CV-Toolbox: Filters, Histograms, Gradients, Frequency, Contours

Table of contents:

• Objectives
• Project Features
• Dependinces
• How to Run The Project
• Project Overview
• Team

Features

• Add various noise types to images such as uniform, guassian & salt-pepper noise.
• Filtration of noisy images using low pass filters such as: average, Gaussian, median filters.
• Edge detection using variety of masks such as: Sobel, Prewitt, Roberts and canny edge detectors.
• Change RGB images to grayscale
• View histograms of images and show equalization & normalization of image.
• Add frequency domain filters such as low-pass, high-pass & Gaussian filters.
• Mix images in frequency domain to get Hybrid images.
• Apply Hough transform for detecting parametric shapes like circles, ellipses and lines
• Apply Active Contour Model for semi-supervised shape delineation.
• Extract the unique features in all images using Harris operator
• Generate feature descriptors using scale invariant features (SIFT).
• Match the image set features using sum of squared differences (SSD) and normalized cross correlations (NCC).
• Apply thresholding using optimal thresholding, Otsu, spectral thresholding (more than 2 modes), local & global thresholding.
• Do unspervised segmentation using k-means, region growing, agglomerative and mean shift methods.
• Detect faces (color or grayscale)
• Recognize faces based on PCA/Eigen analysis

Toolkit Folders

• ActiveContours: Includes active contours class related codes

  • active_contour.cpp /.h
  • image.cpp /.h
  • utils.cpp/.h

• FaceRegonsition: Includes face detection and recognistion code including pca and eigen analysis

  • detectFaces.cpp /.h

• Filters : Includes implementation for filtration functions (requirements 1-3)

  • edgedetectors.cpp/.h
  • noiseaddiatves.cpp/.h
  • imagesmoothers.cpp/ .h
  • threshold.cpp/ .h

• Frequency : Includes implementation for frequency domain related tasks

  • fouriermixer.cpp/ .h

• Harris: Includes Harris Corner Operator Class

  • harrisoperator.cpp/ .h

• Helpers : Includes the helper functions used all over program like converting images to view them.

  • helperfunctions.cpp/ .h

• Histograms : Includes implementation for histogram related tasks.

  • histograms.cpp/ .h

• Hough: Includes implementation classes of hough transform of Lines, Circles & Ellipses

  • houghcircle.cpp/ .h
  • houghellipse.cpp/ .h
  • houghline.cpp/ .h

• Segementation: Includes segementation classes

  • kmeanssegmentation.cpp/ .h
  • meanshift.cpp/ .h
  • agglomerative.cpp/.h
  • regiongrowing.cpp/ .h

• SSD_NCC: Includes SIFT & image matching classes

  • ssd_ncc.cpp/ .h

• Thresholding: Includes Thresholding related tasks

  • Thresholding.cpp/ .h

• UI files: has the ui files

  • mainwindow.cpp/ .h/ .ui
  • filterwidget.cpp/ .h/ .ui
  • histogramswidget.cpp/ .h/ .ui
  • houghwidget.cpp/ .h/ .ui
  • mixingwidget.cpp/ .h/ .ui
  • facerecognitionwidget.cpp/ .h/ .ui
  • matchingWidget.cpp/ .h/ .ui
  • segementationWidget.cpp/ .h/ .ui

• Assets : folder contains program assets like UI SVGs, images to test implementation & Cascading Face Classifer xml file.

Dependenices

- Qt Version 6.4.2
- Qt Charts
- OpenCV 4.7.0

How to run the project

• Download and Install Qt from the online installer using minGW 64 architecture.

www.qt.io

make sure that QtCharts is enabled

• Download and build openCV 4.7.0 from their website

www.opencv.org/releases/

• if you don't know how to build openCV on your PC please refer to the installCV.md

• Go to the project.pro file and make sure that openCV libraries are refered to in your pc like that.

INCLUDEPATH += <yourCV build folder path>\install\include

LIBS += -L<yourCV build folder path>\release\lib \
    -lopencv_calib3d470                 \
    -lopencv_core470                    \
    -lopencv_features2d470              \
    -lopencv_flann470                   \
    -lopencv_highgui470                 \
    -lopencv_imgproc470                 \
    -lopencv_imgcodecs470               \
    -lopencv_photo470                   \
    -lopencv_stitching470               \
    -lopencv_ts470                      \
    -lopencv_video470                   \
    -lopencv_videoio470                 \
    -lopencv_objdetect470               \

• Build the project
• Run the project

Project Overview

• Filters Tab
• Hybrid Image Tab
• Histograms Tab
• Hough Trasform & Active Contours Tab
• Harris & SIFT Tab
• Segementation Tab Face Detection & Recognistion Tab

Team

Second Semester - Computer Vision (SBE3230) class project created by:

Team Members' Names	Section	B.N.
Ibrahim Mohamed	1	2
Mahmoud Yaser	2	30
Maye Khaled	2	40
Marina Nasser	2	12
Omnia Sayed	1	14

Submitted to:

• Eng. Peter Emad & Eng. Laila Abbas All rights reserved © 2022 to Team 5 - Systems & Biomedical Engineering, Cairo University (Class 2024)