sem_size_analysis

KO | EN

gitlite — search

#javascript #python #typescript #llm #android #machine-learning #linux #docker #nodejs #awesome #cli #awesome-list #security #mcp

sem_size_analysis

(★ 12)

This is a repository for using deep learning to segment SEM images for easier and faster analysis of sample structures and properties

sem_size_analysis 최신버젼 다운로드

최종 버전 다운로드 (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# SAEMI

## Description
SAEMI (Size Analysis for Electron Microscopy Images) is a tool for obtaining the size distribution of nanoparticles in electron microscopy images. 
It uses an image segmentation model trained through a neural network to detect nanoparticle sizes and locations. The neural network used is a [u-net](https://arxiv.org/pdf/1505.04597.pdf) 
with a resnet downsample created through the [fastai](https://github.com/fastai/fastai) library. From there, it then obtains a histogram of the sizes of each particle, providing important 
information such as mean particle size, median particle size etc. This tool can be used for characterizing new materials, analyzing environmental effects on chemical systems, 
ensure quality control for routine synthesis and much more.

## Dependencies
- Python == 3.6
- dash == 1.9.1
- dash-canvas == 0.1.0
- fastai == 1.0.60
- gdown == 3.10.1

## Website
You can find the app hosted on Heroku [here](https://saemi.herokuapp.com/)

## To Run Locally
### Using Docker
The easiest way to run the app locally is to run it as a docker container. If you wish to use this method, please ensure you have docker installed on your local machine. You can get
docker from: https://docs.docker.com/get-docker/

After obtaining docker, simply clone the repo into a local directory and run the following command:
```
docker-compose up
```

### Without Docker
If you wish to run the app locally without using docker, first clone the repo into a local directory. As well, please ensure you have the above dependencies installed. If you already have
Python 3.6 installed, you can install the rest of the dependencies by running `pip install -r requirements.txt` after cloning the repo.

Once the dependencies have been succesfully installed, go into the cloned repo and from the command line run:
```
python -m src.backend_api
```
This will run the backend server on `http://0.0.0.0:8000`. To run the frontend server, run the following in a separate command line (also from the cloned repo directory):
```
python -m app
```
This will run the frontend server on `http://127.0.0.1:8050/`. Copy and paste the link into your web browser and you can use the app from there.

## Using the App
Please check the [user docs](https://github.com/lwang94/sem_size_analysis/blob/master/docs/user_docs.md) for more details. 
Below is a summary of the steps a user would take to obtain the size distribution of an electron microscopy image using this app.

1. Preprocess the image: before using the app, remove any metadata such as the scale bar that may interfere with the deep learning prediction

2. Upload the image: upload the image to the app using the buttons on the homepage. The image will then be segmented by the model and a size distribution calculated from the segmentation

3. Postprocess the image: After tbe size distribution is calculated, the user can edit the prediction to give more accurate results in case the model did not satisfactorily segment the image. 
Check the [user docs](https://github.com/lwang94/sem_size_analysis/blob/master/docs/user_docs.md) for more details on post processing

4. Download the calculated particle sizes as a .csv file using the Download link provided. Keep in mind that the particle sizes are in units of pixels so it is up to the user to convert
that to a physical distance.

<!-- 
INCLUDE MEDIUM ARTICLE IN USING THE APP

## Installation
----------------
### Dependencies
saemi requires:
- Python == 3.6
- fastai == 1.0.6.0
- scikit-image == 0.16.2
- dash == 1.8.0
- gdown == 3.10.1

### User installation (TO DO)
To pip install saemi, run:
```
pip install git+git://github.com/lwang94/sem_size_analysis
```

## To Run
----------------
To open the app first cd to the root directory and run in the command shell
```
python -m src.backend_api
```
The above will open the backend api server on localhost:5000. To interact via a front-end UI, run in a different command shell
```
python -m app
```
You should see an output that looks like this.
```
Running on http://127.0.0.1:8050/ (Press CTRL+C to quit)
```
Simply copy and paste the http address into a local browser and press Enter.

### To run tests
With coverage:
```
pytest --cov-config=.coveragerc --cov=sem_size_analysis tests/
```
-->

sem_size_analysis

sem_size_analysis 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

sem_size_analysis

sem_size_analysis 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Download README