Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd Logo
Upload

Welcome to Scribd!

  • 37 views

    Capstone PPT 4

    Uploaded by

    Shweta Bhargude
    This document presents a project on efficient human motion and object detection. It discusses implementing algorithms for face detection, motion detection, and object classification using video analytics on an edge device. The proposed system would detect faces, motion, and objects to apply in security, retail, agriculture, healthcare and manufacturing. It provides an introduction, motivation, problem statement, literature review on related work, proposed system overview and architecture with diagrams, and concludes with future scope. The team aims to enhance security applications by integrating these three algorithms according to customer needs.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd

    Capstone PPT 4

    Uploaded by

    Shweta Bhargude
    37 views17 pages
    This document presents a project on efficient human motion and object detection. It discusses implementing algorithms for face detection, motion detection, and object classification using video analytics on an edge device. The proposed system would detect faces, motion, and objects to apply in security, retail, agriculture, healthcare and manufacturing. It provides an introduction, motivation, problem statement, literature review on related work, proposed system overview and architecture with diagrams, and concludes with future scope. The team aims to enhance security applications by integrating these three algorithms according to customer needs.

    Copyright

    © © All Rights Reserved

    Available Formats

    PPTX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document presents a project on efficient human motion and object detection. It discusses implementing algorithms for face detection, motion detection, and object classification using video analytics on an edge device. The proposed system would detect faces, motion, and objects to apply in security, retail, agriculture, healthcare and manufacturing. It provides an introduction, motivation, problem statement, literature review on related work, proposed system overview and architecture with diagrams, and concludes with future scope. The team aims to enhance security applications by integrating these three algorithms according to customer needs.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    37 views17 pages

    Capstone PPT 4

    Uploaded by

    Shweta Bhargude
    This document presents a project on efficient human motion and object detection. It discusses implementing algorithms for face detection, motion detection, and object classification using video analytics on an edge device. The proposed system would detect faces, motion, and objects to apply in security, retail, agriculture, healthcare and manufacturing. It provides an introduction, motivation, problem statement, literature review on related work, proposed system overview and architecture with diagrams, and concludes with future scope. The team aims to enhance security applications by integrating these three algorithms according to customer needs.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    of 17

    MIT POLYTECHNIC, PUNE

    DEPARTMENT OF COMPUTER ENGINEERING

    Project Title: Efficient Human Motion And Object Detection

    Presented By:
    Team Members name-
    1.Shweta Shinde.
    2.Neha Mahadik.
    3.Shweta Bhargude.
    4.Neha Pawar.

    Guide:
    Prof. M. S. Aware
    Contents
     Introduction
     Motivation
     Problem Statement
     Literature Review
     Existing System
     Proposed System
     System Architecture
     Diagram( DFD, Use Case, Activity)
     Conclusion
     Future Scope
     References
    Introduction.
    • Face detection is a computer technology being used in a variety of applications that identifies human
    faces in digital images.
    • Motion detection is the process of detecting a change in the position of an object relative to its
    surroundings or a change in the surroundings relative to an object.
    • Object detection is a computer technology related to computer vision and image processing that deals
    with detecting instances of semantic objects of a certain class (such as humans, animals, or cars) in
    digital images and videos.
    Motivation.
    • The main motivation behind this project is being able to surveil a particular area and detect
    motion, facial features and objects in the environment.
    • The basic idea behind this project is to detect and identify human beings as well as objects in
    any given environment.
    Problem Statement.
    • Implementation of an edge agent for video analytics to be used in VIMEO applications.
    • It includes implementing and developing algorithms to detect motion and objects in a given
    environment. It also includes execution of algorithms for facial detection and recognition.
    • The idea is to extract a frame from a live video stream and to analyze it using the above
    mentioned algorithms.
    Literature Review.
    Sr.No Paper Name Year of Advantages Disadvantages
    Publication

    1 Human Face 2016 Background subtraction reduces The proposed system fails to
    Detection and search area to detect feature with detect face if human present is
    Recognition in Videos additional computational standing stationary for more
    complexity. Eigenface method time.
    greatly reduces the computations by
    projection image from
    twodimension to one-dimension

    2 A Contour based Procedure for 2016 Algorithm allows to detect faces in Multiple detection recognition face
    Face Detection and all the ranges of moving human needs work. and
    Tracking from face.
    Video
    3 A Surveillance System Based on 2016 Motion is 3D motion estimation with tracking,
    Motion Detection and detected with a new approachedge recognition and classification of moving
    Motion Estimation using region determination which makes objects is not supported by the proposed
    Optical Flow detection faster. system.
    4 Motion detection based on the 2015 The results demonstrate that the The problem of the scene
    combining of the background proposed method can well illustration change is not
    subtraction and spatial color fulfill the solved.The maintenance of the
    information background segmentation in temporal and permanent
    different situations and is robust to variation in background is one
    illumination change. of the principal difficulty

    5 Visual 2015 The real time alert system plays an The captured images are getting
    Surveillance important role in real time so as to stored on HDD which is not secure
    Using Absolute observe the motion of an as compared to storing on cloud.
    Difference object in the motion
    Motion restricted area
    Detection and immediately the user is getting
    the SMS containing text as Motion
    Detected.

    6 Moving Object 2014 The strength of the proposed system The system does not work on
    Detection, Classification and its lies in the ability to separate complex videos like disaster
    Parametric background and foreground in an videos for object detection.
    Evaluation accurate way and it is
    implemented in simple way with low
    time consumption.
    7. A Fast and Robust face Detection 2014 The proposed algorithm can There are a lot of
    and Tracking Algorithm robustly detect and track human improvements to be done on
    face, even in the event of occlusion, face regional technique.
    hence suitable for realtime video
    surveillance

    8. Real-time Object Detection and 2011 The algorithm gave a 99 percent Algorithm does not provide path
    Tracking in an Unknown Environment accuracy in an unknown of motion.
    environment. Fast and Robust.
    Proposed System

    IoT embedded devices are already used as either primary or supplementary eyes for everything
    from driverless cars and traffic monitoring to smart cities and transportation centers. Object
    detection is useful for applications in various industries:
    Retail
    Security
    Agriculture
    Healthcare
    Manufacturing
    •Classification applies a prediction to an image based on an analysis of the contents.
    •Detection locates objects within an image.
    SYSTEM ARCHITECTURE
    DFD Diagram
    Use Case Diagram
    Activity Diagram
    Conclusion.
     We will be able to detect faces, motion and objects. Face and motion detection
    is provided whereas object classification is also present.
     We can apply these three algorithms to various areas according to customer requirements.
    Enhancement of security.
    Future Scope
    Face detection can be improved to detect multiple faces from a side role as well. A model can be
    trained to recognize the detected faces and classify them.
    The scope of the object detection model can be increased. The model can be trained using images
    of different objects. The images may include positive as well as negative images.
    Motion detection module can be further extended to recognize what motion or action can be
    performed. For Example: Standing, Raising hands
    Selection of Region of Interest is applied to a fixed camera. Efforts can be made to select ROI
    from a moving camera.
    References.

     securitymagazine.com/articles/89084-the-many-uses-of-video-analytics
     software.intel.com/en-us/articles/OpenVIMO-RelNotes
     https://software.intel.com/en-us/openvino-toolkit/model-optimizer-developer-guide
    THANK YOU

    You might also like