My solutions to the Computer Vision course, Fall 2022, Dr. Mohammadi
- Homework 1: Basics of image formation and OpenCV
- Homework 2: Concepts of FOV, DOF, lens distortion and implementation of camera calibration
- Homework 3: Implementation of histogram equalization, histogram matching and CLAHE algorithm
- Homework 4: Applying different kernels and filters on pictures such as mean, median and derivative. Additionaly, denoising pictures using fast fourier transform.
- Homework 5: Continuing on the subject of denoising with FFT; Introducing canny and sobel edge detectors; Finding equation of a line inside a picture.
- Homework 6: Using Hough transform to extract lines from an image; Introduction to color spaces and convertion between them.
- Homework 7: Concepts of image alignment and transformation; Implementation of a transform algorithm like the Camscanner app.
- Homework 8: Focusing on image segmentation and morphology image processing.
- Homework 9: Using mophology operations to extract skeleton of shapes in pictures and count number of objects in pictures.
- Homework 10: Using texture descriptors and shape descriptors like compatness, eccenticity, solidity and histogram of LBP for classification.
- Homework 11: Basics of Keras; Implementing CNN models for classification; Fine-tuning Resnet50 model on custom dataset.
- Homework 12: Concepts of overfitting and underfitting and using data augmentation to overcome overfitting problem; Using U-Net for semantic segmentation.