MahmudulAlam/Automatic-Identification-and-Counting-of-Blood-Cells ? reverse-engineered prompt

Build me a Python tool that can detect and count blood cells in smear images, including red blood cells, white blood cells, and platelets. I want it to take an image, run the trained model, draw the detected cells on the image, and show the final counts in a simple way.

Please make sure it works with the existing blood cell dataset style, supports running on normal images and larger high resolution images by splitting them into smaller patches, and includes the duplicate platelet check using nearest neighbor and overlap filtering so repeated predictions get removed. If anything in the setup is outdated, update it so the project runs smoothly with current docs if needed, but keep it compatible with the original TensorFlow based approach.

I’d also like clear setup steps, an easy way to run detection, and a simple way to test it end to end with the provided weights.