Şükrü Burak Çetin's Projects
This script enables users to extract annotation points from binary mask images and write/store them into a .json format file to make it usable in order to feed Mask RCNN model training. It can also show the points on each specific image to control whether the points are extracted in the specific and correct order.
The project basically interferes weighted suitibility analysis to tell users that the area they will mark on the map is a suitable area for a new branch. By placing markers in different areas on the map, it receives information about the locations represented by the markers about the new branch to be opened and compares these areas among themselves
Mitosis detection algorith classifes each nuclei as mitosis or not. It needs nuclei detection algorithm initiation at first. VGG11(132million parameters) and Pytorch are used in this set up.
Nuclei detection algorithm which works on the areas that hard to indentify(mainly for breast cancer tissues). The algorithm base uses Unet structure and Pytorch.
This function takes a panoramic image as input and returns the fisheye version of the image. It extracts the height and width of the input panoramic image. Calculates the dimensions of the fisheye image based on the minimum of its height and width, setting it to have a width of twice its height to maintain a 2:1 aspect ratio.
Modified source code of published article: Real-ESRGAN: A deep learning approach for general image restoration and its application to aerial images
Classifies and eliminates images that don't have any seedlings from the source that are captured for the automation of the agriculture environment by looking at their features that consist of footprints of a seedling.
Segments plant data according to the Unet model. Provides pixel-based segmentation.
Calculates growing rate, manages and regulates essential parameters to increase the efficiency of growing phases via using UNet pixel classification and Pytorch implementation.
The repository provides code for running inference with the SegmentAnything Model (SAM), links for downloading the trained model checkpoints, and example notebooks that show how to use the model.
It calculates the solar energy potential within the selected polygon area and provides the user with information on long-term energy recycling with the classifier.
İstanbul Büyük Şehir Belediyesi Açık Veri Portalı'ndan Elde Edilmiş Veri Üzerinden İstanbul'daki Parklar ve Yeşil Alanların İlçe Bazında Çeşitli Sosyokültürel Değerlerle İlişkilendirilip Analiz Edilmesi
A simple tool for labeling object maks in images, implemented with Python Tkinter.