Home
This repository contains a collection of Phyton Jupiter notebooks that illustrate various core image analysis algorithms. These were developed as part of different lecture courses I have given. These notebooks are groups into sections that are listed below.
This is work in progress and all the acknowledgements still need to be included. Of course none of this would have been possible without the scikit-image libraries. Many of the examples here are just adopted from their documentation.
Thanks to the wonderful work other have done to develop Binder it is also possible to view an interactive version of these notebooks:
Alternatively, you can also use Google Colab to run these scripts. Open a new file I Colab and copy following command to clone the GitHub repository
!git clone https://github.com/engs1258/biomedical-image-analysis-notebooks/
Once you click not the orange folder icon on the left you see all the files in your local drive.
| Notebook | Description |
|---|---|
| biomedia_intro_00_read_image | Some basic code to read and display an image |
| biomedia_intro_01_dicom | Code to read a basic DICOM image |
| biomedia_intro_02_image_contrast | A demo on who to optimise image contrast |
| biomedia_intro_03_threshold | Convert an image to a binary image - and figure out where this can fail. |
| biomedia_intro_04_erosion | Example of the morphological erosion filter |
| biomedia_intro_05_dilation | Example of the morphological dilation filter |
| biomedia_intro_06_opening | Example of the morphological opening filter |
| biomedia_intro_07_closing | Example of the morphological closing filter |
| Notebook | Description |
|---|---|
| biomedia_filters_00_fft | Calculate magnitude and phase of Fourier transform |
| biomedia_filters_01_gaussian | Filter image with Gaussian |
| biomedia_filters_02_median | Filter image with Median |
| biomedia_filters_03_gradient | Calculate partial derivatives and gradients in images |
| biomedia_filters_04_gaussian_pyramid | Use the Gaussian pyramid to generate a multi-scale representation |
| biomedia_filters_05_laplacian_pyramid | Use the Laplacian pyramid to generate a multi-scale representation |
| Notebook | Description |
|---|---|
| biomedia_features_00_sobel | Use a Sobel operator for detecting edges |
| biomedia_features_01_canny | Use the Canny edge detector |
| biomedia_features_02_blobs | Detecting blobs with a Mexican hat filter |
| biomedia_features_03_vessel_enhance | Comparison of different vessel enhancement filters |
| biomedia_features_04_skeleton | Extracting the skeleton from a binary image |
| biomedia_features_05_GLCM | Gray value co-occurrence matrix as a feature descriptor |
| biomedia_features_06_feature_points | Illustration for generating and matching key points |
| Notebook | Description |
|---|---|
| biomedia_seg_00_label_image | Extracting labelled components from an segmented image |
| biomedia_seg_01_otsu | Otsu for global thresholding |
| biomedia_seg_02_connected_comp | Connected components applied to binary image |
| biomedia_seg_03_watershed | Watershed segmentation |
| biomedia_seg_04_kmeans | K-means applied to image segmentation |
| biomedia_seg_05_superpixels | SLIC super pixels applied to a histology image |
| biomedia_seg_06_normalised_cuts | Using superpixels and normalised cut for image segmentation |
| Notebook | Description |
|---|---|
| biomedia_classification_00_feature_extraction_test | Test feature extraction method on a subset of images |
| biomedia_classification_01_generate_features | Calculate features for the entire dataset |
| biomedia_classification_02_pca | Using PCA to visualise the data set |
| biomedia_classification_03_gaussian_mixture | Classifier based on Gaussian mixture model |
| biomedia_classification_04_logistic_regression | Classifier based on logistic regression |
| biomedia_classification_05_svm | Support vector machine classifier predicting class labels |