8000 GitHub - paleopollen/post-processing-cli: A Python-based command-line application for basic post-processing of pollen detection images using Segment Anything Model 2 (SAM 2).
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

A Python-based command-line application for basic post-processing of pollen detection images using Segment Anything Model 2 (SAM 2).

License

Notifications You must be signed in to change notification settings

paleopollen/post-processing-cli

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

26 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 8000  
 
 
 
 
 
 
 
 
 
 

Repository files navigation

A Command Line Interface for Basic Post-Processing of Pollen Detection Images Using Segment Anything Model 2 (SAM 2).

DOI

We present Pollen Detection Post-processor, a Command Line Interface (CLI) for basic post-processing of pollen detection images generated using our Pollen Detection CLI. This CLI uses the Segment Anything Model 2 (SAM 2) model to create segmentation masks and segmented pollen images by using the center of pollen detection as the input prompt.

Installation

Download SAM 2 Checkpoints and Configuration Files

To download the SAM 2 checkpoints and configuration files, run the following commands:

cd checkpoints
./download.sh

cd sam2_configs
./download.sh

Docker Build and Run

To install the package using Docker, run the following commands:

Build Docker Image

docker build -t post-processing-cli .

Run CLI

Example command:

docker run -it --rm -v $(pwd)/data:/data -v $(pwd)/checkpoints:/usr/src/app/checkpoints -v $(pwd)/sam2_configs:/usr/src/app/sam2_configs post-processing-cli -d /data/pollen-detections-dir -m /usr/src/app/checkpoints/sam2_hiera_large.pt -c /usr/src/app/sam2_configs/sam2_hiera_l.yaml

Here, CLI arguments -d contains the path to the directory containing the pollen detection images, -m contains the path to the SAM 2 model checkpoint, and -c contains the path to the SAM 2 configuration file. The CLI saves the output segmentation masks and images in the same directory as the input detection images and updates the metadata JSON file.

An example directory structure for the input pollen detection images that uses the Pollen Detection CLI is as follows:

This example assumes 9 z-plane images with 256 pixels of overlap between each tile.

pollen-detections-dir
├── pollen_detection_slide1_0x_0y
│   ├── 0z.png
│   ├── 1z.png
│   ├── 2z.png
│   ├── 3z.png
│   ├── 4z.png
│   ├── 5z.png
│   ├── 6z.png
│   ├── 7z.png
│   ├── 8z.png
│   ├── mask_1.png
│   ├── metadata_1.json
├── pollen_detection_slide1_0x_768y
│   ├── 0z.png
│   ├── 1z.png
│   ├── 2z.png
│   ├── 3z.png
│   ├── 4z.png
│   ├── 5z.png
│   ├── 6z.png
│   ├── 7z.png
│   ├── 8z.png
│   ├── mask_1.png
│   ├── metadata_1.json
└── ...

Usage

usage: post_processing_cli.py [-h] --detections-dir [DETECTIONS_DIR_PATH] --sam2-model-path [SAM2_MODEL_PATH] --sam2-config-path [SAM2_CONFIG_PATH]

Post-process the pollen detection results.

options:
  -h, --help            show this help message and exit
  --detections-dir [DETECTIONS_DIR_PATH], -d [DETECTIONS_DIR_PATH]
                        Full path of the directory containing the detection results.
  --sam2-model-path [SAM2_MODEL_PATH], -m [SAM2_MODEL_PATH]
                        Full path of the SAM-2 model.
  --sam2-config-path [SAM2_CONFIG_PATH], -c [SAM2_CONFIG_PATH]
                        Full path of the SAM-2 config file.

About

A Python-based command-line application for basic post-processing of pollen detection images using Segment Anything Model 2 (SAM 2).

Resources

License

Stars

Watchers

Forks

Packages

No packages published
0