Scribd Logo
Upload

Welcome to Scribd!

  • 23 views

    CV Lab 3 Muhammad Umer Siddiq

    Uploaded by

    hik um
    This lab report discusses applying histogram equalization to an image. The student first performs histogram equalization manually by calculating the original histogram, normalizing it to a PDF, calculating the CDF, using the CDF to derive an equalization function, and applying the function to equalize the image. They then apply histogram equalization using the cv2 function and compare the results. In the conclusion, they summarize that histogram equalization improves contrast by mapping pixel intensities to a more uniform distribution derived from the image's CDF.

    Copyright:

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

    CV Lab 3 Muhammad Umer Siddiq

    Uploaded by

    hik um
    23 views7 pages
    This lab report discusses applying histogram equalization to an image. The student first performs histogram equalization manually by calculating the original histogram, normalizing it to a PDF, calculating the CDF, using the CDF to derive an equalization function, and applying the function to equalize the image. They then apply histogram equalization using the cv2 function and compare the results. In the conclusion, they summarize that histogram equalization improves contrast by mapping pixel intensities to a more uniform distribution derived from the image's CDF.

    Copyright

    © © All Rights Reserved

    Available Formats

    DOCX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This lab report discusses applying histogram equalization to an image. The student first performs histogram equalization manually by calculating the original histogram, normalizing it to a PDF, calculating the CDF, using the CDF to derive an equalization function, and applying the function to equalize the image. They then apply histogram equalization using the cv2 function and compare the results. In the conclusion, they summarize that histogram equalization improves contrast by mapping pixel intensities to a more uniform distribution derived from the image's CDF.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    23 views7 pages

    CV Lab 3 Muhammad Umer Siddiq

    Uploaded by

    hik um
    This lab report discusses applying histogram equalization to an image. The student first performs histogram equalization manually by calculating the original histogram, normalizing it to a PDF, calculating the CDF, using the CDF to derive an equalization function, and applying the function to equalize the image. They then apply histogram equalization using the cv2 function and compare the results. In the conclusion, they summarize that histogram equalization improves contrast by mapping pixel intensities to a more uniform distribution derived from the image's CDF.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    of 7

    Computer Vision

    Lab Report # 3

    Name: Muhammed Umer Siddiq


    Reg no: 296908
    Degree: DE 41 EE B

    Task: Apply Histogram Equalization, first on your own, then using the cv2
    function.

    Code:
    import sys

    if "google.colab" in sys.modules:
        import subprocess

        subprocess.call("apt-get install subversion".split())
        subprocess.call(
            "svn export https://github.com/YoniChechik/AI_is_Math/trunk/
    c_02a_basic_image_processing/Unequalized_Hawkes_Bay_NZ.jpg".split()
        )

    import numpy as np
    import matplotlib.pyplot as plt
    import cv2

    figsize = (7, 7)
    # read as grayscale
    I = cv2.imread("/content/drive/MyDrive/Colab Notebooks/abc.jpg", 0)

    plt.figure(figsize=figsize)
    plt.imshow(I, cmap="gray", vmin=0, vmax=255)
    plt.title("Original image")
    plt.show()
    # calculating and showing the original histogram

    bins_edges_min_max = [0, 256]
    num_bins = 256
    bin_count, bins_edges = np.histogram(I, num_bins, bins_edges_min_max)
    bins_start = bins_edges[:-1]
    def draw_hist(x_axis, input):
        fig, ax = plt.subplots(figsize=figsize)
        plt.bar(x_axis, input, width=input.shape[0] / (x_axis[-1] - x_axis[0] 
    + 1))
        return fig, ax

    draw_hist(bins_start, bin_count)
    plt.title("Original histogram")
    plt.show()
    # Normalizing the histogram to get PDF

    pdf = bin_count / np.sum(bin_count)

    draw_hist(bins_start, pdf)
    plt.title("Original PDF")
    plt.show()
    # Calculating CDF

    cdf = np.cumsum(pdf)

    plt.figure(figsize=figsize)
    plt.plot(cdf)
    plt.title("Original CDF")
    plt.show()
    # PDF and CDF ploted on a single plot

    fig, ax = draw_hist(bins_start, pdf)
    ax.plot(cdf * np.max(pdf), "r")
    plt.title("Original PDF+ const*CDF to show the connection between the two"
    )
    plt.show()
    # unnormalizing CDF to get equalization function

    f_eq = np.round(cdf * 255).astype(int)

    f_eq
    # Using the equaization function to get the equalized image

    I_eq = f_eq[I]

    plt.figure(figsize=figsize)
    plt.imshow(I_eq, cmap="gray", vmin=0, vmax=255)
    plt.title("equalized image")
    plt.show()

    Output:
    Conclusion:
    Histogram Equalization is a technique used when the contrast of the image is too
    high or too low. It works by mapping the pixel intensities in the image to a new
    distribution that is more uniform. This is achieved by computing the CDF of the
    pixel intensities in the image and then mapping the original pixel intensities to new
    values using a transformation function that is derived by normalizing the CDF and
    multiplying it by the maximum pixel value of an image.

    You might also like