Skip to main content

Public

Project privacy set to public. By default, its content is available to everyone (authenticated or not). Please note that more restrictive permissions might exist on some items.

    •  
      Internship #158713 Differentiate the dataset of genuine and spoof images in graph plot(sklearn):
    Actions
    #158713
    SANKALP PRAJAPATI (sankalp_prajapati)
    2024-05-31 15:43
    2024-05-21 15:06
    Details
    Differentiate the dataset of genuine and spoof images in graph plot(sklearn):
    1. Create a dataset for faces, of spoof and real dataset based
    2. With cropping of 224 width and 224 height size after photo cropped only of face.
    3. Then create a spoof images that manually created by u. or generated by automatically, in the given below code.
    4. after creating the dataset, make a differentiate the spoof images and real images on graph. Note that it should be a balanced dataset. 40 of real and 40 of spoof.
    Empty
    Empty
    State of Progress
    2024-05-21
    Empty
    2024-05-22
    SANKALP PRAJAPATI (sankalp_prajapati)
    Saurabh Kumar Pandey (saurabh)
    2024-05-22
    Closed
    Attachments
    By SANKALP PRAJAPATI (sankalp_prajapati)
    (44 kB)
    lbp&hog.png
    lbp and hog
    By SANKALP PRAJAPATI (sankalp_prajapati)
    (47 kB)
    Pixel_Intensity.png
    pixel intensity
    References
    References list is empty
    Display settings

    Follow-ups

    User avatar
    SANKALP PRAJAPATI (sankalp_prajapati)2024-05-23 09:55
    1. Downloaded the data set of pictures of human faces( first tried and tested on cats and dogs) and then created spoof images from the spoof generator provided, by small alterations and stored it in separate file.
    2. Chose a balanced dataset(40 real + 40 spoof) for plotting.
    3. Now for differentiating and visualization, first used pixel intensity as a measure, but due to unsatisfactory result used binary local pattern and histogram of image for feature extraction and plotted on a graph.
    4. Learned and used orb and surf feature extraction method and plotted graph for both of these.
    5. Studied occlusion method to find out portions of the image contributing to the result set of classes.