This repository contains a collection of computer vision lab tasks completed as part of a coursework or project. These tasks cover various computer vision concepts and techniques, demonstrating proficiency in image processing, object detection, and related fields.
A simple Python program to manage a grocery list, allowing users to add, remove items, and display the current list.
Implementation of a student record system using a Python dictionary, allowing operations like adding students, updating grades, and displaying records.
- Load and display images
- Convert images to grayscale
- Resize images using OpenCV
- Drawing basic shapes on images
- Apply Gaussian blur, crop, and manipulate images
- Add text to images
- Apply binary thresholding and rotation
- Blend two images, convert to grayscale, and apply histogram equalization
- Perform bitwise operations on binary images
- Convert image pixel values to a Pandas DataFrame and apply masks
Explore a dataset containing images of pets categorized into four classes: Angry, Sad, Happy, and Others. Display the number of samples in each class.
Load the pet emotions dataset, resize images, normalize pixel values, and split the dataset into training and testing sets.
Perform EDA on the pet emotions dataset, displaying the distribution of class labels using a bar plot.
Display sample images from each class of the pet emotions dataset along with their labels.
Calculate summary statistics for each class (Angry, Sad, Happy, Others) to understand the distribution of emotions in the dataset.
Collection and basic operations on the "Common Objects-Within University" dataset.
Tasks include loading and displaying X-ray images, contrast enhancement, color mapping, color balance, color filtering, logarithmic and power-law transformations.
Tasks include loading X-ray and MRI images, histogram equalization, color mapping, multi-modal weighted fusion, logarithmic and power-law transformations, and comparative analysis.
Capture live video, apply various image enhancement operations in real-time, and display the original and enhanced video frames.
Implement Gaussian blur, Sobel edge detection, image sharpening, mean filter for noise reduction.
Develop median filter, max filter (dilation), min filter (erosion), bilateral filter, and adaptive median filter.
Calculate 1D and 2D Fourier Transforms, implement high-pass filter, and perform image compression using Fourier Transformation.
Create hybrid images from two input images with different spatial frequencies, experiment with filter combinations, and analyze trade-offs.
Discuss the application of edge detection as a feature extraction technique for tumor detection and propose an additional feature extraction technique.
Implement Harris Corner Detection algorithm, detect corners in an image, and experiment with different threshold values.
Implement corner detection in real-time using the Harris or Shi-Tomasi method on video frames.
Implement Harris or Shi-Tomasi Corner Detection for image stitching by detecting corners in multiple images.
Use ORB (Oriented FAST and Rotated BRIEF) detector and descriptor for feature detection and matching between two images.
Perform thresholding-based segmentation on a medical X-ray image to isolate a bone fracture.
Perform region growing-based segmentation on a microscopic image of cells to identify and separate a specific cell.
Use watershed segmentation to separate and count individual coins in an image of overlapping coins.
Perform cluster-based segmentation on an image of colorful flowers to separate different types of flowers based on color.
Implement screen detection in a computer lab using the Hough Line Transformation to identify boundaries of computer screens.
Implement asset tracking in a computer lab using the Scale-Invariant Feature Transform (SIFT) to recognize and identify individual computer systems and components.
Develop a system for real-time anomaly detection in sensor data using the wavelet transformation.
Implement object recognition using the SIFT algorithm on a set of test images.
Create a panoramic image by stitching multiple overlapping images together using the SIFT algorithm.
Implement lane detection for an autonomous vehicle project using the Hough Line Transformation.
Implement coin detection and counting using the Hough Circle Transformation.
Implement boundary detection in a security system to detect unauthorized objects in a predefined zone in a real-time video stream.
Develop a CNN model for gender classification using a dataset of human faces labeled with gender information.
Create a CNN-based model for recognizing facial expressions in images of animals.
Build a system that estimates the age of a person in a video using a CNN-based architecture.
Implement a system for real-time hand gesture recognition using CNN models.
Perform image classification using pre-trained models, EfficientNet and ResNet50, on a chosen dataset.
Use YOLO (You Only Look Once) to detect home assets using a live web camera.
Implement object detection using Regional Convolutional Neural Networks (R-CNN) on a dataset of your choice.
Classify images using Vision Transformers on a dataset of your choice.