# liyongqingupc/mcan-oilspilldetection architecture

> Architecture diagram of the GitHub repository liyongqingupc/mcan-oilspilldetection, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-08). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/liyongqingupc/mcan-oilspilldetection
- Repository: https://github.com/liyongqingupc/mcan-oilspilldetection

## Overview

MCAN is a PyTorch oil-spill detection project trained from small image datasets. Its documented user workflow is to place training images in Input/TrainingSet and run main_train.py, then place evaluation images in Input/TestSet and run test.py with a trained model. The repository also includes multiscale image-resizing utilities, adversarial network definitions, image-generation helpers, and an Inception-based SIFID metric implementation. Training internals and entry-point wiring were not sampled, so those relationships are left unspecified.

Main entry points: [`main_train.py`](https://github.com/liyongqingupc/mcan-oilspilldetection/blob/master/main_train.py) (Training entry point), [`test.py`](https://github.com/liyongqingupc/mcan-oilspilldetection/blob/master/test.py) (Testing entry point), [`Input/TrainingSet`](https://github.com/liyongqingupc/mcan-oilspilldetection/tree/master/Input/TrainingSet) (Training images), [`Input/TestSet`](https://github.com/liyongqingupc/mcan-oilspilldetection/tree/master/Input/TestSet) (Test images)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_workflow["Train and test"]
  node_trainer["Training images"]
  node_train_entry["Training entry point<br/>[main_train.py]"]
  node_training_loop["Multiscale training<br/>[training.py]"]
  node_tester["Test images"]
  node_test_entry["Testing entry point<br/>[test.py]"]
  node_config["Run configuration<br/>[config.py]"]
end

subgraph group_gan["Adversarial model"]
  node_networks["Generator and discriminator<br/>[models.py]"]
end

subgraph group_image["Image processing"]
  node_image_ops["Image and GAN utilities<br/>[functions.py]"]
  node_resize["Multiscale resizing<br/>[imresize.py]"]
  node_manipulate["Image generation<br/>[manipulate.py]"]
end

subgraph group_metrics["Evaluation metric"]
  node_sifid_score["SIFID scoring<br/>[sifid_score.py]"]
  node_inception["Inception features<br/>[inception.py]"]
end

node_train_command(("Training user"))
node_test_command(("Testing user"))

node_train_command -->|"runs"| node_train_entry
node_trainer -->|"provides images"| node_train_entry
node_test_command -->|"runs"| node_test_entry
node_tester -->|"provides images"| node_test_entry
node_image_ops -->|"calls resizing"| node_resize
node_manipulate -->|"imports training"| node_training_loop

click node_trainer "https://github.com/liyongqingupc/mcan-oilspilldetection/tree/master/Input/TrainingSet"
click node_train_entry "https://github.com/liyongqingupc/mcan-oilspilldetection/blob/master/main_train.py"
click node_training_loop "https://github.com/liyongqingupc/mcan-oilspilldetection/blob/master/MCAN/training.py"
click node_tester "https://github.com/liyongqingupc/mcan-oilspilldetection/tree/master/Input/TestSet"
click node_test_entry "https://github.com/liyongqingupc/mcan-oilspilldetection/blob/master/test.py"
click node_config "https://github.com/liyongqingupc/mcan-oilspilldetection/blob/master/config.py"
click node_networks "https://github.com/liyongqingupc/mcan-oilspilldetection/blob/master/MCAN/models.py"
click node_image_ops "https://github.com/liyongqingupc/mcan-oilspilldetection/blob/master/MCAN/functions.py"
click node_resize "https://github.com/liyongqingupc/mcan-oilspilldetection/blob/master/MCAN/imresize.py"
click node_manipulate "https://github.com/liyongqingupc/mcan-oilspilldetection/blob/master/MCAN/manipulate.py"
click node_sifid_score "https://github.com/liyongqingupc/mcan-oilspilldetection/blob/master/SIFID/sifid_score.py"
click node_inception "https://github.com/liyongqingupc/mcan-oilspilldetection/blob/master/SIFID/inception.py"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_trainer,node_train_entry,node_training_loop,node_tester,node_test_entry,node_config,node_train_command,node_test_command toneBlue
class node_networks toneAmber
class node_image_ops,node_resize,node_manipulate toneMint
class node_sifid_score,node_inception toneRose
```

## Components

### Train and test

- [Training images](https://github.com/liyongqingupc/mcan-oilspilldetection/tree/master/Input/TrainingSet): `Input/TrainingSet`
- [Training entry point](https://github.com/liyongqingupc/mcan-oilspilldetection/blob/master/main_train.py): `main_train.py`
- [Multiscale training](https://github.com/liyongqingupc/mcan-oilspilldetection/blob/master/MCAN/training.py): `MCAN/training.py`
- [Test images](https://github.com/liyongqingupc/mcan-oilspilldetection/tree/master/Input/TestSet): `Input/TestSet`
- [Testing entry point](https://github.com/liyongqingupc/mcan-oilspilldetection/blob/master/test.py): `test.py`
- [Run configuration](https://github.com/liyongqingupc/mcan-oilspilldetection/blob/master/config.py): `config.py`

### Adversarial model

- [Generator and discriminator](https://github.com/liyongqingupc/mcan-oilspilldetection/blob/master/MCAN/models.py): `MCAN/models.py`

### Image processing

- [Image and GAN utilities](https://github.com/liyongqingupc/mcan-oilspilldetection/blob/master/MCAN/functions.py): `MCAN/functions.py`
- [Multiscale resizing](https://github.com/liyongqingupc/mcan-oilspilldetection/blob/master/MCAN/imresize.py): `MCAN/imresize.py`
- [Image generation](https://github.com/liyongqingupc/mcan-oilspilldetection/blob/master/MCAN/manipulate.py): `MCAN/manipulate.py`

### Evaluation metric

- [SIFID scoring](https://github.com/liyongqingupc/mcan-oilspilldetection/blob/master/SIFID/sifid_score.py): `SIFID/sifid_score.py`
- [Inception features](https://github.com/liyongqingupc/mcan-oilspilldetection/blob/master/SIFID/inception.py): `SIFID/inception.py`

### Other components

- **Training user**
- **Testing user**

## Connections

- Training user → Training entry point: runs (evidence: `README.md`)
- Training images → Training entry point: provides images (evidence: `README.md`)
- Testing user → Testing entry point: runs (evidence: `README.md`)
- Test images → Testing entry point: provides images (evidence: `README.md`)
- Image and GAN utilities → Multiscale resizing: calls resizing (evidence: `MCAN/functions.py`)
- Image generation → Multiscale training: imports training (evidence: `MCAN/manipulate.py`)
