recognition ”, which makes the task quite simpler. The main intention of this project is to reunite the missing children with their parents, so we are going to design a web page where one can be able to upload the missing child details and store it in database. If the child who is missing is identified on road by any random person, then he/she will upload the photo of that child on web page then our system can be able to detect that lost child using face recognition algorithms. Whenever the system confirms the matching then it will set up custom alerts along with location and send them directly to family members and concerned investigation officers . And if the match is not found , then the person will be provided with the option of registering child photo as a new entry to our database with the location they found. So, in this way we can find the lost children in more easy and convenient way. In this model, an automated facial recognition system for Missing Child database was proposed using known Haar feature-based cascade classifier. This system will be able to detect face and recognize face automatically in real time. In our model we are going to use Face Recognition module of python.

Attendance System Using Deep Learning

In the traditional system, it is difficult to handle the attendance of a huge number of students in the classroom. Attendance of the students is very important for every university, college as well as school as it keeps the record of every individual. The conventional methodology of taking attendance by calling out the roll number or name of the student and then recording the attendance is a tedious task. Assume the duration of the lecture is 60 minutes or 1 hour and to record the attendance it would take 5 to 10 minutes. Also for every tutor, this is time-consuming. To overcome this problem, real-time face recognition is a real-world solution, which will help in marking the attendance of students in bulk. In this project, face detection and face recognition are used. Attendance marking through face recognition can be implemented in the classroom by capturing the image of students in the classroom via a surveillance camera. Later Haar Cascade algorithm is used for face detection, which helps in locating the position of the face region, and LBPH (Local binary pattern histogram) algorithm for face recognition. The images of all the students in the class are stored in the database and when the face of the individual student matches with one of the faces stored in the database then the attendance is recorded

FACE RECOGNITION ATTENDANCE SYSTEM USING LBPH ALGORITHM

The world is getting refreshed quicker and quicker however a few things are not being refreshed; one of them is the Attendance framework. In ancient days the attendance is taken by signature on the paper. This process is a time consuming and burdensome process. Now some techniques came like RFID based attendance system but there is a chance to cope with the occurrence of malpractice. In this project, new technology is proposed namely Face recognition attendance system. In this system, a camera is placed, that camera will take a picture of the person and compares it with the database. If the image matches the database then attendance is given to that person. In this, the comparison is done by using the Local Binary Pattern Histogram(LBPH) Algorithm. Previously Principal Component Analysis(PCA) is used, compared to PCA, LBPH is the most efciently working algorithm. By using the LBPH algorithm this system gives the best results.

Development of an Employee Attendance System Using Face Recognition and QR-Code Technology

Abstract: The procedure of recording attendance at any organization, including polytechnics and other institutions, is crucial to demonstrating why a particular employee is exceptional. Traditional attendance management systems, which use attendance sheets and signatures, have shown to have some associated issues, including time wastage, impersonation, and attendance sheet misplacement, making the system wasteful and unproductive. Appraising Non-academic staff in Yaba College of Technology has been a great challenge due to the aforementioned problem confronting the management to properly monitored the attendance system put in place with the traditional methods. To solve these problems, the study developed a computerized attendance system that implements face recognition and QR-code was designed and implemented using python programming language, MySQL server database, Tkinter framework was used to build the interface and Open Camera Vision (OpenCV) library. The algorithms for face detection and recognition include Haar cascasde algorithm and Local Binary Pattern Histogram (LBPH) algorithm respectively. The system is equipped with the capability to send instant email attendance reports to the management on daily, weekly or monthly basis. The result findings shows that there was 98.8% detection and recognition rates and 0.12% errors encountered for both face and QR-code. The highest read time was also measured to be 210.30ms. The result finding from our test shows the efficiency and effectiveness of the attendance system and it is therefore recommended for use in Yaba College of Technology for attendance monitoring of Non-academic staff for onward appraisal by the HOD, Dean and the college management.

SAFE: Security Door Lock System Using Haar-Cascade and LBPH Method

Door lock provides numerous benefits and has become indispensable in daily life as it acts as a security guard to prevent burglars and protect home belonging safely. The unlock methods of existing door lock system widely use keys and thumbprint involve touching the object may cause the spread of COVID-19. In this paper, a computer vision based security door lock system using Raspberry Pi (called SAFE) is proposed. Haar-Cascade classifier is employed as face detection classifier, while Local Binary Pattern Histogram (LBPH) is proposed as face recognition classifier. Recognition result is processed based on the usage of user to provide insights of SAFE. The accuracy of SAFE using pre-trained LBPH classifier achieves average of 86% based on the data obtained. The recognition speed outperforms existing work using principal component analysis and eigenfaces.

Face Recognition with Deep Neural Network in Real-World Scenes

The primary aim of this paper is to recognize faces using deep learning models. This method is employed in many areas such as human-computer interaction, automatic image indexing, and ID verification services. Despite its many uses, face recognition still has several difficult features, such as the head position, age, lighting, and facial emotions. In this paper, our method includes three components which are getting the datasets, training the face model, and recognizing faces. Three steps are included in our process. First, the authors introduce the Haar-like features and the first module-get the database by detecting the face, removing the background, and creating the grayscale images of face, then the authors introduce the second module by using LBPH method (Local Binary Patterns Histogram) to detect differences between the model sample face and the detected face. Last, the authors use face recognition predictor to return the recognition result and confidence. Finally, our method successfully recognizes people and show the confidence value at the same time. The experimental results show great potential of our method in face recognition with high accuracy.

Local Binary Pattern Histogram Based Web Facial Authentication System

The objective of the project is to Detect and Recognize Human Facial Features via a Camera. The basic concept of this project is:- To convert the input image to HSV format to extract the binary image and additionally remove the noise from binary image using morphological operations and finally then contours are used to segment out the region of interest and further the region is analyzed to get the final result. While biometric data is generally considered one of the most reliable authentication methods, it also carries significant risk. That’s because if someone’s credit card details are hacked, that person has the option to freeze their credit and take steps to change the personal information that was breached. What do you do if you lose your digital ‘face’? Around the world, biometric information is being captured, stored, and analyzed in increasing quantities, often by organizations and governments, with a mixed record on cybersecurity. A question increasingly being asked is, how safe is the infrastructure that holds and processes all this data? As facial recognition software is still in its relative infancy, the laws governing this area are evolving (and sometimes non-existent).

ANALYSIS OF APPLICATION HAAR CASCADE CLASSIFIER AND LOCAL BINARY PATTERN HISTOGRAM ALGORITHM IN RECOGNIZING FACES WITH REAL-TIME GRAYSCALE IMAGES USING OPENCV

Face detection and recognition systems have been developed with the various application of algorithms. Based on the literature study that has been carried out, the researcher will analyze the performance between the HCC (Haar Cascade Classifier) ​​and LBPH (Local Binary Pattern Histogram) Algorithms with real-time grayscale images using the OpenCV library. The test is carried out based on a sample of facial images with external conditions in the form of lighting conditions which are divided into morning, afternoon, and evening, as well as varying face rotation angles. Parameters observed were accuracy values, FAR (False Acceptance Rate), and FRR (False Rejection Rate). Based on the results of the tests that have been carried out, the average value of accuracy is 56%, while the average value of FAR is 22% and FRR is 23%. Judging from the average accuracy value obtained is 56%, then to be able to be detected and recognized properly the face position must be in frontal condition and with normal lighting. Thus, the final results of this study can be considered for other researchers who want to use a similar algorithm to develop a detection and recognition system.

HAND GESTURE AND DETEKSI WAJAHDETECTION USING RASPBERRY PI

Face detection is currently used for various purposes, one of which is to record employees attendance. This strategy is ineffective since the employees still can hack the attendance by making their own photos and put them in their desks. If they are unable to come to the office,they can always ask their colleagues to submit their already available photos.Therefore, an alternative that can complement the current face detection method is highly needed.One of the methods that can be used is hand gesture detection.This study aims to detect hand gestures made by the employees to ensure whether they really come to work or not,so the chance for manipulation is quite small.For the purpose of hand gesture recognition, this study utilized Local Binary Pattern Histogram algorithm. LBPH is an algorithm used for the image matching process between images that have been given training and images taken in real time.The hand gesture image was first taken using a raspberry pi camera and then processed to examine whether it matches the registered ID or not.The results showed that ID recognition by using hand gestures is detectable and is in accordance with the registered ID.The number recognition in hand gestures includes numbers 1 to 10. The test results showed that, the average time required for reading hand gestures using a laptop was 9.2 seconds, while that of using raspberry was 14,2 seconds.Motion reading using a raspberry takes longer than that of using a laptop because the laptop's performance is higher than that of a raspberry.

Machine learning and uLBP histograms for posture recognition of dependent people via Big Data Hadoop and Spark platform

For dependent population, falls accident are a serious health issue, particularly in a situation of pandemic saturation of health structures. It is, therefore, highly desirable to quarantine patients at home, in order to avoid the spread of contagious diseases. A dedicated surveillance system at home may become an urgent need in order to improve the patients’ living autonomy and significantly reduce assistance costs while preserving their privacy and intimacy. The domestic fall accident is regarded as an abrupt pose transition. Accordingly, normal human postures have to be recognized first. To this end, we proposed a novel big data scalable method for posture recognition using uniform local binary pattern (uLBP) histograms for pattern extraction. Instead of saving the pixels of the entire image, only the patterns were kept for the identification of human postures. By doing so, we tried to preserve people’s intimacy, which is very important in ehealth. To our knowledge, our work is the first to use this approach in a big data platform context for fall event detection while using Random Forest instead of complex deep learning methods. Application results of our conduct are very interesting in comparison to complex architectures such as convolutional deep neural networks (CNN) and feedforward deep neural networks (DFFNN).

CAD System Based on Face Mask Recognition for Respiratory Infections Diseases Hospital

The infection of respiratory diseases can be eliminating and controlling by wearing face mask in contaminated areas. However, to control people about wearing face mask it has been challenging unless the automatic recognitions are applied. Therefore, in this paper, a Face Mask Recognition System by Computer Aided Design (CAD) is introduced. The proposed design system is based on face, mouth and nose detections in captured image. The CAD system considers to be implemented for specialized respiratory diseases hospital with different departments, each department controlled by separated door. The main goal of this paper is to design system based on software programs that helps reduce the spread of respiratory diseases and controlling wearing face mask inside respiratory infection diseases hospital based on mask detection and mask color detection. The proposed system designed for hospital with three respiratory diseases departments and three mask color applied each mask color for each department. The mask recognition system has been used cascaded object detector that is Local Binary Pattern Histogram LBPH algorithm, then color detection as artificial intelligence-based method, Red, Green, and Blue (RGB) color of the face mask images. Finally, by using of Convolutional Neural Network (CNN), the classification accuracy of color recognition achieved 100%, and also the whole system functionality tested successfully obtained all results by testing accuracy 95%. The hardware designing circuit simulation in Proteus software were obtained to control the systems of hospital department doors based on the results obtained from MATLAB software.