The traditional manual roll-call attendance in university classroom and invigilator manually verify the student's identity in examination room is time-consuming and error-prone,so it is urgent to design and develo...
详细信息
The traditional manual roll-call attendance in university classroom and invigilator manually verify the student's identity in examination room is time-consuming and error-prone,so it is urgent to design and develop an efficient and intelligent automatic student identity verification *** recognition based automatic student identity authentication requires facedetection from video images with many faces before face *** this paper,MTCNN algorithm with significant effect is analyzed and improved,including the use of new activation function and non-maximum suppression method,increasing convolution layer and global average pooling layer,and adding real classroom student pictures into the training set for *** addition,a student face timing detection strategy isalso *** experimental results show that the facedetection method has improved in accuracy and detection speed,and can be well applied to facedetection in college classrooms,laying a good foundation for further research on automatic student identity verification in colleges.
facedetection is hotly discussed issues in computer vision, not just because of the difficult nature of the face as an object, mostly because of the numerous implementations that require the incremental approach of t...
详细信息
ISBN:
(纸本)9781665428644
facedetection is hotly discussed issues in computer vision, not just because of the difficult nature of the face as an object, mostly because of the numerous implementations that require the incremental approach of the facedetection program. Important progress has been made over the last 15 years due to the accessibility of data in unrestricted capturing situations (so-called'in-the-wild through the Internet, the public's initiative to establish freely accessible standards, and even success in creating robust machine vision algorithms). Because of the explosive increase of video content, the facedetection issue has attracted extensive interest among researchers. In this study, we look at the most recent advancements in real-world face detectors, beginning with the technique of the pioneering Viola-Jones face detector. This strategies are classified into two sections: rigid structures, which are taught primarily via strategies based on deep learning that are boosted or implemented, and deformable structures, which are defined by their elements and characterize the face. Fair representation techniques will be outlined in detail, as well as a few other efficient strategies that will be discussed shortly after the end. Finally, the most important resources for analyzing face detection algorithms and recent optimization efforts are addressed, as well as the potential of facedetection.
facedetection is a popular research topic to the researchers at present in the Biometric system. The aim of this research is to develop a system which can detect human face or faces from a live video streaming and wa...
详细信息
ISBN:
(数字)9781728151977
ISBN:
(纸本)9781728151977;9781728151960
facedetection is a popular research topic to the researchers at present in the Biometric system. The aim of this research is to develop a system which can detect human face or faces from a live video streaming and warn by giving the siren for ensuring the significant role in the surveillance security aspects. The core of this control system includes ATMega328p Microcontroller based Arduino Uno with Pan-Tilt mechanism and Open Source Computer Vision System (OpenCY) is the platform for the image processing system. For detecting human face or faces existing methods are Haar-Cascade, Camshift, Hausdorff distance, AdaBoost, viola jones, etc. Amongst all of this we used Haar classifier Cascade algorithm for facedetection in this research.
Navigating a large library can be an overwhelming task for many users. The challenge of locating specific books often results in wasted time and diminished user experience. This work introduces an interactive robotic ...
详细信息
ISBN:
(纸本)9798331540913;9798331540906
Navigating a large library can be an overwhelming task for many users. The challenge of locating specific books often results in wasted time and diminished user experience. This work introduces an interactive robotic system to assist library users in borrowing and returning books. The system utilizes facial recognition technology to identify and greet users upon entry into its detection range. For book borrowing, a speech recognition algorithm queries users about the books' titles they wish to borrow. It then accesses a database modeled to mimic the University of Detroit Mercy Library database. Subsequently, the robot guides users to retrieve the requested books from their corresponding bookshelves on the library map using the shortest possible path. After completing the retrieving task, the robot autonomously returns to its initial greeting position within the library. For book returns, a QR code algorithm enables users to scan the QR code on the book cover in front of the robot's camera. The borrowing and returning procedures have been tested in real-time under various scenarios to ensure reliability and effectiveness. Furthermore, an interactive library website has been developed to assist users in checking the availability and precise location of books online. This website features distinct login privileges for students and administrators, facilitating access to essential library information and services.
This paper describes an approach to teach face detection algorithms to beginner level programming learners using a facedetection tool built in Python. Learners are expected to understand and practice their Python cod...
详细信息
暂无评论